From becbef9e489b477b2a3fdd0de6ab754941d14351 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Mon, 1 Apr 2024 02:08:35 +0000 Subject: [PATCH 001/680] sketch of read-in-order optimization --- .../IMergingAlgorithmWithDelayedChunk.cpp | 2 +- .../IMergingAlgorithmWithSharedChunks.cpp | 2 +- .../Algorithms/MergeTreePartLevelInfo.h | 25 ------------- .../Merges/Algorithms/MergeTreeReadInfo.h | 35 +++++++++++++++++++ .../Algorithms/MergingSortedAlgorithm.cpp | 10 ++++++ .../QueryPlan/ReadFromMergeTree.cpp | 12 +++++-- src/Processors/QueryPlan/ReadFromMergeTree.h | 2 +- .../MergeTree/MergeTreeRangeReader.cpp | 17 +++++++-- src/Storages/MergeTree/MergeTreeRangeReader.h | 2 +- src/Storages/MergeTree/MergeTreeReadTask.cpp | 8 ++++- src/Storages/MergeTree/MergeTreeReadTask.h | 6 ++++ .../MergeTree/MergeTreeSelectProcessor.cpp | 13 +++++-- .../MergeTree/MergeTreeSelectProcessor.h | 6 ++++ .../MergeTree/MergeTreeSequentialSource.cpp | 5 +-- 14 files changed, 106 insertions(+), 39 deletions(-) delete mode 100644 src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h create mode 100644 src/Processors/Merges/Algorithms/MergeTreeReadInfo.h diff --git a/src/Processors/Merges/Algorithms/IMergingAlgorithmWithDelayedChunk.cpp b/src/Processors/Merges/Algorithms/IMergingAlgorithmWithDelayedChunk.cpp index cbad6813fbc..13b245717b3 100644 --- a/src/Processors/Merges/Algorithms/IMergingAlgorithmWithDelayedChunk.cpp +++ b/src/Processors/Merges/Algorithms/IMergingAlgorithmWithDelayedChunk.cpp @@ -1,5 +1,5 @@ #include -#include +#include namespace DB diff --git a/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.cpp b/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.cpp index c8b69382e89..4fe50feaede 100644 --- a/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.cpp +++ b/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.cpp @@ -1,5 +1,5 @@ #include -#include +#include namespace DB { diff --git a/src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h b/src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h deleted file mode 100644 index bcf4e759024..00000000000 --- a/src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h +++ /dev/null @@ -1,25 +0,0 @@ -#pragma once - -#include - -namespace DB -{ - -/// To carry part level if chunk is produced by a merge tree source -class MergeTreePartLevelInfo : public ChunkInfo -{ -public: - MergeTreePartLevelInfo() = delete; - explicit MergeTreePartLevelInfo(ssize_t part_level) : origin_merge_tree_part_level(part_level) { } - size_t origin_merge_tree_part_level = 0; -}; - -inline size_t getPartLevelFromChunk(const Chunk & chunk) -{ - const auto & info = chunk.getChunkInfo(); - if (const auto * part_level_info = typeid_cast(info.get())) - return part_level_info->origin_merge_tree_part_level; - return 0; -} - -} diff --git a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h new file mode 100644 index 00000000000..e79df0fb8c8 --- /dev/null +++ b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h @@ -0,0 +1,35 @@ +#pragma once + +#include + +namespace DB +{ + +/// To carry part level and virtual row if chunk is produced by a merge tree source +class MergeTreeReadInfo : public ChunkInfo +{ +public: + MergeTreeReadInfo() = delete; + explicit MergeTreeReadInfo(size_t part_level, bool virtual_row_) : + origin_merge_tree_part_level(part_level), virtual_row(virtual_row_) { } + size_t origin_merge_tree_part_level = 0; + bool virtual_row = false; +}; + +inline size_t getPartLevelFromChunk(const Chunk & chunk) +{ + const auto & info = chunk.getChunkInfo(); + if (const auto * read_info = typeid_cast(info.get())) + return read_info->origin_merge_tree_part_level; + return 0; +} + +inline bool getVirtualRowFromChunk(const Chunk & chunk) +{ + const auto & info = chunk.getChunkInfo(); + if (const auto * read_info = typeid_cast(info.get())) + return read_info->virtual_row; + return 0; +} + +} diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp index 1debfcec8e0..89f0193b05b 100644 --- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp @@ -1,3 +1,4 @@ +#include #include #include #include @@ -239,6 +240,15 @@ IMergingAlgorithm::Status MergingSortedAlgorithm::mergeBatchImpl(TSortingQueue & auto [current_ptr, initial_batch_size] = queue.current(); auto current = *current_ptr; + if (getVirtualRowFromChunk(current_inputs[current.impl->order].chunk)) + { + /// If virtual row is detected, there should be only one row as a single chunk, + /// and always skip this chunk to pull the next one. + assert(initial_batch_size == 1); + queue.removeTop(); + return Status(current.impl->order); + } + bool batch_skip_last_row = false; if (current.impl->isLast(initial_batch_size) && current_inputs[current.impl->order].skip_last_row) { diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index f4607cad040..91cd362f1d9 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -501,7 +501,8 @@ Pipe ReadFromMergeTree::readInOrder( Names required_columns, PoolSettings pool_settings, ReadType read_type, - UInt64 limit) + UInt64 limit, + bool need_virtual_row) { /// For reading in order it makes sense to read only /// one range per task to reduce number of read rows. @@ -596,6 +597,8 @@ Pipe ReadFromMergeTree::readInOrder( processor->addPartLevelToChunk(isQueryWithFinal()); + processor->addVirtualRowToChunk(need_virtual_row); + auto source = std::make_shared(std::move(processor)); if (set_rows_approx) source->addTotalRowsApprox(total_rows); @@ -1028,7 +1031,12 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( } for (auto && item : splitted_parts_and_ranges) - pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit)); + { + /// need_virtual_row = true means a MergingSortedTransform should occur. + /// If so, adding a virtual row might speedup in the case of multiple parts. + bool need_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; + pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit, need_virtual_row)); + } } Block pipe_header; diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.h b/src/Processors/QueryPlan/ReadFromMergeTree.h index 5ed742a9bfd..6a08622af11 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.h +++ b/src/Processors/QueryPlan/ReadFromMergeTree.h @@ -251,7 +251,7 @@ private: Pipe read(RangesInDataParts parts_with_range, Names required_columns, ReadType read_type, size_t max_streams, size_t min_marks_for_concurrent_read, bool use_uncompressed_cache); Pipe readFromPool(RangesInDataParts parts_with_range, Names required_columns, PoolSettings pool_settings); Pipe readFromPoolParallelReplicas(RangesInDataParts parts_with_range, Names required_columns, PoolSettings pool_settings); - Pipe readInOrder(RangesInDataParts parts_with_ranges, Names required_columns, PoolSettings pool_settings, ReadType read_type, UInt64 limit); + Pipe readInOrder(RangesInDataParts parts_with_ranges, Names required_columns, PoolSettings pool_settings, ReadType read_type, UInt64 limit, bool need_virtual_row = false); Pipe spreadMarkRanges(RangesInDataParts && parts_with_ranges, size_t num_streams, AnalysisResult & result, ActionsDAGPtr & result_projection); diff --git a/src/Storages/MergeTree/MergeTreeRangeReader.cpp b/src/Storages/MergeTree/MergeTreeRangeReader.cpp index 6932762f58b..0456a8e2787 100644 --- a/src/Storages/MergeTree/MergeTreeRangeReader.cpp +++ b/src/Storages/MergeTree/MergeTreeRangeReader.cpp @@ -946,7 +946,7 @@ String addDummyColumnWithRowCount(Block & block, size_t num_rows) } -MergeTreeRangeReader::ReadResult MergeTreeRangeReader::read(size_t max_rows, MarkRanges & ranges) +MergeTreeRangeReader::ReadResult MergeTreeRangeReader::read(size_t max_rows, MarkRanges & ranges, bool add_virtual_row) { if (max_rows == 0) throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected at least 1 row to read, got 0."); @@ -961,7 +961,7 @@ MergeTreeRangeReader::ReadResult MergeTreeRangeReader::read(size_t max_rows, Mar if (prev_reader) { - read_result = prev_reader->read(max_rows, ranges); + read_result = prev_reader->read(max_rows, ranges, add_virtual_row); size_t num_read_rows; Columns columns = continueReadingChain(read_result, num_read_rows); @@ -1026,8 +1026,15 @@ MergeTreeRangeReader::ReadResult MergeTreeRangeReader::read(size_t max_rows, Mar } else { + // if (add_virtual_row) + // { + // generate the virtual row + // } + // else + // { read_result = startReadingChain(max_rows, ranges); read_result.num_rows = read_result.numReadRows(); + // } LOG_TEST(log, "First reader returned: {}, requested columns: {}", read_result.dumpInfo(), dumpNames(merge_tree_reader->getColumns())); @@ -1062,7 +1069,11 @@ MergeTreeRangeReader::ReadResult MergeTreeRangeReader::read(size_t max_rows, Mar read_result.addNumBytesRead(total_bytes); } - executePrewhereActionsAndFilterColumns(read_result); + /// If add_virtual_row is enabled, don't turn on prewhere so that virtual row can always pass through. + // if (!add_virtual_row) + // { + executePrewhereActionsAndFilterColumns(read_result); + // } read_result.checkInternalConsistency(); diff --git a/src/Storages/MergeTree/MergeTreeRangeReader.h b/src/Storages/MergeTree/MergeTreeRangeReader.h index 688a6b0922b..d8cf33b0340 100644 --- a/src/Storages/MergeTree/MergeTreeRangeReader.h +++ b/src/Storages/MergeTree/MergeTreeRangeReader.h @@ -300,7 +300,7 @@ public: LoggerPtr log; }; - ReadResult read(size_t max_rows, MarkRanges & ranges); + ReadResult read(size_t max_rows, MarkRanges & ranges, bool add_virtual_row); const Block & getSampleBlock() const { return result_sample_block; } diff --git a/src/Storages/MergeTree/MergeTreeReadTask.cpp b/src/Storages/MergeTree/MergeTreeReadTask.cpp index 08b30e445e2..498c62080a9 100644 --- a/src/Storages/MergeTree/MergeTreeReadTask.cpp +++ b/src/Storages/MergeTree/MergeTreeReadTask.cpp @@ -158,7 +158,13 @@ MergeTreeReadTask::BlockAndProgress MergeTreeReadTask::read(const BlockSizeParam UInt64 recommended_rows = estimateNumRows(params); UInt64 rows_to_read = std::max(static_cast(1), std::min(params.max_block_size_rows, recommended_rows)); - auto read_result = range_readers.main.read(rows_to_read, mark_ranges); + auto read_result = range_readers.main.read(rows_to_read, mark_ranges, add_virtual_row); + + if (add_virtual_row) + { + /// Now we have the virtual row, which is at most once for each part. + add_virtual_row = false; + } /// All rows were filtered. Repeat. if (read_result.num_rows == 0) diff --git a/src/Storages/MergeTree/MergeTreeReadTask.h b/src/Storages/MergeTree/MergeTreeReadTask.h index c8bb501c0e8..73927d62959 100644 --- a/src/Storages/MergeTree/MergeTreeReadTask.h +++ b/src/Storages/MergeTree/MergeTreeReadTask.h @@ -117,6 +117,7 @@ public: size_t row_count = 0; size_t num_read_rows = 0; size_t num_read_bytes = 0; + bool is_virtual_row = false; }; MergeTreeReadTask( @@ -140,6 +141,8 @@ public: static Readers createReaders(const MergeTreeReadTaskInfoPtr & read_info, const Extras & extras, const MarkRanges & ranges); static RangeReaders createRangeReaders(const Readers & readers, const PrewhereExprInfo & prewhere_actions); + void addVirtualRow() { add_virtual_row = true; } + private: UInt64 estimateNumRows(const BlockSizeParams & params) const; @@ -158,6 +161,9 @@ private: /// Used to satistfy preferred_block_size_bytes limitation MergeTreeBlockSizePredictorPtr size_predictor; + + /// If true, add once, and then set false. + bool add_virtual_row = false; }; using MergeTreeReadTaskPtr = std::unique_ptr; diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index fce733d47b7..f61365b0916 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -6,7 +6,7 @@ #include #include #include -#include +#include #include #include #include @@ -134,6 +134,13 @@ ChunkAndProgress MergeTreeSelectProcessor::read() if (!task->getMainRangeReader().isInitialized()) initializeRangeReaders(); + if (add_virtual_row) + { + /// Turn on virtual row just once. + task->addVirtualRow(); + add_virtual_row = false; + } + auto res = algorithm->readFromTask(*task, block_size_params); if (res.row_count) @@ -148,7 +155,9 @@ ChunkAndProgress MergeTreeSelectProcessor::read() } return ChunkAndProgress{ - .chunk = Chunk(ordered_columns, res.row_count, add_part_level ? std::make_shared(task->getInfo().data_part->info.level) : nullptr), + .chunk = Chunk(ordered_columns, res.row_count, + add_part_level || res.is_virtual_row ? std::make_shared( + (add_part_level ? task->getInfo().data_part->info.level : 0), res.is_virtual_row) : nullptr), .num_read_rows = res.num_read_rows, .num_read_bytes = res.num_read_bytes, .is_finished = false}; diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index 01bb3851e04..106190f15c3 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -65,6 +65,8 @@ public: void addPartLevelToChunk(bool add_part_level_) { add_part_level = add_part_level_; } + void addVirtualRowToChunk(bool add_virtual_row_) { add_virtual_row = add_virtual_row_; } + private: /// This struct allow to return block with no columns but with non-zero number of rows similar to Chunk struct BlockAndProgress @@ -99,6 +101,10 @@ private: /// Should we add part level to produced chunk. Part level is useful for next steps if query has FINAL bool add_part_level = false; + /// Should we add a virtual row as the single first chunk. + /// Virtual row is useful for read-in-order optimization when multiple parts exist. + bool add_virtual_row = false; + LoggerPtr log = getLogger("MergeTreeSelectProcessor"); std::atomic is_cancelled{false}; }; diff --git a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp index 81eb166b300..bffea59d5d6 100644 --- a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp +++ b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp @@ -13,7 +13,7 @@ #include #include #include -#include +#include namespace DB { @@ -262,7 +262,8 @@ try ++it; } - return Chunk(std::move(res_columns), rows_read, add_part_level ? std::make_shared(data_part->info.level) : nullptr); + return Chunk(std::move(res_columns), rows_read, + add_part_level ? std::make_shared(data_part->info.level, false) : nullptr); } } else From 72ebd3957251bc0ca9355f80d034b3f7d3083a3e Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Mon, 8 Apr 2024 02:27:54 +0000 Subject: [PATCH 002/680] add simple virtual row --- .../QueryPlan/ReadFromMergeTree.cpp | 2 +- .../MergeTree/MergeTreeRangeReader.cpp | 17 ++----- src/Storages/MergeTree/MergeTreeRangeReader.h | 2 +- src/Storages/MergeTree/MergeTreeReadTask.cpp | 2 +- src/Storages/MergeTree/MergeTreeReadTask.h | 3 -- .../MergeTree/MergeTreeSelectProcessor.cpp | 50 ++++++++++++++----- .../MergeTree/MergeTreeSelectProcessor.h | 8 ++- 7 files changed, 51 insertions(+), 33 deletions(-) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 91cd362f1d9..7f7f2673aee 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -597,7 +597,7 @@ Pipe ReadFromMergeTree::readInOrder( processor->addPartLevelToChunk(isQueryWithFinal()); - processor->addVirtualRowToChunk(need_virtual_row); + processor->addVirtualRowToChunk(need_virtual_row, part_with_ranges.data_part->getIndex()); auto source = std::make_shared(std::move(processor)); if (set_rows_approx) diff --git a/src/Storages/MergeTree/MergeTreeRangeReader.cpp b/src/Storages/MergeTree/MergeTreeRangeReader.cpp index 0456a8e2787..6932762f58b 100644 --- a/src/Storages/MergeTree/MergeTreeRangeReader.cpp +++ b/src/Storages/MergeTree/MergeTreeRangeReader.cpp @@ -946,7 +946,7 @@ String addDummyColumnWithRowCount(Block & block, size_t num_rows) } -MergeTreeRangeReader::ReadResult MergeTreeRangeReader::read(size_t max_rows, MarkRanges & ranges, bool add_virtual_row) +MergeTreeRangeReader::ReadResult MergeTreeRangeReader::read(size_t max_rows, MarkRanges & ranges) { if (max_rows == 0) throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected at least 1 row to read, got 0."); @@ -961,7 +961,7 @@ MergeTreeRangeReader::ReadResult MergeTreeRangeReader::read(size_t max_rows, Mar if (prev_reader) { - read_result = prev_reader->read(max_rows, ranges, add_virtual_row); + read_result = prev_reader->read(max_rows, ranges); size_t num_read_rows; Columns columns = continueReadingChain(read_result, num_read_rows); @@ -1026,15 +1026,8 @@ MergeTreeRangeReader::ReadResult MergeTreeRangeReader::read(size_t max_rows, Mar } else { - // if (add_virtual_row) - // { - // generate the virtual row - // } - // else - // { read_result = startReadingChain(max_rows, ranges); read_result.num_rows = read_result.numReadRows(); - // } LOG_TEST(log, "First reader returned: {}, requested columns: {}", read_result.dumpInfo(), dumpNames(merge_tree_reader->getColumns())); @@ -1069,11 +1062,7 @@ MergeTreeRangeReader::ReadResult MergeTreeRangeReader::read(size_t max_rows, Mar read_result.addNumBytesRead(total_bytes); } - /// If add_virtual_row is enabled, don't turn on prewhere so that virtual row can always pass through. - // if (!add_virtual_row) - // { - executePrewhereActionsAndFilterColumns(read_result); - // } + executePrewhereActionsAndFilterColumns(read_result); read_result.checkInternalConsistency(); diff --git a/src/Storages/MergeTree/MergeTreeRangeReader.h b/src/Storages/MergeTree/MergeTreeRangeReader.h index d8cf33b0340..688a6b0922b 100644 --- a/src/Storages/MergeTree/MergeTreeRangeReader.h +++ b/src/Storages/MergeTree/MergeTreeRangeReader.h @@ -300,7 +300,7 @@ public: LoggerPtr log; }; - ReadResult read(size_t max_rows, MarkRanges & ranges, bool add_virtual_row); + ReadResult read(size_t max_rows, MarkRanges & ranges); const Block & getSampleBlock() const { return result_sample_block; } diff --git a/src/Storages/MergeTree/MergeTreeReadTask.cpp b/src/Storages/MergeTree/MergeTreeReadTask.cpp index 498c62080a9..3c4d121195f 100644 --- a/src/Storages/MergeTree/MergeTreeReadTask.cpp +++ b/src/Storages/MergeTree/MergeTreeReadTask.cpp @@ -158,7 +158,7 @@ MergeTreeReadTask::BlockAndProgress MergeTreeReadTask::read(const BlockSizeParam UInt64 recommended_rows = estimateNumRows(params); UInt64 rows_to_read = std::max(static_cast(1), std::min(params.max_block_size_rows, recommended_rows)); - auto read_result = range_readers.main.read(rows_to_read, mark_ranges, add_virtual_row); + auto read_result = range_readers.main.read(rows_to_read, mark_ranges); if (add_virtual_row) { diff --git a/src/Storages/MergeTree/MergeTreeReadTask.h b/src/Storages/MergeTree/MergeTreeReadTask.h index 73927d62959..709fc73f16e 100644 --- a/src/Storages/MergeTree/MergeTreeReadTask.h +++ b/src/Storages/MergeTree/MergeTreeReadTask.h @@ -117,7 +117,6 @@ public: size_t row_count = 0; size_t num_read_rows = 0; size_t num_read_bytes = 0; - bool is_virtual_row = false; }; MergeTreeReadTask( @@ -141,8 +140,6 @@ public: static Readers createReaders(const MergeTreeReadTaskInfoPtr & read_info, const Extras & extras, const MarkRanges & ranges); static RangeReaders createRangeReaders(const Readers & readers, const PrewhereExprInfo & prewhere_actions); - void addVirtualRow() { add_virtual_row = true; } - private: UInt64 estimateNumRows(const BlockSizeParams & params) const; diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index f61365b0916..d75802c68f3 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -133,38 +133,64 @@ ChunkAndProgress MergeTreeSelectProcessor::read() if (!task->getMainRangeReader().isInitialized()) initializeRangeReaders(); - + add_virtual_row = false; if (add_virtual_row) { /// Turn on virtual row just once. - task->addVirtualRow(); add_virtual_row = false; - } - auto res = algorithm->readFromTask(*task, block_size_params); + const auto & primary_key = storage_snapshot->metadata->primary_key; + + MergeTreeReadTask::BlockAndProgress res; + res.row_count = 1; - if (res.row_count) - { /// Reorder the columns according to result_header Columns ordered_columns; ordered_columns.reserve(result_header.columns()); for (size_t i = 0; i < result_header.columns(); ++i) { - auto name = result_header.getByPosition(i).name; - ordered_columns.push_back(res.block.getByName(name).column); + // TODO: composite pk??? + const ColumnWithTypeAndName & type_and_name = result_header.getByPosition(i); + if (type_and_name.name == primary_key.column_names[0] && type_and_name.type == primary_key.data_types[0]) + ordered_columns.push_back(index[0]->cloneResized(1)); // TODO: use the first range pk whose range might contain results + else + ordered_columns.push_back(type_and_name.type->createColumn()->cloneResized(1)); } return ChunkAndProgress{ - .chunk = Chunk(ordered_columns, res.row_count, - add_part_level || res.is_virtual_row ? std::make_shared( - (add_part_level ? task->getInfo().data_part->info.level : 0), res.is_virtual_row) : nullptr), + .chunk = Chunk(ordered_columns, res.row_count, std::make_shared( + (add_part_level ? task->getInfo().data_part->info.level : 0), true)), .num_read_rows = res.num_read_rows, .num_read_bytes = res.num_read_bytes, .is_finished = false}; } else { - return {Chunk(), res.num_read_rows, res.num_read_bytes, false}; + auto res = algorithm->readFromTask(*task, block_size_params); + + if (res.row_count) + { + /// Reorder the columns according to result_header + Columns ordered_columns; + ordered_columns.reserve(result_header.columns()); + for (size_t i = 0; i < result_header.columns(); ++i) + { + auto name = result_header.getByPosition(i).name; + ordered_columns.push_back(res.block.getByName(name).column); + } + + return ChunkAndProgress{ + .chunk = Chunk(ordered_columns, res.row_count, + add_part_level ? std::make_shared( + (add_part_level ? task->getInfo().data_part->info.level : 0), false) : nullptr), + .num_read_rows = res.num_read_rows, + .num_read_bytes = res.num_read_bytes, + .is_finished = false}; + } + else + { + return {Chunk(), res.num_read_rows, res.num_read_bytes, false}; + } } } diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index 106190f15c3..67a03ca2533 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -65,7 +65,11 @@ public: void addPartLevelToChunk(bool add_part_level_) { add_part_level = add_part_level_; } - void addVirtualRowToChunk(bool add_virtual_row_) { add_virtual_row = add_virtual_row_; } + void addVirtualRowToChunk(bool add_virtual_row_, const Columns& index_) + { + add_virtual_row = add_virtual_row_; + index = index_; + } private: /// This struct allow to return block with no columns but with non-zero number of rows similar to Chunk @@ -105,6 +109,8 @@ private: /// Virtual row is useful for read-in-order optimization when multiple parts exist. bool add_virtual_row = false; + Columns index; + LoggerPtr log = getLogger("MergeTreeSelectProcessor"); std::atomic is_cancelled{false}; }; From 57a2a20900176da28b73f027fc298f7cb7f91781 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Wed, 10 Apr 2024 04:02:15 +0000 Subject: [PATCH 003/680] support composite pk --- src/Storages/MergeTree/MergeTreeSelectProcessor.cpp | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index d75802c68f3..868e757e135 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -133,7 +133,7 @@ ChunkAndProgress MergeTreeSelectProcessor::read() if (!task->getMainRangeReader().isInitialized()) initializeRangeReaders(); - add_virtual_row = false; + if (add_virtual_row) { /// Turn on virtual row just once. @@ -147,12 +147,14 @@ ChunkAndProgress MergeTreeSelectProcessor::read() /// Reorder the columns according to result_header Columns ordered_columns; ordered_columns.reserve(result_header.columns()); - for (size_t i = 0; i < result_header.columns(); ++i) + for (size_t i = 0, j = 0; i < result_header.columns(); ++i) { - // TODO: composite pk??? const ColumnWithTypeAndName & type_and_name = result_header.getByPosition(i); - if (type_and_name.name == primary_key.column_names[0] && type_and_name.type == primary_key.data_types[0]) - ordered_columns.push_back(index[0]->cloneResized(1)); // TODO: use the first range pk whose range might contain results + if (j < index.size() && type_and_name.name == primary_key.column_names[j] && type_and_name.type == primary_key.data_types[j]) + { + ordered_columns.push_back(index[j]->cloneResized(1)); // TODO: use the first range pk whose range might contain results + ++j; + } else ordered_columns.push_back(type_and_name.type->createColumn()->cloneResized(1)); } From bd4385f969c5139870dcfc0ce9d72e6029ab9a59 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Mon, 15 Apr 2024 23:28:16 +0000 Subject: [PATCH 004/680] add test --- .../Merges/Algorithms/MergeTreeReadInfo.h | 2 +- ...03031_read_in_order_optimization.reference | 5 ++ .../03031_read_in_order_optimization.sql | 48 +++++++++++++++++++ 3 files changed, 54 insertions(+), 1 deletion(-) create mode 100644 tests/queries/0_stateless/03031_read_in_order_optimization.reference create mode 100644 tests/queries/0_stateless/03031_read_in_order_optimization.sql diff --git a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h index e79df0fb8c8..ca4bccb235f 100644 --- a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h +++ b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h @@ -29,7 +29,7 @@ inline bool getVirtualRowFromChunk(const Chunk & chunk) const auto & info = chunk.getChunkInfo(); if (const auto * read_info = typeid_cast(info.get())) return read_info->virtual_row; - return 0; + return false; } } diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.reference b/tests/queries/0_stateless/03031_read_in_order_optimization.reference new file mode 100644 index 00000000000..304f7f7a049 --- /dev/null +++ b/tests/queries/0_stateless/03031_read_in_order_optimization.reference @@ -0,0 +1,5 @@ +0 +1 +2 +3 +24578 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.sql b/tests/queries/0_stateless/03031_read_in_order_optimization.sql new file mode 100644 index 00000000000..eecbfe64f6d --- /dev/null +++ b/tests/queries/0_stateless/03031_read_in_order_optimization.sql @@ -0,0 +1,48 @@ + +DROP TABLE IF EXISTS t; + +CREATE TABLE t +( + `x` UInt64, + `y` UInt64, + `z` UInt64, + `k` UInt64 +) +ENGINE = MergeTree +ORDER BY (x, y, z) +SETTINGS index_granularity = 8192; + +INSERT INTO t SELECT + number, + number, + number, + number +FROM numbers(8192 * 3); + +INSERT INTO t SELECT + number + (8192 * 3), + number + (8192 * 3), + number + (8192 * 3), + number + (8192 * 3) +FROM numbers(8192 * 3); + +SELECT x +FROM t +ORDER BY x ASC +LIMIT 4 +SETTINGS max_block_size = 8192, +read_in_order_two_level_merge_threshold = 0, +max_threads = 1, +optimize_read_in_order = 1; + +SYSTEM FLUSH LOGS; + +SELECT read_rows +FROM system.query_log +WHERE current_database = currentDatabase() +AND query like '%SELECT x%' +AND query not like '%system.query_log%' +ORDER BY query_start_time DESC, read_rows DESC +LIMIT 1; + +DROP TABLE t; \ No newline at end of file From cc3fd0e73693e877967b1f5572d8d6779088fa06 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Tue, 23 Apr 2024 02:43:49 +0000 Subject: [PATCH 005/680] minor change --- src/Storages/MergeTree/MergeTreeSelectProcessor.cpp | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index 868e757e135..1f97fec2013 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -150,13 +150,17 @@ ChunkAndProgress MergeTreeSelectProcessor::read() for (size_t i = 0, j = 0; i < result_header.columns(); ++i) { const ColumnWithTypeAndName & type_and_name = result_header.getByPosition(i); + ColumnPtr current_column = type_and_name.type->createColumn(); + if (j < index.size() && type_and_name.name == primary_key.column_names[j] && type_and_name.type == primary_key.data_types[j]) { - ordered_columns.push_back(index[j]->cloneResized(1)); // TODO: use the first range pk whose range might contain results + auto column = current_column->cloneEmpty(); + column->insert((*index[j])[0]); // TODO: use the first range pk whose range might contain results + ordered_columns.push_back(std::move(column)); ++j; } else - ordered_columns.push_back(type_and_name.type->createColumn()->cloneResized(1)); + ordered_columns.push_back(current_column->cloneResized(1)); } return ChunkAndProgress{ From 7f6d6400230eb90e27a99226c89ac0c5acb0d709 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Wed, 24 Apr 2024 01:14:04 +0000 Subject: [PATCH 006/680] use a better range begin in virtual row --- src/Processors/QueryPlan/ReadFromMergeTree.cpp | 3 ++- src/Storages/MergeTree/MergeTreeSelectProcessor.cpp | 2 +- src/Storages/MergeTree/MergeTreeSelectProcessor.h | 7 +++++-- 3 files changed, 8 insertions(+), 4 deletions(-) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 7f7f2673aee..f873bcb6104 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -597,7 +597,8 @@ Pipe ReadFromMergeTree::readInOrder( processor->addPartLevelToChunk(isQueryWithFinal()); - processor->addVirtualRowToChunk(need_virtual_row, part_with_ranges.data_part->getIndex()); + processor->addVirtualRowToChunk(need_virtual_row, part_with_ranges.data_part->getIndex(), + part_with_ranges.ranges.front().begin); auto source = std::make_shared(std::move(processor)); if (set_rows_approx) diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index 1f97fec2013..a3fcfad3bb5 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -155,7 +155,7 @@ ChunkAndProgress MergeTreeSelectProcessor::read() if (j < index.size() && type_and_name.name == primary_key.column_names[j] && type_and_name.type == primary_key.data_types[j]) { auto column = current_column->cloneEmpty(); - column->insert((*index[j])[0]); // TODO: use the first range pk whose range might contain results + column->insert((*index[j])[mark_range_begin]); ordered_columns.push_back(std::move(column)); ++j; } diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index 67a03ca2533..352f771f9ce 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -65,10 +65,11 @@ public: void addPartLevelToChunk(bool add_part_level_) { add_part_level = add_part_level_; } - void addVirtualRowToChunk(bool add_virtual_row_, const Columns& index_) + void addVirtualRowToChunk(bool add_virtual_row_, const Columns& index_, size_t mark_range_begin_) { add_virtual_row = add_virtual_row_; index = index_; + mark_range_begin = mark_range_begin_; } private: @@ -108,8 +109,10 @@ private: /// Should we add a virtual row as the single first chunk. /// Virtual row is useful for read-in-order optimization when multiple parts exist. bool add_virtual_row = false; - + /// PK index used in virtual row. Columns index; + /// The first range that might contain the candidate, used in virtual row. + size_t mark_range_begin; LoggerPtr log = getLogger("MergeTreeSelectProcessor"); std::atomic is_cancelled{false}; From ba049d85b3126b766575f958b61fc2f84bb3a11b Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Fri, 26 Apr 2024 02:45:50 +0000 Subject: [PATCH 007/680] fix test --- tests/queries/0_stateless/03031_read_in_order_optimization.sql | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.sql b/tests/queries/0_stateless/03031_read_in_order_optimization.sql index eecbfe64f6d..f114a838ff3 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization.sql @@ -10,7 +10,8 @@ CREATE TABLE t ) ENGINE = MergeTree ORDER BY (x, y, z) -SETTINGS index_granularity = 8192; +SETTINGS index_granularity = 8192, +index_granularity_bytes = 10485760; INSERT INTO t SELECT number, From 86c7488647750f65d7a75dd4774f84fcf44f763b Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Sat, 4 May 2024 02:09:17 +0000 Subject: [PATCH 008/680] only read one chunk in mergetramsform when meet virtual row --- src/Processors/Merges/IMergingTransform.cpp | 7 +++++-- .../MergeTree/MergeTreeSelectProcessor.h | 2 +- ...03031_read_in_order_optimization.reference | 2 +- .../03031_read_in_order_optimization.sql | 20 +++++++++++++++++++ 4 files changed, 27 insertions(+), 4 deletions(-) diff --git a/src/Processors/Merges/IMergingTransform.cpp b/src/Processors/Merges/IMergingTransform.cpp index fbb47969b2f..50b3e2ca634 100644 --- a/src/Processors/Merges/IMergingTransform.cpp +++ b/src/Processors/Merges/IMergingTransform.cpp @@ -1,3 +1,4 @@ +#include #include namespace DB @@ -101,8 +102,10 @@ IProcessor::Status IMergingTransformBase::prepareInitializeInputs() /// setNotNeeded after reading first chunk, because in optimismtic case /// (e.g. with optimized 'ORDER BY primary_key LIMIT n' and small 'n') /// we won't have to read any chunks anymore; - auto chunk = input.pull(limit_hint != 0); - if ((limit_hint && chunk.getNumRows() < limit_hint) || always_read_till_end) + /// If virtual row exists, test it first, so don't read more chunks. + auto chunk = input.pull(true); + if ((limit_hint == 0 && !getVirtualRowFromChunk(chunk)) + || (limit_hint && chunk.getNumRows() < limit_hint) || always_read_till_end) input.setNeeded(); if (!chunk.hasRows()) diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index 352f771f9ce..255b4c65ff9 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -65,7 +65,7 @@ public: void addPartLevelToChunk(bool add_part_level_) { add_part_level = add_part_level_; } - void addVirtualRowToChunk(bool add_virtual_row_, const Columns& index_, size_t mark_range_begin_) + void addVirtualRowToChunk(bool add_virtual_row_, const Columns & index_, size_t mark_range_begin_) { add_virtual_row = add_virtual_row_; index = index_; diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.reference b/tests/queries/0_stateless/03031_read_in_order_optimization.reference index 304f7f7a049..70d79aecf43 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization.reference +++ b/tests/queries/0_stateless/03031_read_in_order_optimization.reference @@ -2,4 +2,4 @@ 1 2 3 -24578 +16386 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.sql b/tests/queries/0_stateless/03031_read_in_order_optimization.sql index f114a838ff3..999d2e265d0 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization.sql @@ -46,4 +46,24 @@ AND query not like '%system.query_log%' ORDER BY query_start_time DESC, read_rows DESC LIMIT 1; +-- SELECT x +-- FROM t +-- ORDER BY x ASC +-- LIMIT 4 +-- SETTINGS max_block_size = 8192, +-- read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge +-- max_threads = 1, +-- optimize_read_in_order = 1; + +-- SYSTEM FLUSH LOGS; + +-- -- without virtual row 16.38k, but with virtual row 24.58k, becasue read again (why?) in the non-target part after reading its virtual row and before sending the virtual row to the priority queue +-- SELECT read_rows +-- FROM system.query_log +-- WHERE current_database = currentDatabase() +-- AND query like '%SELECT x%' +-- AND query not like '%system.query_log%' +-- ORDER BY query_start_time DESC, read_rows DESC +-- LIMIT 1; + DROP TABLE t; \ No newline at end of file From 04a757eb71a0cdff42f804043025dd6e900e9283 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Sat, 4 May 2024 17:37:56 +0000 Subject: [PATCH 009/680] fix --- src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp | 2 +- src/Processors/QueryPlan/ReadFromMergeTree.cpp | 5 +++-- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp index 7592f37ba22..7da73349c4a 100644 --- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp @@ -234,7 +234,7 @@ IMergingAlgorithm::Status MergingSortedAlgorithm::mergeBatchImpl(TSortingQueue & { /// If virtual row is detected, there should be only one row as a single chunk, /// and always skip this chunk to pull the next one. - assert(initial_batch_size == 1); + chassert(initial_batch_size == 1); queue.removeTop(); return Status(current.impl->order); } diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index cdf301f8044..d7e7f9ae758 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -598,8 +598,9 @@ Pipe ReadFromMergeTree::readInOrder( processor->addPartLevelToChunk(isQueryWithFinal()); - processor->addVirtualRowToChunk(need_virtual_row, part_with_ranges.data_part->getIndex(), - part_with_ranges.ranges.front().begin); + auto primary_key_index = part_with_ranges.data_part->getIndex(); + chassert(primary_key_index); + processor->addVirtualRowToChunk(need_virtual_row, *primary_key_index, part_with_ranges.ranges.front().begin); auto source = std::make_shared(std::move(processor)); if (set_rows_approx) From 1c2c3aed249ea77f0d667e1a9173ca39a5a88858 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Mon, 6 May 2024 13:25:19 +0000 Subject: [PATCH 010/680] support non-preliminary merge case --- src/Processors/Merges/IMergingTransform.cpp | 9 +++-- .../QueryPlan/ReadFromMergeTree.cpp | 8 ++--- .../MergeTree/MergeTreeSelectProcessor.cpp | 4 +-- .../MergeTree/MergeTreeSelectProcessor.h | 4 +-- ...03031_read_in_order_optimization.reference | 5 +++ .../03031_read_in_order_optimization.sql | 33 +++++++++---------- 6 files changed, 34 insertions(+), 29 deletions(-) diff --git a/src/Processors/Merges/IMergingTransform.cpp b/src/Processors/Merges/IMergingTransform.cpp index 50b3e2ca634..3daeca254ed 100644 --- a/src/Processors/Merges/IMergingTransform.cpp +++ b/src/Processors/Merges/IMergingTransform.cpp @@ -102,10 +102,13 @@ IProcessor::Status IMergingTransformBase::prepareInitializeInputs() /// setNotNeeded after reading first chunk, because in optimismtic case /// (e.g. with optimized 'ORDER BY primary_key LIMIT n' and small 'n') /// we won't have to read any chunks anymore; - /// If virtual row exists, test it first, so don't read more chunks. + /// If virtual row exists, let it pass through, so don't read more chunks. auto chunk = input.pull(true); - if ((limit_hint == 0 && !getVirtualRowFromChunk(chunk)) - || (limit_hint && chunk.getNumRows() < limit_hint) || always_read_till_end) + bool virtual_row = getVirtualRowFromChunk(chunk); + if (limit_hint == 0 && !virtual_row) + input.setNeeded(); + + if (!virtual_row && ((limit_hint && chunk.getNumRows() < limit_hint) || always_read_till_end)) input.setNeeded(); if (!chunk.hasRows()) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index d7e7f9ae758..4386d435732 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -597,10 +597,8 @@ Pipe ReadFromMergeTree::readInOrder( actions_settings, block_size, reader_settings); processor->addPartLevelToChunk(isQueryWithFinal()); - - auto primary_key_index = part_with_ranges.data_part->getIndex(); - chassert(primary_key_index); - processor->addVirtualRowToChunk(need_virtual_row, *primary_key_index, part_with_ranges.ranges.front().begin); + processor->addVirtualRowToChunk(need_virtual_row, part_with_ranges.data_part->getIndex(), + part_with_ranges.ranges.front().begin); auto source = std::make_shared(std::move(processor)); if (set_rows_approx) @@ -1037,7 +1035,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( { /// need_virtual_row = true means a MergingSortedTransform should occur. /// If so, adding a virtual row might speedup in the case of multiple parts. - bool need_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; + bool need_virtual_row = item.size() > 1; pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit, need_virtual_row)); } } diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index a3fcfad3bb5..4feef5115bf 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -152,10 +152,10 @@ ChunkAndProgress MergeTreeSelectProcessor::read() const ColumnWithTypeAndName & type_and_name = result_header.getByPosition(i); ColumnPtr current_column = type_and_name.type->createColumn(); - if (j < index.size() && type_and_name.name == primary_key.column_names[j] && type_and_name.type == primary_key.data_types[j]) + if (j < index->size() && type_and_name.name == primary_key.column_names[j] && type_and_name.type == primary_key.data_types[j]) { auto column = current_column->cloneEmpty(); - column->insert((*index[j])[mark_range_begin]); + column->insert((*(*index)[j])[mark_range_begin]); ordered_columns.push_back(std::move(column)); ++j; } diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index 3dab11b556c..7a562c1a115 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -65,7 +65,7 @@ public: void addPartLevelToChunk(bool add_part_level_) { add_part_level = add_part_level_; } - void addVirtualRowToChunk(bool add_virtual_row_, const Columns & index_, size_t mark_range_begin_) + void addVirtualRowToChunk(bool add_virtual_row_, const IMergeTreeDataPart::Index & index_, size_t mark_range_begin_) { add_virtual_row = add_virtual_row_; index = index_; @@ -101,7 +101,7 @@ private: /// Virtual row is useful for read-in-order optimization when multiple parts exist. bool add_virtual_row = false; /// PK index used in virtual row. - Columns index; + IMergeTreeDataPart::Index index; /// The first range that might contain the candidate, used in virtual row. size_t mark_range_begin; diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.reference b/tests/queries/0_stateless/03031_read_in_order_optimization.reference index 70d79aecf43..62e8669fbe0 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization.reference +++ b/tests/queries/0_stateless/03031_read_in_order_optimization.reference @@ -3,3 +3,8 @@ 2 3 16386 +0 +1 +2 +3 +16386 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.sql b/tests/queries/0_stateless/03031_read_in_order_optimization.sql index 999d2e265d0..57f9838392f 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization.sql @@ -46,24 +46,23 @@ AND query not like '%system.query_log%' ORDER BY query_start_time DESC, read_rows DESC LIMIT 1; --- SELECT x --- FROM t --- ORDER BY x ASC --- LIMIT 4 --- SETTINGS max_block_size = 8192, --- read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge --- max_threads = 1, --- optimize_read_in_order = 1; +SELECT x +FROM t +ORDER BY x ASC +LIMIT 4 +SETTINGS max_block_size = 8192, +read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge +max_threads = 1, +optimize_read_in_order = 1; --- SYSTEM FLUSH LOGS; +SYSTEM FLUSH LOGS; --- -- without virtual row 16.38k, but with virtual row 24.58k, becasue read again (why?) in the non-target part after reading its virtual row and before sending the virtual row to the priority queue --- SELECT read_rows --- FROM system.query_log --- WHERE current_database = currentDatabase() --- AND query like '%SELECT x%' --- AND query not like '%system.query_log%' --- ORDER BY query_start_time DESC, read_rows DESC --- LIMIT 1; +SELECT read_rows +FROM system.query_log +WHERE current_database = currentDatabase() +AND query like '%SELECT x%' +AND query not like '%system.query_log%' +ORDER BY query_start_time DESC, read_rows DESC +LIMIT 1; DROP TABLE t; \ No newline at end of file From 0537b8c833b69638c3868497f935f9bb7cf46e0a Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Wed, 8 May 2024 00:17:37 +0000 Subject: [PATCH 011/680] restrict to preliminary merge and add more tests --- .../QueryPlan/ReadFromMergeTree.cpp | 2 +- ...03031_read_in_order_optimization.reference | 7 ++- .../03031_read_in_order_optimization.sql | 48 +++++++++++++++---- 3 files changed, 45 insertions(+), 12 deletions(-) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 4386d435732..9a0469f49a8 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -1035,7 +1035,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( { /// need_virtual_row = true means a MergingSortedTransform should occur. /// If so, adding a virtual row might speedup in the case of multiple parts. - bool need_virtual_row = item.size() > 1; + bool need_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit, need_virtual_row)); } } diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.reference b/tests/queries/0_stateless/03031_read_in_order_optimization.reference index 62e8669fbe0..c73f79d8dce 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization.reference +++ b/tests/queries/0_stateless/03031_read_in_order_optimization.reference @@ -3,8 +3,13 @@ 2 3 16386 +16385 +16386 +16387 +16388 +24578 0 1 2 3 -16386 +16384 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.sql b/tests/queries/0_stateless/03031_read_in_order_optimization.sql index 57f9838392f..597845564e4 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization.sql @@ -24,28 +24,55 @@ INSERT INTO t SELECT number + (8192 * 3), number + (8192 * 3), number + (8192 * 3), - number + (8192 * 3) + number FROM numbers(8192 * 3); +-- Expecting 2 virtual rows + one chunk (8192) for result + one extra chunk for next consumption in merge transform (8192), +-- both chunks come from the same part. SELECT x FROM t ORDER BY x ASC LIMIT 4 SETTINGS max_block_size = 8192, -read_in_order_two_level_merge_threshold = 0, +read_in_order_two_level_merge_threshold = 0, --force preliminary merge max_threads = 1, -optimize_read_in_order = 1; +optimize_read_in_order = 1, +log_comment = 'no filter'; SYSTEM FLUSH LOGS; SELECT read_rows FROM system.query_log WHERE current_database = currentDatabase() -AND query like '%SELECT x%' -AND query not like '%system.query_log%' -ORDER BY query_start_time DESC, read_rows DESC +AND log_comment = 'no filter' +AND type = 'QueryFinish' +ORDER BY query_start_time DESC +limit 1; + +-- Expecting 2 virtual rows + two chunks (8192*2) get filtered out + one chunk for result (8192), +-- all chunks come from the same part. +SELECT k +FROM t +WHERE k > 8192 * 2 +ORDER BY x ASC +LIMIT 4 +SETTINGS max_block_size = 8192, +read_in_order_two_level_merge_threshold = 0, --force preliminary merge +max_threads = 1, +optimize_read_in_order = 1, +log_comment = 'with filter'; + +SYSTEM FLUSH LOGS; + +SELECT read_rows +FROM system.query_log +WHERE current_database = currentDatabase() +AND log_comment = 'with filter' +AND type = 'QueryFinish' +ORDER BY query_start_time DESC LIMIT 1; +-- Should not impact cases without preliminary merge (might read again when chunk row is less than limit) SELECT x FROM t ORDER BY x ASC @@ -53,16 +80,17 @@ LIMIT 4 SETTINGS max_block_size = 8192, read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge max_threads = 1, -optimize_read_in_order = 1; +optimize_read_in_order = 1, +log_comment = 'no impact'; SYSTEM FLUSH LOGS; SELECT read_rows FROM system.query_log WHERE current_database = currentDatabase() -AND query like '%SELECT x%' -AND query not like '%system.query_log%' -ORDER BY query_start_time DESC, read_rows DESC +AND log_comment = 'no impact' +AND type = 'QueryFinish' +ORDER BY query_start_time DESC LIMIT 1; DROP TABLE t; \ No newline at end of file From 8f8ba55ac3cc254d4a890ed6f45cb5a4ef411143 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Tue, 14 May 2024 19:43:47 +0000 Subject: [PATCH 012/680] add check flag --- .../Merges/Algorithms/MergeTreeReadInfo.h | 1 + .../QueryPlan/ReadFromMergeTree.cpp | 5 ++- src/Processors/QueryPlan/SortingStep.cpp | 31 ++++++++++++++ src/QueryPipeline/QueryPipelineBuilder.h | 2 + .../MergeTree/MergeTreeSelectProcessor.cpp | 4 +- .../MergeTree/MergeTreeSelectProcessor.h | 7 ++-- src/Storages/MergeTree/MergeTreeSource.h | 2 + ...03031_read_in_order_optimization.reference | 7 +++- .../03031_read_in_order_optimization.sql | 40 +++++++++++++++---- 9 files changed, 84 insertions(+), 15 deletions(-) diff --git a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h index ca4bccb235f..52ca92b471a 100644 --- a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h +++ b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h @@ -13,6 +13,7 @@ public: explicit MergeTreeReadInfo(size_t part_level, bool virtual_row_) : origin_merge_tree_part_level(part_level), virtual_row(virtual_row_) { } size_t origin_merge_tree_part_level = 0; + /// If virtual_row is true, the chunk must contain the virtual row only. bool virtual_row = false; }; diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 9a0469f49a8..2f1db9539a6 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -597,8 +597,9 @@ Pipe ReadFromMergeTree::readInOrder( actions_settings, block_size, reader_settings); processor->addPartLevelToChunk(isQueryWithFinal()); - processor->addVirtualRowToChunk(need_virtual_row, part_with_ranges.data_part->getIndex(), - part_with_ranges.ranges.front().begin); + processor->addVirtualRowToChunk(part_with_ranges.data_part->getIndex(), part_with_ranges.ranges.front().begin); + if (need_virtual_row) + processor->enableVirtualRow(); auto source = std::make_shared(std::move(processor)); if (set_rows_approx) diff --git a/src/Processors/QueryPlan/SortingStep.cpp b/src/Processors/QueryPlan/SortingStep.cpp index d0491cb4b82..d728e8fb154 100644 --- a/src/Processors/QueryPlan/SortingStep.cpp +++ b/src/Processors/QueryPlan/SortingStep.cpp @@ -13,6 +13,9 @@ #include #include +#include +#include +#include namespace CurrentMetrics { @@ -243,6 +246,34 @@ void SortingStep::mergingSorted(QueryPipelineBuilder & pipeline, const SortDescr /// If there are several streams, then we merge them into one if (pipeline.getNumStreams() > 1) { + /// We check every step of this pipeline, to make sure virtual row can work correctly. + /// Currently ExpressionTransform is supported, should add other processors if possible. + const auto& pipe = pipeline.getPipe(); + bool enable_virtual_row = true; + std::vector> merge_tree_sources; + for (const auto & processor : pipe.getProcessors()) + { + if (auto merge_tree_source = std::dynamic_pointer_cast(processor)) + { + merge_tree_sources.push_back(merge_tree_source); + } + else if (!std::dynamic_pointer_cast(processor)) + { + enable_virtual_row = false; + break; + } + } + + /// If everything is okay, we enable virtual row in MergeTreeSelectProcessor + if (enable_virtual_row && merge_tree_sources.size() >= 2) + { + for (const auto & merge_tree_source : merge_tree_sources) + { + const auto& merge_tree_select_processor = merge_tree_source->getProcessor(); + merge_tree_select_processor->enableVirtualRow(); + } + } + auto transform = std::make_shared( pipeline.getHeader(), pipeline.getNumStreams(), diff --git a/src/QueryPipeline/QueryPipelineBuilder.h b/src/QueryPipeline/QueryPipelineBuilder.h index f0b2ead687e..50a77360d46 100644 --- a/src/QueryPipeline/QueryPipelineBuilder.h +++ b/src/QueryPipeline/QueryPipelineBuilder.h @@ -197,6 +197,8 @@ public: void setQueryIdHolder(std::shared_ptr query_id_holder) { resources.query_id_holders.emplace_back(std::move(query_id_holder)); } void addContext(ContextPtr context) { resources.interpreter_context.emplace_back(std::move(context)); } + const Pipe& getPipe() const { return pipe; } + /// Convert query pipeline to pipe. static Pipe getPipe(QueryPipelineBuilder pipeline, QueryPlanResourceHolder & resources); static QueryPipeline getPipeline(QueryPipelineBuilder builder); diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index 4feef5115bf..0f4b68ddde9 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -134,10 +134,10 @@ ChunkAndProgress MergeTreeSelectProcessor::read() if (!task->getMainRangeReader().isInitialized()) initializeRangeReaders(); - if (add_virtual_row) + if (enable_virtual_row) { /// Turn on virtual row just once. - add_virtual_row = false; + enable_virtual_row = false; const auto & primary_key = storage_snapshot->metadata->primary_key; diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index 7a562c1a115..57da1039ba9 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -65,13 +65,14 @@ public: void addPartLevelToChunk(bool add_part_level_) { add_part_level = add_part_level_; } - void addVirtualRowToChunk(bool add_virtual_row_, const IMergeTreeDataPart::Index & index_, size_t mark_range_begin_) + void addVirtualRowToChunk(const IMergeTreeDataPart::Index & index_, size_t mark_range_begin_) { - add_virtual_row = add_virtual_row_; index = index_; mark_range_begin = mark_range_begin_; } + void enableVirtualRow() { enable_virtual_row = true; } + private: /// Sets up range readers corresponding to data readers void initializeRangeReaders(); @@ -99,7 +100,7 @@ private: /// Should we add a virtual row as the single first chunk. /// Virtual row is useful for read-in-order optimization when multiple parts exist. - bool add_virtual_row = false; + bool enable_virtual_row = false; /// PK index used in virtual row. IMergeTreeDataPart::Index index; /// The first range that might contain the candidate, used in virtual row. diff --git a/src/Storages/MergeTree/MergeTreeSource.h b/src/Storages/MergeTree/MergeTreeSource.h index 655f0ee6ebe..486b8be2fef 100644 --- a/src/Storages/MergeTree/MergeTreeSource.h +++ b/src/Storages/MergeTree/MergeTreeSource.h @@ -19,6 +19,8 @@ public: Status prepare() override; + const MergeTreeSelectProcessorPtr& getProcessor() const { return processor; } + #if defined(OS_LINUX) int schedule() override; #endif diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.reference b/tests/queries/0_stateless/03031_read_in_order_optimization.reference index c73f79d8dce..c7cce7e60e9 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization.reference +++ b/tests/queries/0_stateless/03031_read_in_order_optimization.reference @@ -12,4 +12,9 @@ 1 2 3 -16384 +16386 +16385 +16386 +16387 +16388 +24578 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.sql b/tests/queries/0_stateless/03031_read_in_order_optimization.sql index 597845564e4..332ee7f58dc 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization.sql @@ -27,6 +27,8 @@ INSERT INTO t SELECT number FROM numbers(8192 * 3); +SYSTEM STOP MERGES t; + -- Expecting 2 virtual rows + one chunk (8192) for result + one extra chunk for next consumption in merge transform (8192), -- both chunks come from the same part. SELECT x @@ -37,14 +39,14 @@ SETTINGS max_block_size = 8192, read_in_order_two_level_merge_threshold = 0, --force preliminary merge max_threads = 1, optimize_read_in_order = 1, -log_comment = 'no filter'; +log_comment = 'preliminary merge, no filter'; SYSTEM FLUSH LOGS; SELECT read_rows FROM system.query_log WHERE current_database = currentDatabase() -AND log_comment = 'no filter' +AND log_comment = 'preliminary merge, no filter' AND type = 'QueryFinish' ORDER BY query_start_time DESC limit 1; @@ -60,19 +62,20 @@ SETTINGS max_block_size = 8192, read_in_order_two_level_merge_threshold = 0, --force preliminary merge max_threads = 1, optimize_read_in_order = 1, -log_comment = 'with filter'; +log_comment = 'preliminary merge with filter'; SYSTEM FLUSH LOGS; SELECT read_rows FROM system.query_log WHERE current_database = currentDatabase() -AND log_comment = 'with filter' +AND log_comment = 'preliminary merge with filter' AND type = 'QueryFinish' ORDER BY query_start_time DESC LIMIT 1; --- Should not impact cases without preliminary merge (might read again when chunk row is less than limit) +-- Expecting 2 virtual rows + one chunk (8192) for result + one extra chunk for next consumption in merge transform (8192), +-- both chunks come from the same part. SELECT x FROM t ORDER BY x ASC @@ -81,14 +84,37 @@ SETTINGS max_block_size = 8192, read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge max_threads = 1, optimize_read_in_order = 1, -log_comment = 'no impact'; +log_comment = 'no preliminary merge, no filter'; SYSTEM FLUSH LOGS; SELECT read_rows FROM system.query_log WHERE current_database = currentDatabase() -AND log_comment = 'no impact' +AND log_comment = 'no preliminary merge, no filter' +AND type = 'QueryFinish' +ORDER BY query_start_time DESC +LIMIT 1; + +-- Expecting 2 virtual rows + two chunks (8192*2) get filtered out + one chunk for result (8192), +-- all chunks come from the same part. +SELECT k +FROM t +WHERE k > 8192 * 2 +ORDER BY x ASC +LIMIT 4 +SETTINGS max_block_size = 8192, +read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge +max_threads = 1, +optimize_read_in_order = 1, +log_comment = 'no preliminary merge, with filter'; + +SYSTEM FLUSH LOGS; + +SELECT read_rows +FROM system.query_log +WHERE current_database = currentDatabase() +AND log_comment = 'no preliminary merge, with filter' AND type = 'QueryFinish' ORDER BY query_start_time DESC LIMIT 1; From 3f6cdeb8802c04f39d01e4b048fe0381ff200242 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Wed, 15 May 2024 18:26:28 +0000 Subject: [PATCH 013/680] add more check --- .../Algorithms/MergingSortedAlgorithm.cpp | 8 ++ src/Processors/QueryPlan/SortingStep.cpp | 73 ++++++++++++------- src/Processors/QueryPlan/SortingStep.h | 2 + .../MergeTree/MergeTreeSelectProcessor.cpp | 2 +- .../MergeTree/MergeTreeSelectProcessor.h | 2 + .../02346_fulltext_index_search.sql | 8 +- ...r_optimization_with_virtual_row.reference} | 10 +++ ...n_order_optimization_with_virtual_row.sql} | 20 ++++- 8 files changed, 92 insertions(+), 33 deletions(-) rename tests/queries/0_stateless/{03031_read_in_order_optimization.reference => 03031_read_in_order_optimization_with_virtual_row.reference} (59%) rename tests/queries/0_stateless/{03031_read_in_order_optimization.sql => 03031_read_in_order_optimization_with_virtual_row.sql} (83%) diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp index 7da73349c4a..eb5805087c4 100644 --- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp @@ -8,6 +8,11 @@ namespace DB { +namespace ErrorCodes +{ + extern const int NOT_IMPLEMENTED; +} + MergingSortedAlgorithm::MergingSortedAlgorithm( Block header_, size_t num_inputs, @@ -134,6 +139,9 @@ IMergingAlgorithm::Status MergingSortedAlgorithm::mergeImpl(TSortingHeap & queue auto current = queue.current(); + if (getVirtualRowFromChunk(current_inputs[current.impl->order].chunk)) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Virtual row is not implemented for Non-batch mode."); + if (current.impl->isLast() && current_inputs[current.impl->order].skip_last_row) { /// Get the next block from the corresponding source, if there is one. diff --git a/src/Processors/QueryPlan/SortingStep.cpp b/src/Processors/QueryPlan/SortingStep.cpp index d728e8fb154..97157b06f19 100644 --- a/src/Processors/QueryPlan/SortingStep.cpp +++ b/src/Processors/QueryPlan/SortingStep.cpp @@ -241,38 +241,57 @@ void SortingStep::finishSorting( }); } +void SortingStep::enableVirtualRow(const QueryPipelineBuilder & pipeline) const +{ + /// We check every step of this pipeline, to make sure virtual row can work correctly. + /// Currently ExpressionTransform is supported, should add other processors if possible. + const auto& pipe = pipeline.getPipe(); + bool enable_virtual_row = true; + std::vector> merge_tree_sources; + for (const auto & processor : pipe.getProcessors()) + { + if (auto merge_tree_source = std::dynamic_pointer_cast(processor)) + { + merge_tree_sources.push_back(merge_tree_source); + } + else if (!std::dynamic_pointer_cast(processor)) + { + enable_virtual_row = false; + break; + } + } + + /// If everything is okay, we enable virtual row in MergeTreeSelectProcessor + if (enable_virtual_row && merge_tree_sources.size() >= 2) + { + /// We have to check further in the case of fixed prefix, for example, + /// primary key ab, query SELECT a, b FROM t WHERE a = 1 ORDER BY b, + /// merge sort would sort based on b, leading to wrong result in comparison. + auto extractNameAfterDot = [](const String & name) + { + size_t pos = name.find_last_of('.'); + return (pos != String::npos) ? name.substr(pos + 1) : name; + }; + + const ColumnWithTypeAndName & type_and_name = pipeline.getHeader().getByPosition(0); + String column_name = extractNameAfterDot(type_and_name.name); + for (const auto & merge_tree_source : merge_tree_sources) + { + const auto& merge_tree_select_processor = merge_tree_source->getProcessor(); + + const auto & primary_key = merge_tree_select_processor->getPrimaryKey(); + if (primary_key.column_names[0] == column_name && primary_key.data_types[0] == type_and_name.type) + merge_tree_select_processor->enableVirtualRow(); + } + } +} + void SortingStep::mergingSorted(QueryPipelineBuilder & pipeline, const SortDescription & result_sort_desc, const UInt64 limit_) { /// If there are several streams, then we merge them into one if (pipeline.getNumStreams() > 1) { - /// We check every step of this pipeline, to make sure virtual row can work correctly. - /// Currently ExpressionTransform is supported, should add other processors if possible. - const auto& pipe = pipeline.getPipe(); - bool enable_virtual_row = true; - std::vector> merge_tree_sources; - for (const auto & processor : pipe.getProcessors()) - { - if (auto merge_tree_source = std::dynamic_pointer_cast(processor)) - { - merge_tree_sources.push_back(merge_tree_source); - } - else if (!std::dynamic_pointer_cast(processor)) - { - enable_virtual_row = false; - break; - } - } - - /// If everything is okay, we enable virtual row in MergeTreeSelectProcessor - if (enable_virtual_row && merge_tree_sources.size() >= 2) - { - for (const auto & merge_tree_source : merge_tree_sources) - { - const auto& merge_tree_select_processor = merge_tree_source->getProcessor(); - merge_tree_select_processor->enableVirtualRow(); - } - } + enableVirtualRow(pipeline); auto transform = std::make_shared( pipeline.getHeader(), diff --git a/src/Processors/QueryPlan/SortingStep.h b/src/Processors/QueryPlan/SortingStep.h index 52f48f66a32..5f3820c346b 100644 --- a/src/Processors/QueryPlan/SortingStep.h +++ b/src/Processors/QueryPlan/SortingStep.h @@ -116,6 +116,8 @@ private: UInt64 limit_, bool skip_partial_sort = false); + void enableVirtualRow(const QueryPipelineBuilder & pipeline) const; + Type type; SortDescription prefix_description; diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index 0f4b68ddde9..67b58b53a0d 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -139,7 +139,7 @@ ChunkAndProgress MergeTreeSelectProcessor::read() /// Turn on virtual row just once. enable_virtual_row = false; - const auto & primary_key = storage_snapshot->metadata->primary_key; + const auto & primary_key = getPrimaryKey(); MergeTreeReadTask::BlockAndProgress res; res.row_count = 1; diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index 57da1039ba9..14481be24d3 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -73,6 +73,8 @@ public: void enableVirtualRow() { enable_virtual_row = true; } + const KeyDescription & getPrimaryKey() const { return storage_snapshot->metadata->primary_key; } + private: /// Sets up range readers corresponding to data readers void initializeRangeReaders(); diff --git a/tests/queries/0_stateless/02346_fulltext_index_search.sql b/tests/queries/0_stateless/02346_fulltext_index_search.sql index 3c172bfdaf7..fb6da10a115 100644 --- a/tests/queries/0_stateless/02346_fulltext_index_search.sql +++ b/tests/queries/0_stateless/02346_fulltext_index_search.sql @@ -195,14 +195,14 @@ INSERT INTO tab VALUES (201, 'rick c01'), (202, 'mick c02'), (203, 'nick c03'); SELECT name, type FROM system.data_skipping_indices WHERE table == 'tab' AND database = currentDatabase() LIMIT 1; -- search full_text index -SELECT * FROM tab WHERE s LIKE '%01%' ORDER BY k; +SELECT * FROM tab WHERE s LIKE '%01%' ORDER BY k SETTINGS optimize_read_in_order = 1; --- check the query only read 3 granules (6 rows total; each granule has 2 rows) +-- check the query only read 3 granules (6 rows total; each granule has 2 rows; there are 2 extra virtual rows) SYSTEM FLUSH LOGS; -SELECT read_rows==6 from system.query_log +SELECT read_rows==8 from system.query_log WHERE query_kind ='Select' AND current_database = currentDatabase() - AND endsWith(trimRight(query), 'SELECT * FROM tab WHERE s LIKE \'%01%\' ORDER BY k;') + AND endsWith(trimRight(query), 'SELECT * FROM tab WHERE s LIKE \'%01%\' ORDER BY k SETTINGS optimize_read_in_order = 1;') AND type='QueryFinish' AND result_rows==3 LIMIT 1; diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.reference b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference similarity index 59% rename from tests/queries/0_stateless/03031_read_in_order_optimization.reference rename to tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference index c7cce7e60e9..12c4056ac27 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization.reference +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference @@ -3,18 +3,28 @@ 2 3 16386 +======== 16385 16386 16387 16388 24578 +======== 0 1 2 3 16386 +======== 16385 16386 16387 16388 24578 +======== +1 2 +1 2 +1 3 +1 3 +1 4 +1 4 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql similarity index 83% rename from tests/queries/0_stateless/03031_read_in_order_optimization.sql rename to tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index 332ee7f58dc..ddcc1498af9 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -51,6 +51,7 @@ AND type = 'QueryFinish' ORDER BY query_start_time DESC limit 1; +SELECT '========'; -- Expecting 2 virtual rows + two chunks (8192*2) get filtered out + one chunk for result (8192), -- all chunks come from the same part. SELECT k @@ -74,6 +75,7 @@ AND type = 'QueryFinish' ORDER BY query_start_time DESC LIMIT 1; +SELECT '========'; -- Expecting 2 virtual rows + one chunk (8192) for result + one extra chunk for next consumption in merge transform (8192), -- both chunks come from the same part. SELECT x @@ -96,6 +98,7 @@ AND type = 'QueryFinish' ORDER BY query_start_time DESC LIMIT 1; +SELECT '========'; -- Expecting 2 virtual rows + two chunks (8192*2) get filtered out + one chunk for result (8192), -- all chunks come from the same part. SELECT k @@ -119,4 +122,19 @@ AND type = 'QueryFinish' ORDER BY query_start_time DESC LIMIT 1; -DROP TABLE t; \ No newline at end of file +DROP TABLE t; + +SELECT '========'; +-- from 02149_read_in_order_fixed_prefix +DROP TABLE IF EXISTS t_read_in_order; + +CREATE TABLE t_read_in_order(a UInt32, b UInt32) +ENGINE = MergeTree ORDER BY (a, b) +SETTINGS index_granularity = 3; + +SYSTEM STOP MERGES t_read_in_order; + +INSERT INTO t_read_in_order VALUES (0, 100), (1, 2), (1, 3), (1, 4), (2, 5); +INSERT INTO t_read_in_order VALUES (0, 100), (1, 2), (1, 3), (1, 4), (2, 5); + +SELECT a, b FROM t_read_in_order WHERE a = 1 ORDER BY b SETTINGS max_threads = 1; From 4a0a4c68b2e66c0d4abfef417376d3225965b459 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Sat, 18 May 2024 03:33:42 +0000 Subject: [PATCH 014/680] restrict the case of func pk --- src/Processors/QueryPlan/SortingStep.cpp | 24 +++++-- ...er_optimization_with_virtual_row.reference | 4 ++ ...in_order_optimization_with_virtual_row.sql | 64 +++++++++++++------ 3 files changed, 66 insertions(+), 26 deletions(-) diff --git a/src/Processors/QueryPlan/SortingStep.cpp b/src/Processors/QueryPlan/SortingStep.cpp index 84f90fa782f..addbdd020bb 100644 --- a/src/Processors/QueryPlan/SortingStep.cpp +++ b/src/Processors/QueryPlan/SortingStep.cpp @@ -262,12 +262,9 @@ void SortingStep::enableVirtualRow(const QueryPipelineBuilder & pipeline) const } } - /// If everything is okay, we enable virtual row in MergeTreeSelectProcessor + /// If everything is okay, enable virtual row in MergeTreeSelectProcessor. if (enable_virtual_row && merge_tree_sources.size() >= 2) { - /// We have to check further in the case of fixed prefix, for example, - /// primary key ab, query SELECT a, b FROM t WHERE a = 1 ORDER BY b, - /// merge sort would sort based on b, leading to wrong result in comparison. auto extractNameAfterDot = [](const String & name) { size_t pos = name.find_last_of('.'); @@ -278,10 +275,25 @@ void SortingStep::enableVirtualRow(const QueryPipelineBuilder & pipeline) const String column_name = extractNameAfterDot(type_and_name.name); for (const auto & merge_tree_source : merge_tree_sources) { - const auto& merge_tree_select_processor = merge_tree_source->getProcessor(); + const auto & merge_tree_select_processor = merge_tree_source->getProcessor(); + /// Check pk is not func based, as we only check type and name in filling in primary key of virtual row. const auto & primary_key = merge_tree_select_processor->getPrimaryKey(); - if (primary_key.column_names[0] == column_name && primary_key.data_types[0] == type_and_name.type) + const auto & actions = primary_key.expression->getActions(); + bool is_okay = true; + for (const auto & action : actions) + { + if (action.node->type != ActionsDAG::ActionType::INPUT) + { + is_okay = false; + break; + } + } + + /// We have to check further in the case of fixed prefix, for example, + /// primary key ab, query SELECT a, b FROM t WHERE a = 1 ORDER BY b, + /// merge sort would sort based on b, leading to wrong result in comparison. + if (is_okay && primary_key.column_names[0] == column_name && primary_key.data_types[0] == type_and_name.type) merge_tree_select_processor->enableVirtualRow(); } } diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference index 12c4056ac27..b4b1554a7d4 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference @@ -28,3 +28,7 @@ 1 3 1 4 1 4 +======== +1 3 +1 2 +1 1 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index ddcc1498af9..198bf1eb307 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -13,22 +13,22 @@ ORDER BY (x, y, z) SETTINGS index_granularity = 8192, index_granularity_bytes = 10485760; -INSERT INTO t SELECT - number, - number, - number, - number -FROM numbers(8192 * 3); - -INSERT INTO t SELECT - number + (8192 * 3), - number + (8192 * 3), - number + (8192 * 3), - number -FROM numbers(8192 * 3); - SYSTEM STOP MERGES t; +INSERT INTO t SELECT + number, + number, + number, + number +FROM numbers(8192 * 3); + +INSERT INTO t SELECT + number + (8192 * 3), + number + (8192 * 3), + number + (8192 * 3), + number +FROM numbers(8192 * 3); + -- Expecting 2 virtual rows + one chunk (8192) for result + one extra chunk for next consumption in merge transform (8192), -- both chunks come from the same part. SELECT x @@ -126,15 +126,39 @@ DROP TABLE t; SELECT '========'; -- from 02149_read_in_order_fixed_prefix -DROP TABLE IF EXISTS t_read_in_order; +DROP TABLE IF EXISTS fixed_prefix; -CREATE TABLE t_read_in_order(a UInt32, b UInt32) +CREATE TABLE fixed_prefix(a UInt32, b UInt32) ENGINE = MergeTree ORDER BY (a, b) SETTINGS index_granularity = 3; -SYSTEM STOP MERGES t_read_in_order; +SYSTEM STOP MERGES fixed_prefix; -INSERT INTO t_read_in_order VALUES (0, 100), (1, 2), (1, 3), (1, 4), (2, 5); -INSERT INTO t_read_in_order VALUES (0, 100), (1, 2), (1, 3), (1, 4), (2, 5); +INSERT INTO fixed_prefix VALUES (0, 100), (1, 2), (1, 3), (1, 4), (2, 5); +INSERT INTO fixed_prefix VALUES (0, 100), (1, 2), (1, 3), (1, 4), (2, 5); -SELECT a, b FROM t_read_in_order WHERE a = 1 ORDER BY b SETTINGS max_threads = 1; +SELECT a, b FROM fixed_prefix WHERE a = 1 ORDER BY b SETTINGS max_threads = 1; + +DROP TABLE fixed_prefix; + +SELECT '========'; +-- currently don't support virtual row in this case +DROP TABLE IF EXISTS function_pk; + +CREATE TABLE function_pk +( + `A` Int64, + `B` Int64 +) +ENGINE = MergeTree ORDER BY (A, -B) +SETTINGS index_granularity = 1; + +SYSTEM STOP MERGES function_pk; + +INSERT INTO function_pk values(1,1); +INSERT INTO function_pk values(1,3); +INSERT INTO function_pk values(1,2); + +SELECT * FROM function_pk ORDER BY (A,-B) ASC limit 3 SETTINGS max_threads = 1; + +DROP TABLE function_pk; From bd05771faac142853dde6b1461d34e6d3d47e89e Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Tue, 21 May 2024 04:43:26 +0000 Subject: [PATCH 015/680] temporarily disable a test --- .../03031_read_in_order_optimization_with_virtual_row.sql | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index 198bf1eb307..aff9faf3968 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -159,6 +159,8 @@ INSERT INTO function_pk values(1,1); INSERT INTO function_pk values(1,3); INSERT INTO function_pk values(1,2); +-- TODO: handle preliminary merge for this case, temporarily disable it +SET optimize_read_in_order = 0; SELECT * FROM function_pk ORDER BY (A,-B) ASC limit 3 SETTINGS max_threads = 1; DROP TABLE function_pk; From f8b3987d5292ed1e2acfc7cab2b7bfcd80f1aee1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Tue, 25 Jun 2024 03:26:17 +0300 Subject: [PATCH 016/680] Delete attaching prefix for deduplicated parts --- .../MergeTree/ReplicatedMergeTreeSink.cpp | 9 ++- .../__init__.py | 0 .../test.py | 61 +++++++++++++++++++ 3 files changed, 69 insertions(+), 1 deletion(-) create mode 100644 tests/integration/test_deduplicated_attached_part_rename/__init__.py create mode 100644 tests/integration/test_deduplicated_attached_part_rename/test.py diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp index 4b4f4c33e7d..4190e3cce5e 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp @@ -561,8 +561,15 @@ bool ReplicatedMergeTreeSinkImpl::writeExistingPart(MergeTreeData::Mutabl String block_id = deduplicate ? fmt::format("{}_{}", part->info.partition_id, part->checksums.getTotalChecksumHex()) : ""; bool deduplicated = commitPart(zookeeper, part, block_id, replicas_num).second; + int error = 0; /// Set a special error code if the block is duplicate - int error = (deduplicate && deduplicated) ? ErrorCodes::INSERT_WAS_DEDUPLICATED : 0; + /// And remove attaching_ prefix + if (deduplicate && deduplicated) + { + error = ErrorCodes::INSERT_WAS_DEDUPLICATED; + fs::path new_relative_path = fs::path("detached") / part->getNewName(part->info); + part->renameTo(new_relative_path, false); + } PartLog::addNewPart(storage.getContext(), PartLog::PartLogEntry(part, watch.elapsed(), profile_events_scope.getSnapshot()), ExecutionStatus(error)); return deduplicated; } diff --git a/tests/integration/test_deduplicated_attached_part_rename/__init__.py b/tests/integration/test_deduplicated_attached_part_rename/__init__.py new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/integration/test_deduplicated_attached_part_rename/test.py b/tests/integration/test_deduplicated_attached_part_rename/test.py new file mode 100644 index 00000000000..362b2bad37a --- /dev/null +++ b/tests/integration/test_deduplicated_attached_part_rename/test.py @@ -0,0 +1,61 @@ +import pytest +from helpers.cluster import ClickHouseCluster + +cluster = ClickHouseCluster(__file__) +ch1 = cluster.add_instance( + "ch1", + with_zookeeper=True, + macros={"replica": "node1"}, + stay_alive=True, +) + +database_name = "dedup_attach" + +@pytest.fixture(scope="module") +def started_cluster(): + try: + cluster.start() + yield cluster + + finally: + cluster.shutdown() + + +def q(query): + return ch1.query(database=database_name, sql=query) + + +def test_deduplicated_attached_part_renamed_after_attach(started_cluster): + ch1.query(f"CREATE DATABASE {database_name}") + + q("CREATE TABLE dedup (id UInt32) ENGINE=ReplicatedMergeTree('/clickhouse/tables/dedup_attach/dedup/s1', 'r1') ORDER BY id;") + q("INSERT INTO dedup VALUES (1),(2),(3);") + + table_data_path = q("SELECT data_paths FROM system.tables WHERE database=currentDatabase() AND table='dedup'").strip("'[]\n") + + ch1.exec_in_container( + [ + "bash", + "-c", + f"cp -r {table_data_path}/all_0_0_0 {table_data_path}/detached/all_0_0_0", + ] + ) + # Part is attached as all_1_1_0 + q("ALTER TABLE dedup ATTACH PART 'all_0_0_0'") + + assert 2 == int(q(f"SELECT count() FROM system.parts WHERE database='{database_name}' AND table = 'dedup'").strip()) + + ch1.exec_in_container( + [ + "bash", + "-c", + f"cp -r {table_data_path}/all_1_1_0 {table_data_path}/detached/all_1_1_0", + ] + ) + # Part is deduplicated and not attached + q("ALTER TABLE dedup ATTACH PART 'all_1_1_0'") + + assert 2 == int(q(f"SELECT count() FROM system.parts WHERE database='{database_name}' AND table = 'dedup'").strip()) + assert 1 == int(q(f"SELECT count() FROM system.detached_parts WHERE database='{database_name}' AND table = 'dedup'").strip()) + # Check that it is not 'attaching_all_1_1_0' + assert "all_1_1_0" == q(f"SELECT name FROM system.detached_parts WHERE database='{database_name}' AND table = 'dedup'").strip() From 6601ded4a1332548ae4cfe35c7ba8f276214d153 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Wed, 10 Jul 2024 23:02:11 +0300 Subject: [PATCH 017/680] Fix black --- .../test.py | 34 +++++++++++++++---- 1 file changed, 28 insertions(+), 6 deletions(-) diff --git a/tests/integration/test_deduplicated_attached_part_rename/test.py b/tests/integration/test_deduplicated_attached_part_rename/test.py index 362b2bad37a..2b7ab0934d1 100644 --- a/tests/integration/test_deduplicated_attached_part_rename/test.py +++ b/tests/integration/test_deduplicated_attached_part_rename/test.py @@ -11,6 +11,7 @@ ch1 = cluster.add_instance( database_name = "dedup_attach" + @pytest.fixture(scope="module") def started_cluster(): try: @@ -28,10 +29,14 @@ def q(query): def test_deduplicated_attached_part_renamed_after_attach(started_cluster): ch1.query(f"CREATE DATABASE {database_name}") - q("CREATE TABLE dedup (id UInt32) ENGINE=ReplicatedMergeTree('/clickhouse/tables/dedup_attach/dedup/s1', 'r1') ORDER BY id;") + q( + "CREATE TABLE dedup (id UInt32) ENGINE=ReplicatedMergeTree('/clickhouse/tables/dedup_attach/dedup/s1', 'r1') ORDER BY id;" + ) q("INSERT INTO dedup VALUES (1),(2),(3);") - table_data_path = q("SELECT data_paths FROM system.tables WHERE database=currentDatabase() AND table='dedup'").strip("'[]\n") + table_data_path = q( + "SELECT data_paths FROM system.tables WHERE database=currentDatabase() AND table='dedup'" + ).strip("'[]\n") ch1.exec_in_container( [ @@ -43,7 +48,11 @@ def test_deduplicated_attached_part_renamed_after_attach(started_cluster): # Part is attached as all_1_1_0 q("ALTER TABLE dedup ATTACH PART 'all_0_0_0'") - assert 2 == int(q(f"SELECT count() FROM system.parts WHERE database='{database_name}' AND table = 'dedup'").strip()) + assert 2 == int( + q( + f"SELECT count() FROM system.parts WHERE database='{database_name}' AND table = 'dedup'" + ).strip() + ) ch1.exec_in_container( [ @@ -55,7 +64,20 @@ def test_deduplicated_attached_part_renamed_after_attach(started_cluster): # Part is deduplicated and not attached q("ALTER TABLE dedup ATTACH PART 'all_1_1_0'") - assert 2 == int(q(f"SELECT count() FROM system.parts WHERE database='{database_name}' AND table = 'dedup'").strip()) - assert 1 == int(q(f"SELECT count() FROM system.detached_parts WHERE database='{database_name}' AND table = 'dedup'").strip()) + assert 2 == int( + q( + f"SELECT count() FROM system.parts WHERE database='{database_name}' AND table = 'dedup'" + ).strip() + ) + assert 1 == int( + q( + f"SELECT count() FROM system.detached_parts WHERE database='{database_name}' AND table = 'dedup'" + ).strip() + ) # Check that it is not 'attaching_all_1_1_0' - assert "all_1_1_0" == q(f"SELECT name FROM system.detached_parts WHERE database='{database_name}' AND table = 'dedup'").strip() + assert ( + "all_1_1_0" + == q( + f"SELECT name FROM system.detached_parts WHERE database='{database_name}' AND table = 'dedup'" + ).strip() + ) From 351ba3ef102979714d546e7575a9f9f54325498a Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Thu, 8 Aug 2024 10:07:39 +0200 Subject: [PATCH 018/680] Revert "Revert "Use `Atomic` database by default in `clickhouse-local`"" --- programs/local/LocalServer.cpp | 21 +++++---- src/Databases/DatabaseAtomic.cpp | 24 ++++++++-- src/Databases/DatabaseAtomic.h | 3 ++ src/Databases/DatabaseLazy.cpp | 3 +- src/Databases/DatabaseLazy.h | 2 +- src/Databases/DatabaseOnDisk.cpp | 28 ++++++++--- src/Databases/DatabaseOnDisk.h | 7 ++- src/Databases/DatabaseOrdinary.cpp | 4 +- src/Databases/DatabasesOverlay.cpp | 47 +++++++++++++++++++ src/Databases/DatabasesOverlay.h | 9 ++++ src/Databases/IDatabase.h | 1 + .../MySQL/DatabaseMaterializedMySQL.cpp | 1 + src/Interpreters/StorageID.h | 1 - .../0_stateless/01191_rename_dictionary.sql | 1 + ...ickhouse_local_interactive_table.reference | 4 +- ...2141_clickhouse_local_interactive_table.sh | 4 +- .../03199_atomic_clickhouse_local.reference | 6 +++ .../03199_atomic_clickhouse_local.sh | 24 ++++++++++ 18 files changed, 161 insertions(+), 29 deletions(-) create mode 100644 tests/queries/0_stateless/03199_atomic_clickhouse_local.reference create mode 100755 tests/queries/0_stateless/03199_atomic_clickhouse_local.sh diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp index 6b0b8fc5b50..0d731ed0e14 100644 --- a/programs/local/LocalServer.cpp +++ b/programs/local/LocalServer.cpp @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -50,7 +51,6 @@ #include #include #include -#include #include #include #include @@ -216,12 +216,12 @@ static DatabasePtr createMemoryDatabaseIfNotExists(ContextPtr context, const Str return system_database; } -static DatabasePtr createClickHouseLocalDatabaseOverlay(const String & name_, ContextPtr context_) +static DatabasePtr createClickHouseLocalDatabaseOverlay(const String & name_, ContextPtr context) { - auto databaseCombiner = std::make_shared(name_, context_); - databaseCombiner->registerNextDatabase(std::make_shared(name_, "", context_)); - databaseCombiner->registerNextDatabase(std::make_shared(name_, context_)); - return databaseCombiner; + auto overlay = std::make_shared(name_, context); + overlay->registerNextDatabase(std::make_shared(name_, fs::weakly_canonical(context->getPath()), UUIDHelpers::generateV4(), context)); + overlay->registerNextDatabase(std::make_shared(name_, "", context)); + return overlay; } /// If path is specified and not empty, will try to setup server environment and load existing metadata @@ -367,7 +367,7 @@ std::string LocalServer::getInitialCreateTableQuery() else table_structure = "(" + table_structure + ")"; - return fmt::format("CREATE TABLE {} {} ENGINE = File({}, {});", + return fmt::format("CREATE TEMPORARY TABLE {} {} ENGINE = File({}, {});", table_name, table_structure, data_format, table_file); } @@ -761,7 +761,12 @@ void LocalServer::processConfig() DatabaseCatalog::instance().initializeAndLoadTemporaryDatabase(); std::string default_database = server_settings.default_database; - DatabaseCatalog::instance().attachDatabase(default_database, createClickHouseLocalDatabaseOverlay(default_database, global_context)); + { + DatabasePtr database = createClickHouseLocalDatabaseOverlay(default_database, global_context); + if (UUID uuid = database->getUUID(); uuid != UUIDHelpers::Nil) + DatabaseCatalog::instance().addUUIDMapping(uuid); + DatabaseCatalog::instance().attachDatabase(default_database, database); + } global_context->setCurrentDatabase(default_database); if (getClientConfiguration().has("path")) diff --git a/src/Databases/DatabaseAtomic.cpp b/src/Databases/DatabaseAtomic.cpp index d86e29ca915..83b82976e4f 100644 --- a/src/Databases/DatabaseAtomic.cpp +++ b/src/Databases/DatabaseAtomic.cpp @@ -53,9 +53,6 @@ DatabaseAtomic::DatabaseAtomic(String name_, String metadata_path_, UUID uuid, c , db_uuid(uuid) { assert(db_uuid != UUIDHelpers::Nil); - fs::create_directories(fs::path(getContext()->getPath()) / "metadata"); - fs::create_directories(path_to_table_symlinks); - tryCreateMetadataSymlink(); } DatabaseAtomic::DatabaseAtomic(String name_, String metadata_path_, UUID uuid, ContextPtr context_) @@ -63,6 +60,16 @@ DatabaseAtomic::DatabaseAtomic(String name_, String metadata_path_, UUID uuid, C { } +void DatabaseAtomic::createDirectories() +{ + if (database_atomic_directories_created.test_and_set()) + return; + DatabaseOnDisk::createDirectories(); + fs::create_directories(fs::path(getContext()->getPath()) / "metadata"); + fs::create_directories(path_to_table_symlinks); + tryCreateMetadataSymlink(); +} + String DatabaseAtomic::getTableDataPath(const String & table_name) const { std::lock_guard lock(mutex); @@ -99,6 +106,7 @@ void DatabaseAtomic::drop(ContextPtr) void DatabaseAtomic::attachTable(ContextPtr /* context_ */, const String & name, const StoragePtr & table, const String & relative_table_path) { assert(relative_table_path != data_path && !relative_table_path.empty()); + createDirectories(); DetachedTables not_in_use; std::lock_guard lock(mutex); not_in_use = cleanupDetachedTables(); @@ -200,11 +208,15 @@ void DatabaseAtomic::renameTable(ContextPtr local_context, const String & table_ if (exchange && !supportsAtomicRename()) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "RENAME EXCHANGE is not supported"); + createDirectories(); waitDatabaseStarted(); auto & other_db = dynamic_cast(to_database); bool inside_database = this == &other_db; + if (!inside_database) + other_db.createDirectories(); + String old_metadata_path = getObjectMetadataPath(table_name); String new_metadata_path = to_database.getObjectMetadataPath(to_table_name); @@ -325,6 +337,7 @@ void DatabaseAtomic::commitCreateTable(const ASTCreateQuery & query, const Stora const String & table_metadata_tmp_path, const String & table_metadata_path, ContextPtr query_context) { + createDirectories(); DetachedTables not_in_use; auto table_data_path = getTableDataPath(query); try @@ -461,6 +474,9 @@ void DatabaseAtomic::beforeLoadingMetadata(ContextMutablePtr /*context*/, Loadin if (mode < LoadingStrictnessLevel::FORCE_RESTORE) return; + if (!fs::exists(path_to_table_symlinks)) + return; + /// Recreate symlinks to table data dirs in case of force restore, because some of them may be broken for (const auto & table_path : fs::directory_iterator(path_to_table_symlinks)) { @@ -588,6 +604,7 @@ void DatabaseAtomic::renameDatabase(ContextPtr query_context, const String & new { /// CREATE, ATTACH, DROP, DETACH and RENAME DATABASE must hold DDLGuard + createDirectories(); waitDatabaseStarted(); bool check_ref_deps = query_context->getSettingsRef().check_referential_table_dependencies; @@ -679,4 +696,5 @@ void registerDatabaseAtomic(DatabaseFactory & factory) }; factory.registerDatabase("Atomic", create_fn); } + } diff --git a/src/Databases/DatabaseAtomic.h b/src/Databases/DatabaseAtomic.h index 4a4ccfa2573..ca24494f600 100644 --- a/src/Databases/DatabaseAtomic.h +++ b/src/Databases/DatabaseAtomic.h @@ -76,6 +76,9 @@ protected: using DetachedTables = std::unordered_map; [[nodiscard]] DetachedTables cleanupDetachedTables() TSA_REQUIRES(mutex); + std::atomic_flag database_atomic_directories_created = ATOMIC_FLAG_INIT; + void createDirectories(); + void tryCreateMetadataSymlink(); virtual bool allowMoveTableToOtherDatabaseEngine(IDatabase & /*to_database*/) const { return false; } diff --git a/src/Databases/DatabaseLazy.cpp b/src/Databases/DatabaseLazy.cpp index 3fb6d30fcb8..e43adfc5d37 100644 --- a/src/Databases/DatabaseLazy.cpp +++ b/src/Databases/DatabaseLazy.cpp @@ -47,12 +47,13 @@ DatabaseLazy::DatabaseLazy(const String & name_, const String & metadata_path_, : DatabaseOnDisk(name_, metadata_path_, std::filesystem::path("data") / escapeForFileName(name_) / "", "DatabaseLazy (" + name_ + ")", context_) , expiration_time(expiration_time_) { + createDirectories(); } void DatabaseLazy::loadStoredObjects(ContextMutablePtr local_context, LoadingStrictnessLevel /*mode*/) { - iterateMetadataFiles(local_context, [this, &local_context](const String & file_name) + iterateMetadataFiles([this, &local_context](const String & file_name) { const std::string table_name = unescapeForFileName(file_name.substr(0, file_name.size() - 4)); diff --git a/src/Databases/DatabaseLazy.h b/src/Databases/DatabaseLazy.h index 41cfb751141..aeac130594f 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 the cache. + * Works like DatabaseOrdinary, but stores only recently accessed tables in memory. * Can be used only with *Log engines. */ class DatabaseLazy final : public DatabaseOnDisk diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index 734f354d9a5..82a81b0b32d 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -172,7 +172,14 @@ DatabaseOnDisk::DatabaseOnDisk( , metadata_path(metadata_path_) , data_path(data_path_) { - fs::create_directories(local_context->getPath() + data_path); +} + + +void DatabaseOnDisk::createDirectories() +{ + if (directories_created.test_and_set()) + return; + fs::create_directories(std::filesystem::path(getContext()->getPath()) / data_path); fs::create_directories(metadata_path); } @@ -190,6 +197,8 @@ void DatabaseOnDisk::createTable( const StoragePtr & table, const ASTPtr & query) { + createDirectories(); + const auto & settings = local_context->getSettingsRef(); const auto & create = query->as(); assert(table_name == create.getTable()); @@ -257,7 +266,6 @@ void DatabaseOnDisk::createTable( } commitCreateTable(create, table, table_metadata_tmp_path, table_metadata_path, local_context); - removeDetachedPermanentlyFlag(local_context, table_name, table_metadata_path, false); } @@ -285,6 +293,8 @@ void DatabaseOnDisk::commitCreateTable(const ASTCreateQuery & query, const Stora { try { + createDirectories(); + /// Add a table to the map of known tables. attachTable(query_context, query.getTable(), table, getTableDataPath(query)); @@ -420,6 +430,7 @@ void DatabaseOnDisk::renameTable( throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Moving tables between databases of different engines is not supported"); } + createDirectories(); waitDatabaseStarted(); auto table_data_relative_path = getTableDataPath(table_name); @@ -568,14 +579,14 @@ void DatabaseOnDisk::drop(ContextPtr local_context) assert(TSA_SUPPRESS_WARNING_FOR_READ(tables).empty()); if (local_context->getSettingsRef().force_remove_data_recursively_on_drop) { - (void)fs::remove_all(local_context->getPath() + getDataPath()); + (void)fs::remove_all(std::filesystem::path(getContext()->getPath()) / data_path); (void)fs::remove_all(getMetadataPath()); } else { try { - (void)fs::remove(local_context->getPath() + getDataPath()); + (void)fs::remove(std::filesystem::path(getContext()->getPath()) / data_path); (void)fs::remove(getMetadataPath()); } catch (const fs::filesystem_error & e) @@ -613,15 +624,18 @@ time_t DatabaseOnDisk::getObjectMetadataModificationTime(const String & object_n } } -void DatabaseOnDisk::iterateMetadataFiles(ContextPtr local_context, const IteratingFunction & process_metadata_file) const +void DatabaseOnDisk::iterateMetadataFiles(const IteratingFunction & process_metadata_file) const { + if (!fs::exists(metadata_path)) + return; + auto process_tmp_drop_metadata_file = [&](const String & file_name) { assert(getUUID() == UUIDHelpers::Nil); static const char * tmp_drop_ext = ".sql.tmp_drop"; const std::string object_name = file_name.substr(0, file_name.size() - strlen(tmp_drop_ext)); - if (fs::exists(local_context->getPath() + getDataPath() + '/' + object_name)) + if (fs::exists(std::filesystem::path(getContext()->getPath()) / data_path / object_name)) { fs::rename(getMetadataPath() + file_name, getMetadataPath() + object_name + ".sql"); LOG_WARNING(log, "Object {} was not dropped previously and will be restored", backQuote(object_name)); @@ -638,7 +652,7 @@ void DatabaseOnDisk::iterateMetadataFiles(ContextPtr local_context, const Iterat std::vector> metadata_files; fs::directory_iterator dir_end; - for (fs::directory_iterator dir_it(getMetadataPath()); dir_it != dir_end; ++dir_it) + for (fs::directory_iterator dir_it(metadata_path); dir_it != dir_end; ++dir_it) { String file_name = dir_it->path().filename(); /// For '.svn', '.gitignore' directory and similar. diff --git a/src/Databases/DatabaseOnDisk.h b/src/Databases/DatabaseOnDisk.h index 12656068643..0c0ecf76a26 100644 --- a/src/Databases/DatabaseOnDisk.h +++ b/src/Databases/DatabaseOnDisk.h @@ -64,7 +64,7 @@ public: time_t getObjectMetadataModificationTime(const String & object_name) const override; String getDataPath() const override { return data_path; } - String getTableDataPath(const String & table_name) const override { return data_path + escapeForFileName(table_name) + "/"; } + String getTableDataPath(const String & table_name) const override { return std::filesystem::path(data_path) / escapeForFileName(table_name) / ""; } String getTableDataPath(const ASTCreateQuery & query) const override { return getTableDataPath(query.getTable()); } String getMetadataPath() const override { return metadata_path; } @@ -83,7 +83,7 @@ protected: using IteratingFunction = std::function; - void iterateMetadataFiles(ContextPtr context, const IteratingFunction & process_metadata_file) const; + void iterateMetadataFiles(const IteratingFunction & process_metadata_file) const; ASTPtr getCreateTableQueryImpl( const String & table_name, @@ -99,6 +99,9 @@ protected: virtual void removeDetachedPermanentlyFlag(ContextPtr context, const String & table_name, const String & table_metadata_path, bool attach); virtual void setDetachedTableNotInUseForce(const UUID & /*uuid*/) {} + std::atomic_flag directories_created = ATOMIC_FLAG_INIT; + void createDirectories(); + const String metadata_path; const String data_path; }; diff --git a/src/Databases/DatabaseOrdinary.cpp b/src/Databases/DatabaseOrdinary.cpp index 8808261654f..dd8a3f42ea8 100644 --- a/src/Databases/DatabaseOrdinary.cpp +++ b/src/Databases/DatabaseOrdinary.cpp @@ -55,7 +55,7 @@ static constexpr size_t METADATA_FILE_BUFFER_SIZE = 32768; static constexpr const char * const CONVERT_TO_REPLICATED_FLAG_NAME = "convert_to_replicated"; DatabaseOrdinary::DatabaseOrdinary(const String & name_, const String & metadata_path_, ContextPtr context_) - : DatabaseOrdinary(name_, metadata_path_, "data/" + escapeForFileName(name_) + "/", "DatabaseOrdinary (" + name_ + ")", context_) + : DatabaseOrdinary(name_, metadata_path_, std::filesystem::path("data") / escapeForFileName(name_) / "", "DatabaseOrdinary (" + name_ + ")", context_) { } @@ -265,7 +265,7 @@ void DatabaseOrdinary::loadTablesMetadata(ContextPtr local_context, ParsedTables } }; - iterateMetadataFiles(local_context, process_metadata); + iterateMetadataFiles(process_metadata); size_t objects_in_database = metadata.parsed_tables.size() - prev_tables_count; size_t dictionaries_in_database = metadata.total_dictionaries - prev_total_dictionaries; diff --git a/src/Databases/DatabasesOverlay.cpp b/src/Databases/DatabasesOverlay.cpp index 801356b3dd7..495733e15fd 100644 --- a/src/Databases/DatabasesOverlay.cpp +++ b/src/Databases/DatabasesOverlay.cpp @@ -14,6 +14,8 @@ namespace ErrorCodes { extern const int LOGICAL_ERROR; extern const int CANNOT_GET_CREATE_TABLE_QUERY; + extern const int BAD_ARGUMENTS; + extern const int UNKNOWN_TABLE; } DatabasesOverlay::DatabasesOverlay(const String & name_, ContextPtr context_) @@ -124,6 +126,39 @@ StoragePtr DatabasesOverlay::detachTable(ContextPtr context_, const String & tab getEngineName()); } +void DatabasesOverlay::renameTable( + ContextPtr current_context, + const String & name, + IDatabase & to_database, + const String & to_name, + bool exchange, + bool dictionary) +{ + for (auto & db : databases) + { + if (db->isTableExist(name, current_context)) + { + if (DatabasesOverlay * to_overlay_database = typeid_cast(&to_database)) + { + /// Renaming from Overlay database inside itself or into another Overlay database. + /// Just use the first database in the overlay as a destination. + if (to_overlay_database->databases.empty()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "The destination Overlay database {} does not have any members", to_database.getDatabaseName()); + + db->renameTable(current_context, name, *to_overlay_database->databases[0], to_name, exchange, dictionary); + } + else + { + /// Renaming into a different type of database. E.g. from Overlay on top of Atomic database into just Atomic database. + db->renameTable(current_context, name, to_database, to_name, exchange, dictionary); + } + + return; + } + } + throw Exception(ErrorCodes::UNKNOWN_TABLE, "Table {}.{} doesn't exist", backQuote(getDatabaseName()), backQuote(name)); +} + ASTPtr DatabasesOverlay::getCreateTableQueryImpl(const String & name, ContextPtr context_, bool throw_on_error) const { ASTPtr result = nullptr; @@ -178,6 +213,18 @@ String DatabasesOverlay::getTableDataPath(const ASTCreateQuery & query) const return result; } +UUID DatabasesOverlay::getUUID() const +{ + UUID result = UUIDHelpers::Nil; + for (const auto & db : databases) + { + result = db->getUUID(); + if (result != UUIDHelpers::Nil) + break; + } + return result; +} + UUID DatabasesOverlay::tryGetTableUUID(const String & table_name) const { UUID result = UUIDHelpers::Nil; diff --git a/src/Databases/DatabasesOverlay.h b/src/Databases/DatabasesOverlay.h index b0c7e7e4032..40c653e5cb5 100644 --- a/src/Databases/DatabasesOverlay.h +++ b/src/Databases/DatabasesOverlay.h @@ -35,12 +35,21 @@ public: StoragePtr detachTable(ContextPtr context, const String & table_name) override; + void renameTable( + ContextPtr current_context, + const String & name, + IDatabase & to_database, + const String & to_name, + bool exchange, + bool dictionary) override; + ASTPtr getCreateTableQueryImpl(const String & name, ContextPtr context, bool throw_on_error) const override; ASTPtr getCreateDatabaseQuery() const override; String getTableDataPath(const String & table_name) const override; String getTableDataPath(const ASTCreateQuery & query) const override; + UUID getUUID() const override; UUID tryGetTableUUID(const String & table_name) const override; void drop(ContextPtr context) override; diff --git a/src/Databases/IDatabase.h b/src/Databases/IDatabase.h index f94326d220e..02418abb2b0 100644 --- a/src/Databases/IDatabase.h +++ b/src/Databases/IDatabase.h @@ -416,6 +416,7 @@ public: std::lock_guard lock{mutex}; return database_name; } + /// Get UUID of database. virtual UUID getUUID() const { return UUIDHelpers::Nil; } diff --git a/src/Databases/MySQL/DatabaseMaterializedMySQL.cpp b/src/Databases/MySQL/DatabaseMaterializedMySQL.cpp index 2f5477a6b9d..8b3850c4e0c 100644 --- a/src/Databases/MySQL/DatabaseMaterializedMySQL.cpp +++ b/src/Databases/MySQL/DatabaseMaterializedMySQL.cpp @@ -46,6 +46,7 @@ DatabaseMaterializedMySQL::DatabaseMaterializedMySQL( , settings(std::move(settings_)) , materialize_thread(context_, database_name_, mysql_database_name_, std::move(pool_), std::move(client_), binlog_client_, settings.get()) { + createDirectories(); } void DatabaseMaterializedMySQL::rethrowExceptionIfNeeded() const diff --git a/src/Interpreters/StorageID.h b/src/Interpreters/StorageID.h index f9afbc7b98d..ad55d16e284 100644 --- a/src/Interpreters/StorageID.h +++ b/src/Interpreters/StorageID.h @@ -27,7 +27,6 @@ class ASTQueryWithTableAndOutput; class ASTTableIdentifier; class Context; -// TODO(ilezhankin): refactor and merge |ASTTableIdentifier| struct StorageID { String database_name; diff --git a/tests/queries/0_stateless/01191_rename_dictionary.sql b/tests/queries/0_stateless/01191_rename_dictionary.sql index c5012dabc81..be95e5a7d4b 100644 --- a/tests/queries/0_stateless/01191_rename_dictionary.sql +++ b/tests/queries/0_stateless/01191_rename_dictionary.sql @@ -27,6 +27,7 @@ RENAME DICTIONARY test_01191.t TO test_01191.dict1; -- {serverError INCORRECT_QU DROP DICTIONARY test_01191.t; -- {serverError INCORRECT_QUERY} DROP TABLE test_01191.t; +DROP DATABASE IF EXISTS dummy_db; CREATE DATABASE dummy_db ENGINE=Atomic; RENAME DICTIONARY test_01191.dict TO dummy_db.dict1; RENAME DICTIONARY dummy_db.dict1 TO test_01191.dict; diff --git a/tests/queries/0_stateless/02141_clickhouse_local_interactive_table.reference b/tests/queries/0_stateless/02141_clickhouse_local_interactive_table.reference index 0bb8966cbe4..0e74c0a083e 100644 --- a/tests/queries/0_stateless/02141_clickhouse_local_interactive_table.reference +++ b/tests/queries/0_stateless/02141_clickhouse_local_interactive_table.reference @@ -1,2 +1,2 @@ -CREATE TABLE default.`table`\n(\n `key` String\n)\nENGINE = File(\'TSVWithNamesAndTypes\', \'/dev/null\') -CREATE TABLE foo.`table`\n(\n `key` String\n)\nENGINE = File(\'TSVWithNamesAndTypes\', \'/dev/null\') +CREATE TEMPORARY TABLE `table`\n(\n `key` String\n)\nENGINE = File(TSVWithNamesAndTypes, \'/dev/null\') +CREATE TEMPORARY TABLE `table`\n(\n `key` String\n)\nENGINE = File(TSVWithNamesAndTypes, \'/dev/null\') diff --git a/tests/queries/0_stateless/02141_clickhouse_local_interactive_table.sh b/tests/queries/0_stateless/02141_clickhouse_local_interactive_table.sh index 934d87616ac..3a95e59416a 100755 --- a/tests/queries/0_stateless/02141_clickhouse_local_interactive_table.sh +++ b/tests/queries/0_stateless/02141_clickhouse_local_interactive_table.sh @@ -4,5 +4,5 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_LOCAL --file /dev/null --structure "key String" --input-format TSVWithNamesAndTypes --interactive --send_logs_level=trace <<<'show create table table' -$CLICKHOUSE_LOCAL --database foo --file /dev/null --structure "key String" --input-format TSVWithNamesAndTypes --interactive --send_logs_level=trace <<<'show create table table' +$CLICKHOUSE_LOCAL --file /dev/null --structure "key String" --input-format TSVWithNamesAndTypes --interactive --send_logs_level=trace <<<'show create temporary table table' +$CLICKHOUSE_LOCAL --database foo --file /dev/null --structure "key String" --input-format TSVWithNamesAndTypes --interactive --send_logs_level=trace <<<'show create temporary table table' diff --git a/tests/queries/0_stateless/03199_atomic_clickhouse_local.reference b/tests/queries/0_stateless/03199_atomic_clickhouse_local.reference new file mode 100644 index 00000000000..1975397394b --- /dev/null +++ b/tests/queries/0_stateless/03199_atomic_clickhouse_local.reference @@ -0,0 +1,6 @@ +123 +Hello +['Hello','world'] +Hello +Hello +['Hello','world'] diff --git a/tests/queries/0_stateless/03199_atomic_clickhouse_local.sh b/tests/queries/0_stateless/03199_atomic_clickhouse_local.sh new file mode 100755 index 00000000000..edaa83b8f95 --- /dev/null +++ b/tests/queries/0_stateless/03199_atomic_clickhouse_local.sh @@ -0,0 +1,24 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +${CLICKHOUSE_LOCAL} -n " +CREATE TABLE test (x UInt8) ORDER BY x; +INSERT INTO test VALUES (123); +SELECT * FROM test; +CREATE OR REPLACE TABLE test (s String) ORDER BY s; +INSERT INTO test VALUES ('Hello'); +SELECT * FROM test; +RENAME TABLE test TO test2; +CREATE OR REPLACE TABLE test (s Array(String)) ORDER BY s; +INSERT INTO test VALUES (['Hello', 'world']); +SELECT * FROM test; +SELECT * FROM test2; +EXCHANGE TABLES test AND test2; +SELECT * FROM test; +SELECT * FROM test2; +DROP TABLE test; +DROP TABLE test2; +" From 848285eabc5accf96084f847c86be5e583ab80a0 Mon Sep 17 00:00:00 2001 From: vdimir Date: Mon, 12 Aug 2024 15:40:03 +0000 Subject: [PATCH 019/680] Fix OrderByLimitByDuplicateEliminationPass with IGNORE NULLS --- src/Analyzer/FunctionNode.cpp | 7 +++ src/Analyzer/Passes/FuseFunctionsPass.cpp | 5 +- ...ore_nulls_query_tree_elimination.reference | 3 ++ ...22_ignore_nulls_query_tree_elimination.sql | 51 +++++++++++++++++++ 4 files changed, 63 insertions(+), 3 deletions(-) create mode 100644 tests/queries/0_stateless/03222_ignore_nulls_query_tree_elimination.reference create mode 100644 tests/queries/0_stateless/03222_ignore_nulls_query_tree_elimination.sql diff --git a/src/Analyzer/FunctionNode.cpp b/src/Analyzer/FunctionNode.cpp index e98b04fe9a9..f402309c7be 100644 --- a/src/Analyzer/FunctionNode.cpp +++ b/src/Analyzer/FunctionNode.cpp @@ -88,6 +88,7 @@ void FunctionNode::resolveAsFunction(FunctionBasePtr function_value) function_name = function_value->getName(); function = std::move(function_value); kind = FunctionKind::ORDINARY; + nulls_action = NullsAction::EMPTY; } void FunctionNode::resolveAsAggregateFunction(AggregateFunctionPtr aggregate_function_value) @@ -95,6 +96,12 @@ void FunctionNode::resolveAsAggregateFunction(AggregateFunctionPtr aggregate_fun function_name = aggregate_function_value->getName(); function = std::move(aggregate_function_value); kind = FunctionKind::AGGREGATE; + /** When the function is resolved, we do not need the nulls action anymore. + * The only thing that the nulls action does is map from one function to another. + * Thus, the nulls action is encoded in the function name and does not make sense anymore. + * Keeping the nulls action may lead to incorrect comparison of functions, e.g., count() and count() IGNORE NULLS are the same function. + */ + nulls_action = NullsAction::EMPTY; } void FunctionNode::resolveAsWindowFunction(AggregateFunctionPtr window_function_value) diff --git a/src/Analyzer/Passes/FuseFunctionsPass.cpp b/src/Analyzer/Passes/FuseFunctionsPass.cpp index 0175e304a2b..1009e7981ea 100644 --- a/src/Analyzer/Passes/FuseFunctionsPass.cpp +++ b/src/Analyzer/Passes/FuseFunctionsPass.cpp @@ -81,10 +81,9 @@ QueryTreeNodePtr createResolvedFunction(const ContextPtr & context, const String } FunctionNodePtr createResolvedAggregateFunction( - const String & name, const QueryTreeNodePtr & argument, const Array & parameters = {}, NullsAction action = NullsAction::EMPTY) + const String & name, const QueryTreeNodePtr & argument, const Array & parameters = {}) { auto function_node = std::make_shared(name); - function_node->setNullsAction(action); if (!parameters.empty()) { @@ -96,7 +95,7 @@ FunctionNodePtr createResolvedAggregateFunction( function_node->getArguments().getNodes() = { argument }; AggregateFunctionProperties properties; - auto aggregate_function = AggregateFunctionFactory::instance().get(name, action, {argument->getResultType()}, parameters, properties); + auto aggregate_function = AggregateFunctionFactory::instance().get(name, NullsAction::EMPTY, {argument->getResultType()}, parameters, properties); function_node->resolveAsAggregateFunction(std::move(aggregate_function)); return function_node; diff --git a/tests/queries/0_stateless/03222_ignore_nulls_query_tree_elimination.reference b/tests/queries/0_stateless/03222_ignore_nulls_query_tree_elimination.reference new file mode 100644 index 00000000000..1f242fa6f00 --- /dev/null +++ b/tests/queries/0_stateless/03222_ignore_nulls_query_tree_elimination.reference @@ -0,0 +1,3 @@ +3 +3 +3 diff --git a/tests/queries/0_stateless/03222_ignore_nulls_query_tree_elimination.sql b/tests/queries/0_stateless/03222_ignore_nulls_query_tree_elimination.sql new file mode 100644 index 00000000000..72f9781ed45 --- /dev/null +++ b/tests/queries/0_stateless/03222_ignore_nulls_query_tree_elimination.sql @@ -0,0 +1,51 @@ +#!/usr/bin/env -S ${HOME}/clickhouse-client --queries-file + +DROP TABLE IF EXISTS with_fill_date__fuzz_0; + +CREATE TABLE with_fill_date__fuzz_0 +( + `d` Date, + `d32` Nullable(Int32), + `d33` Int32 +) +ENGINE = Memory; + + +INSERT INTO with_fill_date__fuzz_0 VALUES (toDate('2020-03-03'), 1, 3), (toDate('2020-03-03'), NULL, 3), (toDate('2020-02-05'), 1, 1); + + +SELECT count() +FROM with_fill_date__fuzz_0 +ORDER BY + count(), + count() IGNORE NULLS, + max(d) +WITH FILL STEP toIntervalDay(10) +; + + +SELECT count() +FROM with_fill_date__fuzz_0 +ORDER BY + any(d32) RESPECT NULLS, + any_respect_nulls(d32), + max(d) +WITH FILL STEP toIntervalDay(10) +; + + +SELECT count() +FROM with_fill_date__fuzz_0 +ORDER BY + any(d32), + any(d32) IGNORE NULLS, + any(d32) RESPECT NULLS, + any_respect_nulls(d32) IGNORE NULLS, + any_respect_nulls(d32), + sum(d33), + sum(d33) IGNORE NULLS, + max(d) +WITH FILL STEP toIntervalDay(10) +; + + From 0abb330356245b27d929c750101dcfd1925cb6a4 Mon Sep 17 00:00:00 2001 From: vdimir Date: Tue, 13 Aug 2024 09:21:39 +0000 Subject: [PATCH 020/680] fix 03010_sum_to_to_count_if_nullable.reference --- .../0_stateless/03010_sum_to_to_count_if_nullable.reference | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference index 79ebc7a5c0c..db8d26ccfea 100644 --- a/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference +++ b/tests/queries/0_stateless/03010_sum_to_to_count_if_nullable.reference @@ -83,7 +83,7 @@ QUERY id: 0 FUNCTION id: 4, function_name: tuple, function_type: ordinary, result_type: Tuple(Nullable(UInt64)) ARGUMENTS LIST id: 5, nodes: 1 - FUNCTION id: 6, function_name: sum, function_type: aggregate, nulls_action : IGNORE_NULLS, result_type: Nullable(UInt64) + FUNCTION id: 6, function_name: sum, function_type: aggregate, result_type: Nullable(UInt64) ARGUMENTS LIST id: 7, nodes: 1 FUNCTION id: 8, function_name: if, function_type: ordinary, result_type: Nullable(UInt8) From 1ba1efe3a77fc5181d3c8e228c93e5f20a087c86 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Tue, 20 Aug 2024 17:01:41 +0000 Subject: [PATCH 021/680] fix --- .../Algorithms/MergeTreePartLevelInfo.h | 29 ------------------- 1 file changed, 29 deletions(-) delete mode 100644 src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h diff --git a/src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h b/src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h deleted file mode 100644 index e4f22deec8d..00000000000 --- a/src/Processors/Merges/Algorithms/MergeTreePartLevelInfo.h +++ /dev/null @@ -1,29 +0,0 @@ -#pragma once - -#include - -namespace DB -{ - -/// To carry part level if chunk is produced by a merge tree source -class MergeTreePartLevelInfo : public ChunkInfoCloneable -{ -public: - MergeTreePartLevelInfo() = delete; - explicit MergeTreePartLevelInfo(ssize_t part_level) - : origin_merge_tree_part_level(part_level) - { } - MergeTreePartLevelInfo(const MergeTreePartLevelInfo & other) = default; - - size_t origin_merge_tree_part_level = 0; -}; - -inline size_t getPartLevelFromChunk(const Chunk & chunk) -{ - const auto part_level_info = chunk.getChunkInfos().get(); - if (part_level_info) - return part_level_info->origin_merge_tree_part_level; - return 0; -} - -} From 57996cc68463d750d31ff26071176b3f8cbfa9ae Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Tue, 3 Sep 2024 02:38:48 +0000 Subject: [PATCH 022/680] temp fix --- src/Storages/MergeTree/MergeTreeSelectProcessor.cpp | 6 +++--- tests/queries/0_stateless/02346_fulltext_index_search.sql | 8 ++++---- .../03031_read_in_order_optimization_with_virtual_row.sql | 1 + 3 files changed, 8 insertions(+), 7 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index cc28884df24..4f1df44f68a 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -164,8 +164,8 @@ ChunkAndProgress MergeTreeSelectProcessor::read() } auto chunk = Chunk(ordered_columns, res.row_count); - if (add_part_level) - chunk.getChunkInfos().add(std::make_shared(task->getInfo().data_part->info.level, true)); + chunk.getChunkInfos().add(std::make_shared( + add_part_level ? task->getInfo().data_part->info.level : 0, true)); return ChunkAndProgress{ .chunk = std::move(chunk), @@ -190,7 +190,7 @@ ChunkAndProgress MergeTreeSelectProcessor::read() auto chunk = Chunk(ordered_columns, res.row_count); if (add_part_level) - chunk.getChunkInfos().add(std::make_shared(task->getInfo().data_part->info.level, true)); + chunk.getChunkInfos().add(std::make_shared(task->getInfo().data_part->info.level, false)); return ChunkAndProgress{ .chunk = std::move(chunk), diff --git a/tests/queries/0_stateless/02346_fulltext_index_search.sql b/tests/queries/0_stateless/02346_fulltext_index_search.sql index 179d98a161b..f0505f63124 100644 --- a/tests/queries/0_stateless/02346_fulltext_index_search.sql +++ b/tests/queries/0_stateless/02346_fulltext_index_search.sql @@ -195,14 +195,14 @@ INSERT INTO tab VALUES (201, 'rick c01'), (202, 'mick c02'), (203, 'nick c03'); SELECT name, type FROM system.data_skipping_indices WHERE table == 'tab' AND database = currentDatabase() LIMIT 1; -- search full_text index -SELECT * FROM tab WHERE s LIKE '%01%' ORDER BY k SETTINGS optimize_read_in_order = 1; +SELECT * FROM tab WHERE s LIKE '%01%' ORDER BY k SETTINGS optimize_read_in_order = 0; --- check the query only read 3 granules (6 rows total; each granule has 2 rows; there are 2 extra virtual rows) +-- check the query only read 3 granules (6 rows total; each granule has 2 rows) SYSTEM FLUSH LOGS; -SELECT read_rows==8 from system.query_log +SELECT read_rows==6 from system.query_log WHERE query_kind ='Select' AND current_database = currentDatabase() - AND endsWith(trimRight(query), 'SELECT * FROM tab WHERE s LIKE \'%01%\' ORDER BY k SETTINGS optimize_read_in_order = 1;') + AND endsWith(trimRight(query), 'SELECT * FROM tab WHERE s LIKE \'%01%\' ORDER BY k SETTINGS optimize_read_in_order = 0;') AND type='QueryFinish' AND result_rows==3 LIMIT 1; diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index aff9faf3968..5bae739bc51 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -108,6 +108,7 @@ ORDER BY x ASC LIMIT 4 SETTINGS max_block_size = 8192, read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge +read_in_order_use_buffering = false, --avoid buffer max_threads = 1, optimize_read_in_order = 1, log_comment = 'no preliminary merge, with filter'; From 87c7a8b4fbfbea4e9b02ae6494b5baad7dd30b42 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Wed, 4 Sep 2024 23:08:02 +0000 Subject: [PATCH 023/680] virtualrow sketch --- .../QueryPlan/ReadFromMergeTree.cpp | 3 + src/Processors/QueryPlan/SortingStep.cpp | 3 +- .../Transforms/VirtualRowTransform.cpp | 99 +++++++++++++++++++ .../Transforms/VirtualRowTransform.h | 28 ++++++ 4 files changed, 132 insertions(+), 1 deletion(-) create mode 100644 src/Processors/Transforms/VirtualRowTransform.cpp create mode 100644 src/Processors/Transforms/VirtualRowTransform.h diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index fd1f09f1df8..90e499d02f7 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -635,6 +636,8 @@ Pipe ReadFromMergeTree::readInOrder( }); } + pipe.addSimpleTransform([](const Block & header){ return std::make_shared(header); }); + return pipe; } diff --git a/src/Processors/QueryPlan/SortingStep.cpp b/src/Processors/QueryPlan/SortingStep.cpp index f1ee68d64cf..aa909bef8a9 100644 --- a/src/Processors/QueryPlan/SortingStep.cpp +++ b/src/Processors/QueryPlan/SortingStep.cpp @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include @@ -259,7 +260,7 @@ void SortingStep::enableVirtualRow(const QueryPipelineBuilder & pipeline) const { merge_tree_sources.push_back(merge_tree_source); } - else if (!std::dynamic_pointer_cast(processor)) + else if (!std::dynamic_pointer_cast(processor) && !std::dynamic_pointer_cast(processor)) { enable_virtual_row = false; break; diff --git a/src/Processors/Transforms/VirtualRowTransform.cpp b/src/Processors/Transforms/VirtualRowTransform.cpp new file mode 100644 index 00000000000..2e486616e8e --- /dev/null +++ b/src/Processors/Transforms/VirtualRowTransform.cpp @@ -0,0 +1,99 @@ +#include +#include "Processors/Chunk.h" + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + +VirtualRowTransform::VirtualRowTransform(const Block & header) + : IInflatingTransform(header, header) +{ +} + +IInflatingTransform::Status VirtualRowTransform::prepare() +{ + /// Check can output. + + if (output.isFinished()) + { + input.close(); + return Status::Finished; + } + + if (!output.canPush()) + { + input.setNotNeeded(); + return Status::PortFull; + } + + /// Output if has data. + if (generated) + { + output.push(std::move(current_chunk)); + generated = false; + return Status::PortFull; + } + + if (can_generate) + return Status::Ready; + + /// Check can input. + if (!has_input) + { + if (input.isFinished()) + { + if (is_finished) + { + output.finish(); + return Status::Finished; + } + is_finished = true; + return Status::Ready; + } + + input.setNeeded(); + + if (!input.hasData()) + return Status::NeedData; + + /// Set input port NotNeeded after chunk was pulled. + current_chunk = input.pull(true); + has_input = true; + } + + /// Now transform. + return Status::Ready; +} + +void VirtualRowTransform::consume(Chunk chunk) +{ + if (!is_first) + { + temp_chunk = std::move(chunk); + return; + } + + is_first = false; + temp_chunk = std::move(chunk); +} + +Chunk VirtualRowTransform::generate() +{ + if (temp_chunk.empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Can't generate chunk in VirtualRowTransform"); + + Chunk result; + result.swap(temp_chunk); + return result; +} + +bool VirtualRowTransform::canGenerate() +{ + return !temp_chunk.empty(); +} + +} diff --git a/src/Processors/Transforms/VirtualRowTransform.h b/src/Processors/Transforms/VirtualRowTransform.h new file mode 100644 index 00000000000..d054c798345 --- /dev/null +++ b/src/Processors/Transforms/VirtualRowTransform.h @@ -0,0 +1,28 @@ +#pragma once + +#include +#include + +namespace DB +{ + +class VirtualRowTransform : public IInflatingTransform +{ +public: + explicit VirtualRowTransform(const Block & header); + + String getName() const override { return "VirtualRowTransform"; } + + Status prepare() override; + +protected: + void consume(Chunk chunk) override; + bool canGenerate() override; + Chunk generate() override; + +private: + bool is_first = false; + Chunk temp_chunk; +}; + +} From 67ad7b592ce5152496bc8ddc5f3dce3cb7e9d571 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Fri, 6 Sep 2024 04:12:03 +0000 Subject: [PATCH 024/680] better --- .../Transforms/VirtualRowTransform.cpp | 41 ++++++++++++------- .../Transforms/VirtualRowTransform.h | 19 ++++++--- 2 files changed, 40 insertions(+), 20 deletions(-) diff --git a/src/Processors/Transforms/VirtualRowTransform.cpp b/src/Processors/Transforms/VirtualRowTransform.cpp index 2e486616e8e..e79ede2abec 100644 --- a/src/Processors/Transforms/VirtualRowTransform.cpp +++ b/src/Processors/Transforms/VirtualRowTransform.cpp @@ -10,11 +10,12 @@ namespace ErrorCodes } VirtualRowTransform::VirtualRowTransform(const Block & header) - : IInflatingTransform(header, header) + : IProcessor({header}, {header}) + , input(inputs.front()), output(outputs.front()) { } -IInflatingTransform::Status VirtualRowTransform::prepare() +VirtualRowTransform::Status VirtualRowTransform::prepare() { /// Check can output. @@ -46,13 +47,8 @@ IInflatingTransform::Status VirtualRowTransform::prepare() { if (input.isFinished()) { - if (is_finished) - { - output.finish(); - return Status::Finished; - } - is_finished = true; - return Status::Ready; + output.finish(); + return Status::Finished; } input.setNeeded(); @@ -69,6 +65,28 @@ IInflatingTransform::Status VirtualRowTransform::prepare() return Status::Ready; } +void VirtualRowTransform::work() +{ + if (can_generate) + { + if (generated) + throw Exception(ErrorCodes::LOGICAL_ERROR, "VirtualRowTransform cannot consume chunk because it already was generated"); + + current_chunk = generate(); + generated = true; + can_generate = false; + } + else + { + if (!has_input) + throw Exception(ErrorCodes::LOGICAL_ERROR, "VirtualRowTransform cannot consume chunk because it wasn't read"); + + consume(std::move(current_chunk)); + has_input = false; + can_generate = true; + } +} + void VirtualRowTransform::consume(Chunk chunk) { if (!is_first) @@ -91,9 +109,4 @@ Chunk VirtualRowTransform::generate() return result; } -bool VirtualRowTransform::canGenerate() -{ - return !temp_chunk.empty(); -} - } diff --git a/src/Processors/Transforms/VirtualRowTransform.h b/src/Processors/Transforms/VirtualRowTransform.h index d054c798345..7f6be5d792e 100644 --- a/src/Processors/Transforms/VirtualRowTransform.h +++ b/src/Processors/Transforms/VirtualRowTransform.h @@ -6,7 +6,7 @@ namespace DB { -class VirtualRowTransform : public IInflatingTransform +class VirtualRowTransform : public IProcessor { public: explicit VirtualRowTransform(const Block & header); @@ -14,13 +14,20 @@ public: String getName() const override { return "VirtualRowTransform"; } Status prepare() override; - -protected: - void consume(Chunk chunk) override; - bool canGenerate() override; - Chunk generate() override; + void work() override; private: + void consume(Chunk chunk); + Chunk generate(); + + InputPort & input; + OutputPort & output; + + Chunk current_chunk; + bool has_input = false; + bool generated = false; + bool can_generate = false; + bool is_first = false; Chunk temp_chunk; }; From 384617cfdf26539d5478120caa25f7e57a28d6b3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Fri, 6 Sep 2024 18:12:16 +0300 Subject: [PATCH 025/680] Check for unexpected relative path --- src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp index cf5537452f3..68aa370959c 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp @@ -582,6 +582,8 @@ bool ReplicatedMergeTreeSinkImpl::writeExistingPart(MergeTreeData::Mutabl if (deduplicate && deduplicated) { error = ErrorCodes::INSERT_WAS_DEDUPLICATED; + if (!startsWith(part->getDataPartStorage().getRelativePath(), "detached/attaching_")) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected relative path for a part: {}", part->getDataPartStorage().getRelativePath()); fs::path new_relative_path = fs::path("detached") / part->getNewName(part->info); part->renameTo(new_relative_path, false); } From 35e263a4205afa405a5f819fdf91e102ad0cd088 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Fri, 6 Sep 2024 18:12:44 +0300 Subject: [PATCH 026/680] Cleanup for flaky tests --- .../test_deduplicated_attached_part_rename/test.py | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tests/integration/test_deduplicated_attached_part_rename/test.py b/tests/integration/test_deduplicated_attached_part_rename/test.py index 2b7ab0934d1..7afd85c62dc 100644 --- a/tests/integration/test_deduplicated_attached_part_rename/test.py +++ b/tests/integration/test_deduplicated_attached_part_rename/test.py @@ -81,3 +81,7 @@ def test_deduplicated_attached_part_renamed_after_attach(started_cluster): f"SELECT name FROM system.detached_parts WHERE database='{database_name}' AND table = 'dedup'" ).strip() ) + + q("DROP TABLE dedup") + q("SYSTEM DROP REPLICA 'r1' FROM ZKPATH '/clickhouse/tables/dedup_attach/dedup/s1'") + ch1.query(f"DROP DATABASE {database_name}") From 8e2f98a032378588e932e929fa1a46680846f367 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Sat, 7 Sep 2024 15:47:39 +0000 Subject: [PATCH 027/680] Make a clean start with v1.21.2 --- contrib/krb5 | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/krb5 b/contrib/krb5 index 71b06c22760..878cf51ff05 160000 --- a/contrib/krb5 +++ b/contrib/krb5 @@ -1 +1 @@ -Subproject commit 71b06c2276009ae649c7703019f3b4605f66fd3d +Subproject commit 878cf51ff0516da8e50235e770f52c75e8dc11d8 From 35f27bf36db43d67121584bcf7bfc407c05ae2c8 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Sat, 7 Sep 2024 15:59:48 +0000 Subject: [PATCH 028/680] Bump krb5 to v1.21.3 --- contrib/krb5 | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/krb5 b/contrib/krb5 index 878cf51ff05..c5b4b994c18 160000 --- a/contrib/krb5 +++ b/contrib/krb5 @@ -1 +1 @@ -Subproject commit 878cf51ff0516da8e50235e770f52c75e8dc11d8 +Subproject commit c5b4b994c18db86933255907a97eee5993fd18fe From 36f62334c40610ad062a41ef3edbb8ecd535afff Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Sun, 8 Sep 2024 00:31:02 +0000 Subject: [PATCH 029/680] move logic to virtualrow transform --- .../QueryPlan/ReadFromMergeTree.cpp | 20 ++++-- .../Transforms/VirtualRowTransform.cpp | 68 +++++++++++-------- .../Transforms/VirtualRowTransform.h | 25 ++++--- .../MergeTree/MergeTreeSelectProcessor.cpp | 60 +++------------- .../MergeTree/MergeTreeSelectProcessor.h | 8 --- ...1_mergetree_read_in_order_spread.reference | 7 +- ...er_optimization_with_virtual_row.reference | 16 +---- ...in_order_optimization_with_virtual_row.sql | 58 ++-------------- 8 files changed, 90 insertions(+), 172 deletions(-) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 90e499d02f7..264d4cd095d 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -615,15 +615,25 @@ Pipe ReadFromMergeTree::readInOrder( actions_settings, block_size, reader_settings); processor->addPartLevelToChunk(isQueryWithFinal()); - processor->addVirtualRowToChunk(part_with_ranges.data_part->getIndex(), part_with_ranges.ranges.front().begin); - if (need_virtual_row) - processor->enableVirtualRow(); auto source = std::make_shared(std::move(processor), data.getLogName()); if (set_total_rows_approx) source->addTotalRowsApprox(total_rows); - pipes.emplace_back(std::move(source)); + Pipe pipe(source); + + if (need_virtual_row) + { + pipe.addSimpleTransform([&](const Block & header) + { + return std::make_shared(header, + storage_snapshot->metadata->primary_key, + part_with_ranges.data_part->getIndex(), + part_with_ranges.ranges.front().begin); + }); + } + + pipes.emplace_back(std::move(pipe)); } auto pipe = Pipe::unitePipes(std::move(pipes)); @@ -636,8 +646,6 @@ Pipe ReadFromMergeTree::readInOrder( }); } - pipe.addSimpleTransform([](const Block & header){ return std::make_shared(header); }); - return pipe; } diff --git a/src/Processors/Transforms/VirtualRowTransform.cpp b/src/Processors/Transforms/VirtualRowTransform.cpp index e79ede2abec..55b442cefb6 100644 --- a/src/Processors/Transforms/VirtualRowTransform.cpp +++ b/src/Processors/Transforms/VirtualRowTransform.cpp @@ -1,5 +1,5 @@ #include -#include "Processors/Chunk.h" +#include namespace DB { @@ -9,9 +9,14 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; } -VirtualRowTransform::VirtualRowTransform(const Block & header) - : IProcessor({header}, {header}) +VirtualRowTransform::VirtualRowTransform(const Block & header_, + const KeyDescription & primary_key_, + const IMergeTreeDataPart::Index & index_, + size_t mark_range_begin_) + : IProcessor({header_}, {header_}) , input(inputs.front()), output(outputs.front()) + , header(header_), primary_key(primary_key_) + , index(index_), mark_range_begin(mark_range_begin_) { } @@ -72,41 +77,50 @@ void VirtualRowTransform::work() if (generated) throw Exception(ErrorCodes::LOGICAL_ERROR, "VirtualRowTransform cannot consume chunk because it already was generated"); - current_chunk = generate(); generated = true; can_generate = false; + + if (!is_first) + { + if (current_chunk.empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Can't generate chunk in VirtualRowTransform"); + return; + } + + is_first = false; + + /// Reorder the columns according to result_header + Columns ordered_columns; + ordered_columns.reserve(header.columns()); + for (size_t i = 0, j = 0; i < header.columns(); ++i) + { + const ColumnWithTypeAndName & type_and_name = header.getByPosition(i); + ColumnPtr current_column = type_and_name.type->createColumn(); + // ordered_columns.push_back(current_column->cloneResized(1)); + + if (j < index->size() && type_and_name.name == primary_key.column_names[j] + && type_and_name.type == primary_key.data_types[j]) + { + auto column = current_column->cloneEmpty(); + column->insert((*(*index)[j])[mark_range_begin]); + ordered_columns.push_back(std::move(column)); + ++j; + } + else + ordered_columns.push_back(current_column->cloneResized(1)); + } + + current_chunk.setColumns(ordered_columns, 1); + current_chunk.getChunkInfos().add(std::make_shared(0, true)); } else { if (!has_input) throw Exception(ErrorCodes::LOGICAL_ERROR, "VirtualRowTransform cannot consume chunk because it wasn't read"); - consume(std::move(current_chunk)); has_input = false; can_generate = true; } } -void VirtualRowTransform::consume(Chunk chunk) -{ - if (!is_first) - { - temp_chunk = std::move(chunk); - return; - } - - is_first = false; - temp_chunk = std::move(chunk); -} - -Chunk VirtualRowTransform::generate() -{ - if (temp_chunk.empty()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Can't generate chunk in VirtualRowTransform"); - - Chunk result; - result.swap(temp_chunk); - return result; -} - } diff --git a/src/Processors/Transforms/VirtualRowTransform.h b/src/Processors/Transforms/VirtualRowTransform.h index 7f6be5d792e..b9f0cb46242 100644 --- a/src/Processors/Transforms/VirtualRowTransform.h +++ b/src/Processors/Transforms/VirtualRowTransform.h @@ -1,15 +1,20 @@ #pragma once -#include -#include +#include +#include +#include namespace DB { +/// Virtual row is useful for read-in-order optimization when multiple parts exist. class VirtualRowTransform : public IProcessor { public: - explicit VirtualRowTransform(const Block & header); + explicit VirtualRowTransform(const Block & header_, + const KeyDescription & primary_key_, + const IMergeTreeDataPart::Index & index_, + size_t mark_range_begin_); String getName() const override { return "VirtualRowTransform"; } @@ -17,19 +22,21 @@ public: void work() override; private: - void consume(Chunk chunk); - Chunk generate(); - InputPort & input; OutputPort & output; Chunk current_chunk; bool has_input = false; bool generated = false; - bool can_generate = false; + bool can_generate = true; + bool is_first = true; - bool is_first = false; - Chunk temp_chunk; + Block header; + KeyDescription primary_key; + /// PK index used in virtual row. + IMergeTreeDataPart::Index index; + /// The first range that might contain the candidate. + size_t mark_range_begin; }; } diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index 4f1df44f68a..ca368a94bd4 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -134,38 +134,22 @@ ChunkAndProgress MergeTreeSelectProcessor::read() if (!task->getMainRangeReader().isInitialized()) initializeRangeReaders(); - if (enable_virtual_row) + auto res = algorithm->readFromTask(*task, block_size_params); + + if (res.row_count) { - /// Turn on virtual row just once. - enable_virtual_row = false; - - const auto & primary_key = getPrimaryKey(); - - MergeTreeReadTask::BlockAndProgress res; - res.row_count = 1; - /// Reorder the columns according to result_header Columns ordered_columns; ordered_columns.reserve(result_header.columns()); - for (size_t i = 0, j = 0; i < result_header.columns(); ++i) + for (size_t i = 0; i < result_header.columns(); ++i) { - const ColumnWithTypeAndName & type_and_name = result_header.getByPosition(i); - ColumnPtr current_column = type_and_name.type->createColumn(); - - if (j < index->size() && type_and_name.name == primary_key.column_names[j] && type_and_name.type == primary_key.data_types[j]) - { - auto column = current_column->cloneEmpty(); - column->insert((*(*index)[j])[mark_range_begin]); - ordered_columns.push_back(std::move(column)); - ++j; - } - else - ordered_columns.push_back(current_column->cloneResized(1)); + auto name = result_header.getByPosition(i).name; + ordered_columns.push_back(res.block.getByName(name).column); } auto chunk = Chunk(ordered_columns, res.row_count); - chunk.getChunkInfos().add(std::make_shared( - add_part_level ? task->getInfo().data_part->info.level : 0, true)); + if (add_part_level) + chunk.getChunkInfos().add(std::make_shared(task->getInfo().data_part->info.level, false)); return ChunkAndProgress{ .chunk = std::move(chunk), @@ -175,33 +159,7 @@ ChunkAndProgress MergeTreeSelectProcessor::read() } else { - auto res = algorithm->readFromTask(*task, block_size_params); - - if (res.row_count) - { - /// Reorder the columns according to result_header - Columns ordered_columns; - ordered_columns.reserve(result_header.columns()); - for (size_t i = 0; i < result_header.columns(); ++i) - { - auto name = result_header.getByPosition(i).name; - ordered_columns.push_back(res.block.getByName(name).column); - } - - auto chunk = Chunk(ordered_columns, res.row_count); - if (add_part_level) - chunk.getChunkInfos().add(std::make_shared(task->getInfo().data_part->info.level, false)); - - return ChunkAndProgress{ - .chunk = std::move(chunk), - .num_read_rows = res.num_read_rows, - .num_read_bytes = res.num_read_bytes, - .is_finished = false}; - } - else - { - return {Chunk(), res.num_read_rows, res.num_read_bytes, false}; - } + return {Chunk(), res.num_read_rows, res.num_read_bytes, false}; } } diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index d790d1e266f..6dcb6ca73d2 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -60,12 +60,6 @@ public: void addPartLevelToChunk(bool add_part_level_) { add_part_level = add_part_level_; } - void addVirtualRowToChunk(const IMergeTreeDataPart::Index & index_, size_t mark_range_begin_) - { - index = index_; - mark_range_begin = mark_range_begin_; - } - void enableVirtualRow() { enable_virtual_row = true; } const KeyDescription & getPrimaryKey() const { return storage_snapshot->metadata->primary_key; } @@ -100,8 +94,6 @@ private: bool enable_virtual_row = false; /// PK index used in virtual row. IMergeTreeDataPart::Index index; - /// The first range that might contain the candidate, used in virtual row. - size_t mark_range_begin; LoggerPtr log = getLogger("MergeTreeSelectProcessor"); std::atomic is_cancelled{false}; diff --git a/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference b/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference index 443f6d3ae93..44e61566deb 100644 --- a/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference +++ b/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference @@ -12,6 +12,7 @@ ExpressionTransform × 3 MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 - ExpressionTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform × 2 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + ExpressionTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 \ No newline at end of file diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference index b4b1554a7d4..3c3a9cf532e 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference @@ -2,25 +2,13 @@ 1 2 3 -16386 +16384 ======== 16385 16386 16387 16388 -24578 -======== -0 -1 -2 -3 -16386 -======== -16385 -16386 -16387 -16388 -24578 +24576 ======== 1 2 1 2 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index 5bae739bc51..688e427d19d 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -39,14 +39,14 @@ SETTINGS max_block_size = 8192, read_in_order_two_level_merge_threshold = 0, --force preliminary merge max_threads = 1, optimize_read_in_order = 1, -log_comment = 'preliminary merge, no filter'; +log_comment = 'no filter'; SYSTEM FLUSH LOGS; SELECT read_rows FROM system.query_log WHERE current_database = currentDatabase() -AND log_comment = 'preliminary merge, no filter' +AND log_comment = 'no filter' AND type = 'QueryFinish' ORDER BY query_start_time DESC limit 1; @@ -63,68 +63,18 @@ SETTINGS max_block_size = 8192, read_in_order_two_level_merge_threshold = 0, --force preliminary merge max_threads = 1, optimize_read_in_order = 1, -log_comment = 'preliminary merge with filter'; +log_comment = 'with filter'; SYSTEM FLUSH LOGS; SELECT read_rows FROM system.query_log WHERE current_database = currentDatabase() -AND log_comment = 'preliminary merge with filter' +AND log_comment = 'with filter' AND type = 'QueryFinish' ORDER BY query_start_time DESC LIMIT 1; -SELECT '========'; --- Expecting 2 virtual rows + one chunk (8192) for result + one extra chunk for next consumption in merge transform (8192), --- both chunks come from the same part. -SELECT x -FROM t -ORDER BY x ASC -LIMIT 4 -SETTINGS max_block_size = 8192, -read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge -max_threads = 1, -optimize_read_in_order = 1, -log_comment = 'no preliminary merge, no filter'; - -SYSTEM FLUSH LOGS; - -SELECT read_rows -FROM system.query_log -WHERE current_database = currentDatabase() -AND log_comment = 'no preliminary merge, no filter' -AND type = 'QueryFinish' -ORDER BY query_start_time DESC -LIMIT 1; - -SELECT '========'; --- Expecting 2 virtual rows + two chunks (8192*2) get filtered out + one chunk for result (8192), --- all chunks come from the same part. -SELECT k -FROM t -WHERE k > 8192 * 2 -ORDER BY x ASC -LIMIT 4 -SETTINGS max_block_size = 8192, -read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge -read_in_order_use_buffering = false, --avoid buffer -max_threads = 1, -optimize_read_in_order = 1, -log_comment = 'no preliminary merge, with filter'; - -SYSTEM FLUSH LOGS; - -SELECT read_rows -FROM system.query_log -WHERE current_database = currentDatabase() -AND log_comment = 'no preliminary merge, with filter' -AND type = 'QueryFinish' -ORDER BY query_start_time DESC -LIMIT 1; - -DROP TABLE t; - SELECT '========'; -- from 02149_read_in_order_fixed_prefix DROP TABLE IF EXISTS fixed_prefix; From 503e7490d439e2f0969ef7b09cc2134af154fa1a Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Sun, 8 Sep 2024 00:55:10 +0000 Subject: [PATCH 030/680] tidy --- src/Processors/Transforms/VirtualRowTransform.cpp | 2 +- src/Storages/MergeTree/MergeTreeReadTask.cpp | 6 ------ src/Storages/MergeTree/MergeTreeReadTask.h | 3 --- tests/queries/0_stateless/02346_fulltext_index_search.sql | 4 ++-- 4 files changed, 3 insertions(+), 12 deletions(-) diff --git a/src/Processors/Transforms/VirtualRowTransform.cpp b/src/Processors/Transforms/VirtualRowTransform.cpp index 55b442cefb6..9b904fc4ae2 100644 --- a/src/Processors/Transforms/VirtualRowTransform.cpp +++ b/src/Processors/Transforms/VirtualRowTransform.cpp @@ -98,7 +98,7 @@ void VirtualRowTransform::work() ColumnPtr current_column = type_and_name.type->createColumn(); // ordered_columns.push_back(current_column->cloneResized(1)); - if (j < index->size() && type_and_name.name == primary_key.column_names[j] + if (j < index->size() && type_and_name.name == primary_key.column_names[j] && type_and_name.type == primary_key.data_types[j]) { auto column = current_column->cloneEmpty(); diff --git a/src/Storages/MergeTree/MergeTreeReadTask.cpp b/src/Storages/MergeTree/MergeTreeReadTask.cpp index 491aa26343d..177a325ea5a 100644 --- a/src/Storages/MergeTree/MergeTreeReadTask.cpp +++ b/src/Storages/MergeTree/MergeTreeReadTask.cpp @@ -161,12 +161,6 @@ MergeTreeReadTask::BlockAndProgress MergeTreeReadTask::read(const BlockSizeParam auto read_result = range_readers.main.read(rows_to_read, mark_ranges); - if (add_virtual_row) - { - /// Now we have the virtual row, which is at most once for each part. - add_virtual_row = false; - } - /// All rows were filtered. Repeat. if (read_result.num_rows == 0) read_result.columns.clear(); diff --git a/src/Storages/MergeTree/MergeTreeReadTask.h b/src/Storages/MergeTree/MergeTreeReadTask.h index a44d4e4fabd..e90a07e0b55 100644 --- a/src/Storages/MergeTree/MergeTreeReadTask.h +++ b/src/Storages/MergeTree/MergeTreeReadTask.h @@ -162,9 +162,6 @@ private: /// Used to satistfy preferred_block_size_bytes limitation MergeTreeBlockSizePredictorPtr size_predictor; - - /// If true, add once, and then set false. - bool add_virtual_row = false; }; using MergeTreeReadTaskPtr = std::unique_ptr; diff --git a/tests/queries/0_stateless/02346_fulltext_index_search.sql b/tests/queries/0_stateless/02346_fulltext_index_search.sql index f0505f63124..80f49790201 100644 --- a/tests/queries/0_stateless/02346_fulltext_index_search.sql +++ b/tests/queries/0_stateless/02346_fulltext_index_search.sql @@ -195,14 +195,14 @@ INSERT INTO tab VALUES (201, 'rick c01'), (202, 'mick c02'), (203, 'nick c03'); SELECT name, type FROM system.data_skipping_indices WHERE table == 'tab' AND database = currentDatabase() LIMIT 1; -- search full_text index -SELECT * FROM tab WHERE s LIKE '%01%' ORDER BY k SETTINGS optimize_read_in_order = 0; +SELECT * FROM tab WHERE s LIKE '%01%' ORDER BY k; -- check the query only read 3 granules (6 rows total; each granule has 2 rows) SYSTEM FLUSH LOGS; SELECT read_rows==6 from system.query_log WHERE query_kind ='Select' AND current_database = currentDatabase() - AND endsWith(trimRight(query), 'SELECT * FROM tab WHERE s LIKE \'%01%\' ORDER BY k SETTINGS optimize_read_in_order = 0;') + AND endsWith(trimRight(query), 'SELECT * FROM tab WHERE s LIKE \'%01%\' ORDER BY k;') AND type='QueryFinish' AND result_rows==3 LIMIT 1; From b232205b4407e185b3a17bc261c9fd977d0c0e11 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Sun, 8 Sep 2024 22:22:06 +0300 Subject: [PATCH 031/680] Fix unexpected part path check --- src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp index 68aa370959c..fb2bc2fada7 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp @@ -582,7 +582,7 @@ bool ReplicatedMergeTreeSinkImpl::writeExistingPart(MergeTreeData::Mutabl if (deduplicate && deduplicated) { error = ErrorCodes::INSERT_WAS_DEDUPLICATED; - if (!startsWith(part->getDataPartStorage().getRelativePath(), "detached/attaching_")) + if (!endsWith(part->getDataPartStorage().getRelativePath(), "detached/attaching_" + part->name + "/")) throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected relative path for a part: {}", part->getDataPartStorage().getRelativePath()); fs::path new_relative_path = fs::path("detached") / part->getNewName(part->info); part->renameTo(new_relative_path, false); From 26e74bc9eec77da69c727fa2946041257bc877ce Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Mon, 9 Sep 2024 14:29:41 +0000 Subject: [PATCH 032/680] move virtual row flag to class member --- src/Processors/QueryPlan/ReadFromMergeTree.cpp | 11 +++++------ src/Processors/QueryPlan/ReadFromMergeTree.h | 4 +++- src/Storages/MergeTree/MergeTreeSequentialSource.cpp | 1 - .../01551_mergetree_read_in_order_spread.reference | 10 ++++++---- 4 files changed, 14 insertions(+), 12 deletions(-) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 264d4cd095d..599a33f1777 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -513,8 +513,7 @@ Pipe ReadFromMergeTree::readInOrder( Names required_columns, PoolSettings pool_settings, ReadType read_type, - UInt64 read_limit, - bool need_virtual_row) + UInt64 read_limit) { /// For reading in order it makes sense to read only /// one range per task to reduce number of read rows. @@ -622,7 +621,7 @@ Pipe ReadFromMergeTree::readInOrder( Pipe pipe(source); - if (need_virtual_row) + if (enable_virtual_row) { pipe.addSimpleTransform([&](const Block & header) { @@ -1061,10 +1060,10 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( for (auto && item : splitted_parts_and_ranges) { - /// need_virtual_row = true means a MergingSortedTransform should occur. + /// enable_virtual_row = true means a MergingSortedTransform should occur. /// If so, adding a virtual row might speedup in the case of multiple parts. - bool need_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; - pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit, need_virtual_row)); + enable_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; + pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit)); } } diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.h b/src/Processors/QueryPlan/ReadFromMergeTree.h index 20c9cfafc7e..7a0b22d87c4 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.h +++ b/src/Processors/QueryPlan/ReadFromMergeTree.h @@ -239,7 +239,7 @@ private: Pipe read(RangesInDataParts parts_with_range, Names required_columns, ReadType read_type, size_t max_streams, size_t min_marks_for_concurrent_read, bool use_uncompressed_cache); Pipe readFromPool(RangesInDataParts parts_with_range, Names required_columns, PoolSettings pool_settings); Pipe readFromPoolParallelReplicas(RangesInDataParts parts_with_range, Names required_columns, PoolSettings pool_settings); - Pipe readInOrder(RangesInDataParts parts_with_ranges, Names required_columns, PoolSettings pool_settings, ReadType read_type, UInt64 limit, bool need_virtual_row = false); + Pipe readInOrder(RangesInDataParts parts_with_ranges, Names required_columns, PoolSettings pool_settings, ReadType read_type, UInt64 limit); Pipe spreadMarkRanges(RangesInDataParts && parts_with_ranges, size_t num_streams, AnalysisResult & result, std::optional & result_projection); @@ -269,6 +269,8 @@ private: std::optional read_task_callback; bool enable_vertical_final = false; bool enable_remove_parts_from_snapshot_optimization = true; + + bool enable_virtual_row = false; }; } diff --git a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp index edeac12a1df..e799dc0b20e 100644 --- a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp +++ b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp @@ -14,7 +14,6 @@ #include #include #include - #include #include diff --git a/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference b/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference index 44e61566deb..e83c2e906d1 100644 --- a/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference +++ b/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference @@ -12,7 +12,9 @@ ExpressionTransform × 3 MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - VirtualRowTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 - ExpressionTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 \ No newline at end of file + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + ExpressionTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 From 4a67c68d0bc6ef337a011c044ac56899265f3b0e Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Tue, 10 Sep 2024 01:31:01 +0000 Subject: [PATCH 033/680] only focus on the direct mergesort case --- .../QueryPlan/ReadFromMergeTree.cpp | 6 +- src/Processors/QueryPlan/SortingStep.cpp | 64 ---------- src/Processors/QueryPlan/SortingStep.h | 2 - src/QueryPipeline/QueryPipelineBuilder.h | 2 - .../MergeTree/MergeTreeSelectProcessor.cpp | 2 - .../MergeTree/MergeTreeSelectProcessor.h | 12 -- src/Storages/MergeTree/MergeTreeSource.h | 2 - .../02521_aggregation_by_partitions.reference | 112 +++++++++++++----- 8 files changed, 83 insertions(+), 119 deletions(-) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 599a33f1777..a5c7af01d55 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -392,7 +392,7 @@ Pipe ReadFromMergeTree::readFromPoolParallelReplicas( auto algorithm = std::make_unique(i); auto processor = std::make_unique( - pool, std::move(algorithm), storage_snapshot, prewhere_info, + pool, std::move(algorithm), prewhere_info, actions_settings, block_size_copy, reader_settings); auto source = std::make_shared(std::move(processor), data.getLogName()); @@ -491,7 +491,7 @@ Pipe ReadFromMergeTree::readFromPool( auto algorithm = std::make_unique(i); auto processor = std::make_unique( - pool, std::move(algorithm), storage_snapshot, prewhere_info, + pool, std::move(algorithm), prewhere_info, actions_settings, block_size_copy, reader_settings); auto source = std::make_shared(std::move(processor), data.getLogName()); @@ -610,7 +610,7 @@ Pipe ReadFromMergeTree::readInOrder( algorithm = std::make_unique(i); auto processor = std::make_unique( - pool, std::move(algorithm), storage_snapshot, prewhere_info, + pool, std::move(algorithm), prewhere_info, actions_settings, block_size, reader_settings); processor->addPartLevelToChunk(isQueryWithFinal()); diff --git a/src/Processors/QueryPlan/SortingStep.cpp b/src/Processors/QueryPlan/SortingStep.cpp index aa909bef8a9..48fad9f5fdb 100644 --- a/src/Processors/QueryPlan/SortingStep.cpp +++ b/src/Processors/QueryPlan/SortingStep.cpp @@ -6,7 +6,6 @@ #include #include #include -#include #include #include #include @@ -14,9 +13,6 @@ #include #include -#include -#include -#include #include @@ -247,69 +243,11 @@ void SortingStep::finishSorting( }); } -void SortingStep::enableVirtualRow(const QueryPipelineBuilder & pipeline) const -{ - /// We check every step of this pipeline, to make sure virtual row can work correctly. - /// Currently ExpressionTransform is supported, should add other processors if possible. - const auto& pipe = pipeline.getPipe(); - bool enable_virtual_row = true; - std::vector> merge_tree_sources; - for (const auto & processor : pipe.getProcessors()) - { - if (auto merge_tree_source = std::dynamic_pointer_cast(processor)) - { - merge_tree_sources.push_back(merge_tree_source); - } - else if (!std::dynamic_pointer_cast(processor) && !std::dynamic_pointer_cast(processor)) - { - enable_virtual_row = false; - break; - } - } - - /// If everything is okay, enable virtual row in MergeTreeSelectProcessor. - if (enable_virtual_row && merge_tree_sources.size() >= 2) - { - auto extractNameAfterDot = [](const String & name) - { - size_t pos = name.find_last_of('.'); - return (pos != String::npos) ? name.substr(pos + 1) : name; - }; - - const ColumnWithTypeAndName & type_and_name = pipeline.getHeader().getByPosition(0); - String column_name = extractNameAfterDot(type_and_name.name); - for (const auto & merge_tree_source : merge_tree_sources) - { - const auto & merge_tree_select_processor = merge_tree_source->getProcessor(); - - /// Check pk is not func based, as we only check type and name in filling in primary key of virtual row. - const auto & primary_key = merge_tree_select_processor->getPrimaryKey(); - const auto & actions = primary_key.expression->getActions(); - bool is_okay = true; - for (const auto & action : actions) - { - if (action.node->type != ActionsDAG::ActionType::INPUT) - { - is_okay = false; - break; - } - } - - /// We have to check further in the case of fixed prefix, for example, - /// primary key ab, query SELECT a, b FROM t WHERE a = 1 ORDER BY b, - /// merge sort would sort based on b, leading to wrong result in comparison. - if (is_okay && primary_key.column_names[0] == column_name && primary_key.data_types[0] == type_and_name.type) - merge_tree_select_processor->enableVirtualRow(); - } - } -} - void SortingStep::mergingSorted(QueryPipelineBuilder & pipeline, const SortDescription & result_sort_desc, const UInt64 limit_) { /// If there are several streams, then we merge them into one if (pipeline.getNumStreams() > 1) { - if (use_buffering && sort_settings.read_in_order_use_buffering) { pipeline.addSimpleTransform([&](const Block & header) @@ -318,8 +256,6 @@ void SortingStep::mergingSorted(QueryPipelineBuilder & pipeline, const SortDescr }); } - enableVirtualRow(pipeline); - auto transform = std::make_shared( pipeline.getHeader(), pipeline.getNumStreams(), diff --git a/src/Processors/QueryPlan/SortingStep.h b/src/Processors/QueryPlan/SortingStep.h index e6f3a07b907..b4a49394a13 100644 --- a/src/Processors/QueryPlan/SortingStep.h +++ b/src/Processors/QueryPlan/SortingStep.h @@ -118,8 +118,6 @@ private: UInt64 limit_, bool skip_partial_sort = false); - void enableVirtualRow(const QueryPipelineBuilder & pipeline) const; - Type type; SortDescription prefix_description; diff --git a/src/QueryPipeline/QueryPipelineBuilder.h b/src/QueryPipeline/QueryPipelineBuilder.h index 22df1d8ea48..a9e5b1535c0 100644 --- a/src/QueryPipeline/QueryPipelineBuilder.h +++ b/src/QueryPipeline/QueryPipelineBuilder.h @@ -197,8 +197,6 @@ public: void setQueryIdHolder(std::shared_ptr query_id_holder) { resources.query_id_holders.emplace_back(std::move(query_id_holder)); } void addContext(ContextPtr context) { resources.interpreter_context.emplace_back(std::move(context)); } - const Pipe& getPipe() const { return pipe; } - /// Convert query pipeline to pipe. static Pipe getPipe(QueryPipelineBuilder pipeline, QueryPlanResourceHolder & resources); static QueryPipeline getPipeline(QueryPipelineBuilder builder); diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index ca368a94bd4..85f545d2a51 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -26,14 +26,12 @@ namespace ErrorCodes MergeTreeSelectProcessor::MergeTreeSelectProcessor( MergeTreeReadPoolPtr pool_, MergeTreeSelectAlgorithmPtr algorithm_, - const StorageSnapshotPtr & storage_snapshot_, const PrewhereInfoPtr & prewhere_info_, const ExpressionActionsSettings & actions_settings_, const MergeTreeReadTask::BlockSizeParams & block_size_params_, const MergeTreeReaderSettings & reader_settings_) : pool(std::move(pool_)) , algorithm(std::move(algorithm_)) - , storage_snapshot(storage_snapshot_) , prewhere_info(prewhere_info_) , actions_settings(actions_settings_) , prewhere_actions(getPrewhereActions(prewhere_info, actions_settings, reader_settings_.enable_multiple_prewhere_read_steps)) diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index 6dcb6ca73d2..7a9cebbcb2e 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -36,7 +36,6 @@ public: MergeTreeSelectProcessor( MergeTreeReadPoolPtr pool_, MergeTreeSelectAlgorithmPtr algorithm_, - const StorageSnapshotPtr & storage_snapshot_, const PrewhereInfoPtr & prewhere_info_, const ExpressionActionsSettings & actions_settings_, const MergeTreeReadTask::BlockSizeParams & block_size_params_, @@ -60,17 +59,12 @@ public: void addPartLevelToChunk(bool add_part_level_) { add_part_level = add_part_level_; } - void enableVirtualRow() { enable_virtual_row = true; } - - const KeyDescription & getPrimaryKey() const { return storage_snapshot->metadata->primary_key; } - private: /// Sets up range readers corresponding to data readers void initializeRangeReaders(); const MergeTreeReadPoolPtr pool; const MergeTreeSelectAlgorithmPtr algorithm; - const StorageSnapshotPtr storage_snapshot; const PrewhereInfoPtr prewhere_info; const ExpressionActionsSettings actions_settings; @@ -89,12 +83,6 @@ private: /// Should we add part level to produced chunk. Part level is useful for next steps if query has FINAL bool add_part_level = false; - /// Should we add a virtual row as the single first chunk. - /// Virtual row is useful for read-in-order optimization when multiple parts exist. - bool enable_virtual_row = false; - /// PK index used in virtual row. - IMergeTreeDataPart::Index index; - LoggerPtr log = getLogger("MergeTreeSelectProcessor"); std::atomic is_cancelled{false}; }; diff --git a/src/Storages/MergeTree/MergeTreeSource.h b/src/Storages/MergeTree/MergeTreeSource.h index 287f2f5ac63..7506af4f9b8 100644 --- a/src/Storages/MergeTree/MergeTreeSource.h +++ b/src/Storages/MergeTree/MergeTreeSource.h @@ -19,8 +19,6 @@ public: Status prepare() override; - const MergeTreeSelectProcessorPtr& getProcessor() const { return processor; } - #if defined(OS_LINUX) int schedule() override; #endif diff --git a/tests/queries/0_stateless/02521_aggregation_by_partitions.reference b/tests/queries/0_stateless/02521_aggregation_by_partitions.reference index 87b2d5c3430..addc36421c3 100644 --- a/tests/queries/0_stateless/02521_aggregation_by_partitions.reference +++ b/tests/queries/0_stateless/02521_aggregation_by_partitions.reference @@ -160,52 +160,100 @@ ExpressionTransform × 16 (ReadFromMergeTree) MergingSortedTransform 2 → 1 ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 1000000 Skip merging: 1 Skip merging: 1 From 79e1ce1d4bd1e032b7890f27386dbf9c043e49c0 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Thu, 12 Sep 2024 23:54:16 +0000 Subject: [PATCH 034/680] fix --- src/Processors/QueryPlan/ReadFromMergeTree.h | 4 ---- 1 file changed, 4 deletions(-) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.h b/src/Processors/QueryPlan/ReadFromMergeTree.h index a09d31155dc..b43217db598 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.h +++ b/src/Processors/QueryPlan/ReadFromMergeTree.h @@ -282,12 +282,8 @@ private: std::optional read_task_callback; bool enable_vertical_final = false; bool enable_remove_parts_from_snapshot_optimization = true; -<<<<<<< LessReadInOrder - bool enable_virtual_row = false; -======= std::optional number_of_current_replica; ->>>>>>> master }; } From 084c8115fe55440d363999dd498f77c02306c467 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Fri, 13 Sep 2024 21:09:03 +0000 Subject: [PATCH 035/680] support non-preliminary merge --- .../Optimizations/optimizeReadInOrder.cpp | 2 + .../QueryPlan/ReadFromMergeTree.cpp | 8 ++- src/Processors/QueryPlan/ReadFromMergeTree.h | 2 + ...er_optimization_with_virtual_row.reference | 12 ++++ ...in_order_optimization_with_virtual_row.sql | 58 +++++++++++++++++-- 5 files changed, 75 insertions(+), 7 deletions(-) diff --git a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp index ac7fcdcf83f..c41122c26b2 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp @@ -820,6 +820,8 @@ InputOrderInfoPtr buildInputOrderInfo(SortingStep & sorting, QueryPlan::Node & n bool can_read = reading->requestReadingInOrder(order_info->used_prefix_of_sorting_key_size, order_info->direction, order_info->limit); if (!can_read) return nullptr; + + reading->enableVirtualRow(); } return order_info; diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 43b034b476a..7a297f6db3b 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -1099,9 +1099,11 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( for (auto && item : splitted_parts_and_ranges) { - /// enable_virtual_row = true means a MergingSortedTransform should occur. - /// If so, adding a virtual row might speedup in the case of multiple parts. - enable_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; + /// If not enabled before, try to enable it when conditions meet as in the following section of preliminary merge, + /// only ExpressionTransform is added between MergingSortedTransform and readFromMergeTree. + if (!enable_virtual_row) + enable_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; + pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit)); } } diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.h b/src/Processors/QueryPlan/ReadFromMergeTree.h index b43217db598..ccb56c3f31a 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.h +++ b/src/Processors/QueryPlan/ReadFromMergeTree.h @@ -210,6 +210,8 @@ public: void applyFilters(ActionDAGNodes added_filter_nodes) override; + void enableVirtualRow() { enable_virtual_row = true; } + private: int getSortDirection() const { diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference index 3c3a9cf532e..7106ddc157c 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference @@ -10,6 +10,18 @@ 16388 24576 ======== +0 +1 +2 +3 +16384 +======== +16385 +16386 +16387 +16388 +24578 +======== 1 2 1 2 1 3 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index 688e427d19d..5bae739bc51 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -39,14 +39,14 @@ SETTINGS max_block_size = 8192, read_in_order_two_level_merge_threshold = 0, --force preliminary merge max_threads = 1, optimize_read_in_order = 1, -log_comment = 'no filter'; +log_comment = 'preliminary merge, no filter'; SYSTEM FLUSH LOGS; SELECT read_rows FROM system.query_log WHERE current_database = currentDatabase() -AND log_comment = 'no filter' +AND log_comment = 'preliminary merge, no filter' AND type = 'QueryFinish' ORDER BY query_start_time DESC limit 1; @@ -63,18 +63,68 @@ SETTINGS max_block_size = 8192, read_in_order_two_level_merge_threshold = 0, --force preliminary merge max_threads = 1, optimize_read_in_order = 1, -log_comment = 'with filter'; +log_comment = 'preliminary merge with filter'; SYSTEM FLUSH LOGS; SELECT read_rows FROM system.query_log WHERE current_database = currentDatabase() -AND log_comment = 'with filter' +AND log_comment = 'preliminary merge with filter' AND type = 'QueryFinish' ORDER BY query_start_time DESC LIMIT 1; +SELECT '========'; +-- Expecting 2 virtual rows + one chunk (8192) for result + one extra chunk for next consumption in merge transform (8192), +-- both chunks come from the same part. +SELECT x +FROM t +ORDER BY x ASC +LIMIT 4 +SETTINGS max_block_size = 8192, +read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge +max_threads = 1, +optimize_read_in_order = 1, +log_comment = 'no preliminary merge, no filter'; + +SYSTEM FLUSH LOGS; + +SELECT read_rows +FROM system.query_log +WHERE current_database = currentDatabase() +AND log_comment = 'no preliminary merge, no filter' +AND type = 'QueryFinish' +ORDER BY query_start_time DESC +LIMIT 1; + +SELECT '========'; +-- Expecting 2 virtual rows + two chunks (8192*2) get filtered out + one chunk for result (8192), +-- all chunks come from the same part. +SELECT k +FROM t +WHERE k > 8192 * 2 +ORDER BY x ASC +LIMIT 4 +SETTINGS max_block_size = 8192, +read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge +read_in_order_use_buffering = false, --avoid buffer +max_threads = 1, +optimize_read_in_order = 1, +log_comment = 'no preliminary merge, with filter'; + +SYSTEM FLUSH LOGS; + +SELECT read_rows +FROM system.query_log +WHERE current_database = currentDatabase() +AND log_comment = 'no preliminary merge, with filter' +AND type = 'QueryFinish' +ORDER BY query_start_time DESC +LIMIT 1; + +DROP TABLE t; + SELECT '========'; -- from 02149_read_in_order_fixed_prefix DROP TABLE IF EXISTS fixed_prefix; From 2aba6f5b36d959c65d6daaaacb855a7ccf9b26b2 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Fri, 13 Sep 2024 21:44:03 +0000 Subject: [PATCH 036/680] avoid conflict with buffering --- .../QueryPlan/Optimizations/optimizeReadInOrder.cpp | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp index c41122c26b2..29453acca41 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp @@ -821,7 +821,10 @@ InputOrderInfoPtr buildInputOrderInfo(SortingStep & sorting, QueryPlan::Node & n if (!can_read) return nullptr; - reading->enableVirtualRow(); + bool use_buffering = (order_info->limit == 0) && sorting.getSettings().read_in_order_use_buffering; + /// Avoid conflict with buffering. + if (!use_buffering) + reading->enableVirtualRow(); } return order_info; From c8d6c177688783b25eb8f88e1f891e9839dac8a7 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Sat, 14 Sep 2024 02:22:11 +0000 Subject: [PATCH 037/680] fix --- .../Optimizations/optimizeReadInOrder.cpp | 2 +- src/Processors/QueryPlan/ReadFromMergeTree.cpp | 17 ++++++++++++----- src/Processors/QueryPlan/ReadFromMergeTree.h | 2 +- .../02149_read_in_order_fixed_prefix.reference | 18 +++++++++++++----- ...der_optimization_with_virtual_row.reference | 2 +- 5 files changed, 28 insertions(+), 13 deletions(-) diff --git a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp index 29453acca41..b302534e2f4 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp @@ -822,7 +822,7 @@ InputOrderInfoPtr buildInputOrderInfo(SortingStep & sorting, QueryPlan::Node & n return nullptr; bool use_buffering = (order_info->limit == 0) && sorting.getSettings().read_in_order_use_buffering; - /// Avoid conflict with buffering. + /// Avoid conflict with buffering. if (!use_buffering) reading->enableVirtualRow(); } diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 7a297f6db3b..ac5db8277c2 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -549,7 +549,8 @@ Pipe ReadFromMergeTree::readInOrder( Names required_columns, PoolSettings pool_settings, ReadType read_type, - UInt64 read_limit) + UInt64 read_limit, + bool enable_current_virtual_row) { /// For reading in order it makes sense to read only /// one range per task to reduce number of read rows. @@ -660,7 +661,7 @@ Pipe ReadFromMergeTree::readInOrder( Pipe pipe(source); - if (enable_virtual_row) + if (enable_current_virtual_row) { pipe.addSimpleTransform([&](const Block & header) { @@ -1097,14 +1098,20 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( splitted_parts_and_ranges.emplace_back(std::move(new_parts)); } + /// If enabled in the optimization stage, check whether there are more than one branch. + if (enable_virtual_row) + enable_virtual_row = splitted_parts_and_ranges.size() > 1 + || (splitted_parts_and_ranges.size() == 1 && splitted_parts_and_ranges[0].size() > 1); + for (auto && item : splitted_parts_and_ranges) { - /// If not enabled before, try to enable it when conditions meet as in the following section of preliminary merge, + /// If not enabled before, try to enable it when conditions meet, as in the following section of preliminary merge, /// only ExpressionTransform is added between MergingSortedTransform and readFromMergeTree. + bool enable_current_virtual_row = enable_virtual_row; if (!enable_virtual_row) - enable_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; + enable_current_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; - pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit)); + pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit, enable_current_virtual_row)); } } diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.h b/src/Processors/QueryPlan/ReadFromMergeTree.h index ccb56c3f31a..7c0bbdc8dec 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.h +++ b/src/Processors/QueryPlan/ReadFromMergeTree.h @@ -254,7 +254,7 @@ private: Pipe read(RangesInDataParts parts_with_range, Names required_columns, ReadType read_type, size_t max_streams, size_t min_marks_for_concurrent_read, bool use_uncompressed_cache); Pipe readFromPool(RangesInDataParts parts_with_range, Names required_columns, PoolSettings pool_settings); Pipe readFromPoolParallelReplicas(RangesInDataParts parts_with_range, Names required_columns, PoolSettings pool_settings); - Pipe readInOrder(RangesInDataParts parts_with_ranges, Names required_columns, PoolSettings pool_settings, ReadType read_type, UInt64 limit); + Pipe readInOrder(RangesInDataParts parts_with_ranges, Names required_columns, PoolSettings pool_settings, ReadType read_type, UInt64 limit, bool enable_current_virtual_row = false); Pipe spreadMarkRanges(RangesInDataParts && parts_with_ranges, size_t num_streams, AnalysisResult & result, std::optional & result_projection); diff --git a/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference b/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference index d608364e01b..f7966645e8a 100644 --- a/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference +++ b/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference @@ -14,7 +14,10 @@ ExpressionTransform (Expression) ExpressionTransform × 2 (ReadFromMergeTree) - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 2020-10-01 9 2020-10-01 9 2020-10-01 9 @@ -32,9 +35,11 @@ ExpressionTransform ExpressionTransform × 2 (ReadFromMergeTree) ReverseTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InReverseOrder) 0 → 1 - ReverseTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InReverseOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InReverseOrder) 0 → 1 + ReverseTransform + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InReverseOrder) 0 → 1 2020-10-01 9 2020-10-01 9 2020-10-01 9 @@ -51,7 +56,10 @@ ExpressionTransform (Expression) ExpressionTransform × 2 (ReadFromMergeTree) - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 2020-10-11 0 2020-10-11 0 2020-10-11 0 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference index 7106ddc157c..ef9f06ec21a 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference @@ -20,7 +20,7 @@ 16386 16387 16388 -24578 +24576 ======== 1 2 1 2 From 105639c0878e896b59bea098a51f4354cf831846 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Sat, 14 Sep 2024 20:41:36 +0000 Subject: [PATCH 038/680] disable pk function --- .../QueryPlan/ReadFromMergeTree.cpp | 14 ++++++++++++- ...1_mergetree_read_in_order_spread.reference | 9 +++------ ...in_order_optimization_with_virtual_row.sql | 20 +++++++++++++++---- 3 files changed, 32 insertions(+), 11 deletions(-) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index ac5db8277c2..02d10dcb46b 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -1098,6 +1098,17 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( splitted_parts_and_ranges.emplace_back(std::move(new_parts)); } + bool primary_key_type_supports_virtual_row = true; + const auto & actions = storage_snapshot->metadata->getPrimaryKey().expression->getActions(); + for (const auto & action : actions) + { + if (action.node->type != ActionsDAG::ActionType::INPUT) + { + primary_key_type_supports_virtual_row = false; + break; + } + } + /// If enabled in the optimization stage, check whether there are more than one branch. if (enable_virtual_row) enable_virtual_row = splitted_parts_and_ranges.size() > 1 @@ -1111,7 +1122,8 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( if (!enable_virtual_row) enable_current_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; - pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit, enable_current_virtual_row)); + pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit, + enable_current_virtual_row && primary_key_type_supports_virtual_row)); } } diff --git a/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference b/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference index e83c2e906d1..443f6d3ae93 100644 --- a/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference +++ b/tests/queries/0_stateless/01551_mergetree_read_in_order_spread.reference @@ -12,9 +12,6 @@ ExpressionTransform × 3 MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - ExpressionTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + ExpressionTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index 5bae739bc51..159f38903e3 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -138,7 +138,14 @@ SYSTEM STOP MERGES fixed_prefix; INSERT INTO fixed_prefix VALUES (0, 100), (1, 2), (1, 3), (1, 4), (2, 5); INSERT INTO fixed_prefix VALUES (0, 100), (1, 2), (1, 3), (1, 4), (2, 5); -SELECT a, b FROM fixed_prefix WHERE a = 1 ORDER BY b SETTINGS max_threads = 1; +SELECT a, b +FROM fixed_prefix +WHERE a = 1 +ORDER BY b +SETTINGS max_threads = 1, +read_in_order_use_buffering = false, +optimize_read_in_order = 1, +read_in_order_two_level_merge_threshold = 0; --force preliminary merge DROP TABLE fixed_prefix; @@ -160,8 +167,13 @@ INSERT INTO function_pk values(1,1); INSERT INTO function_pk values(1,3); INSERT INTO function_pk values(1,2); --- TODO: handle preliminary merge for this case, temporarily disable it -SET optimize_read_in_order = 0; -SELECT * FROM function_pk ORDER BY (A,-B) ASC limit 3 SETTINGS max_threads = 1; +SELECT * +FROM function_pk +ORDER BY (A,-B) ASC +limit 3 +SETTINGS max_threads = 1, +read_in_order_use_buffering = false, +optimize_read_in_order = 1, +read_in_order_two_level_merge_threshold = 0; --force preliminary merge DROP TABLE function_pk; From 45471d841bd906cbd7c4b4e88581c049e759d9f1 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Mon, 16 Sep 2024 17:41:38 +0000 Subject: [PATCH 039/680] remove default value of enable_current_virtual_row --- src/Processors/QueryPlan/ReadFromMergeTree.cpp | 4 ++-- src/Processors/QueryPlan/ReadFromMergeTree.h | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 02d10dcb46b..fb69bdd5aaa 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -716,7 +716,7 @@ Pipe ReadFromMergeTree::read( if (read_type == ReadType::Default && (max_streams > 1 || checkAllPartsOnRemoteFS(parts_with_range))) return readFromPool(std::move(parts_with_range), std::move(required_columns), std::move(pool_settings)); - auto pipe = readInOrder(parts_with_range, required_columns, pool_settings, read_type, /*limit=*/ 0); + auto pipe = readInOrder(parts_with_range, required_columns, pool_settings, read_type, /*limit=*/ 0, false); /// Use ConcatProcessor to concat sources together. /// It is needed to read in parts order (and so in PK order) if single thread is used. @@ -1025,7 +1025,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( /// For parallel replicas the split will be performed on the initiator side. if (is_parallel_reading_from_replicas) { - pipes.emplace_back(readInOrder(std::move(parts_with_ranges), column_names, pool_settings, read_type, input_order_info->limit)); + pipes.emplace_back(readInOrder(std::move(parts_with_ranges), column_names, pool_settings, read_type, input_order_info->limit, false)); } else { diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.h b/src/Processors/QueryPlan/ReadFromMergeTree.h index 7c0bbdc8dec..aa1b9dcfdcb 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.h +++ b/src/Processors/QueryPlan/ReadFromMergeTree.h @@ -254,7 +254,7 @@ private: Pipe read(RangesInDataParts parts_with_range, Names required_columns, ReadType read_type, size_t max_streams, size_t min_marks_for_concurrent_read, bool use_uncompressed_cache); Pipe readFromPool(RangesInDataParts parts_with_range, Names required_columns, PoolSettings pool_settings); Pipe readFromPoolParallelReplicas(RangesInDataParts parts_with_range, Names required_columns, PoolSettings pool_settings); - Pipe readInOrder(RangesInDataParts parts_with_ranges, Names required_columns, PoolSettings pool_settings, ReadType read_type, UInt64 limit, bool enable_current_virtual_row = false); + Pipe readInOrder(RangesInDataParts parts_with_ranges, Names required_columns, PoolSettings pool_settings, ReadType read_type, UInt64 limit, bool enable_current_virtual_row); Pipe spreadMarkRanges(RangesInDataParts && parts_with_ranges, size_t num_streams, AnalysisResult & result, std::optional & result_projection); From 6af5fe48ba2f7d22447056fd148665f350830fe4 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Mon, 16 Sep 2024 19:43:00 +0000 Subject: [PATCH 040/680] handle the case first prefix fixed --- .../QueryPlan/Optimizations/optimizeReadInOrder.cpp | 10 +++++++--- src/Processors/QueryPlan/ReadFromMergeTree.cpp | 2 +- src/Storages/ReadInOrderOptimizer.cpp | 2 +- src/Storages/SelectQueryInfo.h | 11 ++++++++++- ...d_in_order_optimization_with_virtual_row.reference | 6 ++++++ ...31_read_in_order_optimization_with_virtual_row.sql | 9 +++++++++ 6 files changed, 34 insertions(+), 6 deletions(-) diff --git a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp index b302534e2f4..909645098b1 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp @@ -370,6 +370,7 @@ InputOrderInfoPtr buildInputOrderInfo( int read_direction = 0; size_t next_description_column = 0; size_t next_sort_key = 0; + bool first_prefix_fixed = false; while (next_description_column < description.size() && next_sort_key < sorting_key.column_names.size()) { @@ -447,6 +448,9 @@ InputOrderInfoPtr buildInputOrderInfo( } else if (fixed_key_columns.contains(sort_column_node)) { + if (next_sort_key == 0) + first_prefix_fixed = true; + //std::cerr << "+++++++++ Found fixed key by match" << std::endl; ++next_sort_key; } @@ -481,7 +485,7 @@ InputOrderInfoPtr buildInputOrderInfo( if (read_direction == 0 || order_key_prefix_descr.empty()) return nullptr; - return std::make_shared(order_key_prefix_descr, next_sort_key, read_direction, limit); + return std::make_shared(order_key_prefix_descr, next_sort_key, read_direction, limit, first_prefix_fixed); } /// We really need three different sort descriptions here. @@ -685,7 +689,7 @@ AggregationInputOrder buildInputOrderInfo( for (const auto & key : not_matched_group_by_keys) group_by_sort_description.emplace_back(SortColumnDescription(std::string(key))); - auto input_order = std::make_shared(order_key_prefix_descr, next_sort_key, /*read_direction*/ 1, /* limit */ 0); + auto input_order = std::make_shared(order_key_prefix_descr, next_sort_key, /*read_direction*/ 1, /* limit */ 0, false); return { std::move(input_order), std::move(sort_description_for_merging), std::move(group_by_sort_description) }; } @@ -823,7 +827,7 @@ InputOrderInfoPtr buildInputOrderInfo(SortingStep & sorting, QueryPlan::Node & n bool use_buffering = (order_info->limit == 0) && sorting.getSettings().read_in_order_use_buffering; /// Avoid conflict with buffering. - if (!use_buffering) + if (!use_buffering && !order_info->first_prefix_fixed) reading->enableVirtualRow(); } diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index fb69bdd5aaa..b507172597e 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -1808,7 +1808,7 @@ bool ReadFromMergeTree::requestReadingInOrder(size_t prefix_size, int direction, if (direction != 1 && query_info.isFinal()) return false; - query_info.input_order_info = std::make_shared(SortDescription{}, prefix_size, direction, read_limit); + query_info.input_order_info = std::make_shared(SortDescription{}, prefix_size, direction, read_limit, false); reader_settings.read_in_order = true; /// In case or read-in-order, don't create too many reading streams. diff --git a/src/Storages/ReadInOrderOptimizer.cpp b/src/Storages/ReadInOrderOptimizer.cpp index 9c8c4c2fe79..ea7ea218feb 100644 --- a/src/Storages/ReadInOrderOptimizer.cpp +++ b/src/Storages/ReadInOrderOptimizer.cpp @@ -249,7 +249,7 @@ InputOrderInfoPtr ReadInOrderOptimizer::getInputOrderImpl( if (sort_description_for_merging.empty()) return {}; - return std::make_shared(std::move(sort_description_for_merging), key_pos, read_direction, limit); + return std::make_shared(std::move(sort_description_for_merging), key_pos, read_direction, limit, false); } InputOrderInfoPtr ReadInOrderOptimizer::getInputOrder( diff --git a/src/Storages/SelectQueryInfo.h b/src/Storages/SelectQueryInfo.h index 7ad6a733c6f..bf1229f7a3a 100644 --- a/src/Storages/SelectQueryInfo.h +++ b/src/Storages/SelectQueryInfo.h @@ -119,13 +119,22 @@ struct InputOrderInfo const int direction; const UInt64 limit; + /** For virtual row optimization only + * for example, when pk is (a,b), a = 1, order by b, virtual row should be + * disabled in the following case: + * 1st part (0, 100), (1, 2), (1, 3), (1, 4) + * 2nd part (0, 100), (1, 2), (1, 3), (1, 4). + */ + bool first_prefix_fixed; + InputOrderInfo( const SortDescription & sort_description_for_merging_, size_t used_prefix_of_sorting_key_size_, - int direction_, UInt64 limit_) + int direction_, UInt64 limit_, bool first_prefix_fixed_) : sort_description_for_merging(sort_description_for_merging_) , used_prefix_of_sorting_key_size(used_prefix_of_sorting_key_size_) , direction(direction_), limit(limit_) + , first_prefix_fixed(first_prefix_fixed_) { } diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference index ef9f06ec21a..08dabf3ee06 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference @@ -28,6 +28,12 @@ 1 3 1 4 1 4 +1 2 +1 2 +1 3 +1 3 +1 4 +1 4 ======== 1 3 1 2 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index 159f38903e3..b26f3a48eef 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -147,6 +147,15 @@ read_in_order_use_buffering = false, optimize_read_in_order = 1, read_in_order_two_level_merge_threshold = 0; --force preliminary merge +SELECT a, b +FROM fixed_prefix +WHERE a = 1 +ORDER BY b +SETTINGS max_threads = 1, +read_in_order_use_buffering = false, +optimize_read_in_order = 1, +read_in_order_two_level_merge_threshold = 5; --avoid preliminary merge + DROP TABLE fixed_prefix; SELECT '========'; From 81a7927b8a24d6e7686ed6bf9bd6f7452428b492 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Tue, 17 Sep 2024 15:06:21 +0000 Subject: [PATCH 041/680] handle virtual row in BufferChunksTransform --- src/Processors/QueryPlan/BufferChunksTransform.cpp | 14 ++++++++++++++ .../Optimizations/optimizeReadInOrder.cpp | 4 +--- ...read_in_order_optimization_with_virtual_row.sql | 4 ---- 3 files changed, 15 insertions(+), 7 deletions(-) diff --git a/src/Processors/QueryPlan/BufferChunksTransform.cpp b/src/Processors/QueryPlan/BufferChunksTransform.cpp index 3601a68d36e..0d9cee28619 100644 --- a/src/Processors/QueryPlan/BufferChunksTransform.cpp +++ b/src/Processors/QueryPlan/BufferChunksTransform.cpp @@ -1,4 +1,5 @@ #include +#include namespace DB { @@ -49,13 +50,26 @@ IProcessor::Status BufferChunksTransform::prepare() else if (input.hasData()) { auto chunk = pullChunk(); + bool virtual_row = getVirtualRowFromChunk(chunk); output.push(std::move(chunk)); + if (virtual_row) + { + input.setNotNeeded(); + return Status::PortFull; + } } } if (input.hasData() && (num_buffered_rows < max_rows_to_buffer || num_buffered_bytes < max_bytes_to_buffer)) { auto chunk = pullChunk(); + bool virtual_row = getVirtualRowFromChunk(chunk); + if (virtual_row) + { + output.push(std::move(chunk)); + input.setNotNeeded(); + return Status::PortFull; + } num_buffered_rows += chunk.getNumRows(); num_buffered_bytes += chunk.bytes(); chunks.push(std::move(chunk)); diff --git a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp index 909645098b1..e7468a3a3f2 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp @@ -825,9 +825,7 @@ InputOrderInfoPtr buildInputOrderInfo(SortingStep & sorting, QueryPlan::Node & n if (!can_read) return nullptr; - bool use_buffering = (order_info->limit == 0) && sorting.getSettings().read_in_order_use_buffering; - /// Avoid conflict with buffering. - if (!use_buffering && !order_info->first_prefix_fixed) + if (!order_info->first_prefix_fixed) reading->enableVirtualRow(); } diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index b26f3a48eef..7e3af6c057a 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -108,7 +108,6 @@ ORDER BY x ASC LIMIT 4 SETTINGS max_block_size = 8192, read_in_order_two_level_merge_threshold = 5, --avoid preliminary merge -read_in_order_use_buffering = false, --avoid buffer max_threads = 1, optimize_read_in_order = 1, log_comment = 'no preliminary merge, with filter'; @@ -143,7 +142,6 @@ FROM fixed_prefix WHERE a = 1 ORDER BY b SETTINGS max_threads = 1, -read_in_order_use_buffering = false, optimize_read_in_order = 1, read_in_order_two_level_merge_threshold = 0; --force preliminary merge @@ -152,7 +150,6 @@ FROM fixed_prefix WHERE a = 1 ORDER BY b SETTINGS max_threads = 1, -read_in_order_use_buffering = false, optimize_read_in_order = 1, read_in_order_two_level_merge_threshold = 5; --avoid preliminary merge @@ -181,7 +178,6 @@ FROM function_pk ORDER BY (A,-B) ASC limit 3 SETTINGS max_threads = 1, -read_in_order_use_buffering = false, optimize_read_in_order = 1, read_in_order_two_level_merge_threshold = 0; --force preliminary merge From a48bd922d9122aa18a4cf1fe196a3418f798c7a4 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Tue, 17 Sep 2024 20:27:59 +0000 Subject: [PATCH 042/680] fix limit in BufferChunksTransform with virtual row --- src/Processors/QueryPlan/BufferChunksTransform.cpp | 14 ++++++++------ src/Processors/QueryPlan/BufferChunksTransform.h | 2 +- src/Processors/QueryPlan/ReadFromMergeTree.cpp | 2 +- .../02149_read_in_order_fixed_prefix.reference | 8 +++----- 4 files changed, 13 insertions(+), 13 deletions(-) diff --git a/src/Processors/QueryPlan/BufferChunksTransform.cpp b/src/Processors/QueryPlan/BufferChunksTransform.cpp index 0d9cee28619..47e2c2ba0d5 100644 --- a/src/Processors/QueryPlan/BufferChunksTransform.cpp +++ b/src/Processors/QueryPlan/BufferChunksTransform.cpp @@ -49,8 +49,8 @@ IProcessor::Status BufferChunksTransform::prepare() } else if (input.hasData()) { - auto chunk = pullChunk(); - bool virtual_row = getVirtualRowFromChunk(chunk); + bool virtual_row; + auto chunk = pullChunk(virtual_row); output.push(std::move(chunk)); if (virtual_row) { @@ -62,8 +62,8 @@ IProcessor::Status BufferChunksTransform::prepare() if (input.hasData() && (num_buffered_rows < max_rows_to_buffer || num_buffered_bytes < max_bytes_to_buffer)) { - auto chunk = pullChunk(); - bool virtual_row = getVirtualRowFromChunk(chunk); + bool virtual_row; + auto chunk = pullChunk(virtual_row); if (virtual_row) { output.push(std::move(chunk)); @@ -85,10 +85,12 @@ IProcessor::Status BufferChunksTransform::prepare() return Status::NeedData; } -Chunk BufferChunksTransform::pullChunk() +Chunk BufferChunksTransform::pullChunk(bool & virtual_row) { auto chunk = input.pull(); - num_processed_rows += chunk.getNumRows(); + virtual_row = getVirtualRowFromChunk(chunk); + if (!virtual_row) + num_processed_rows += chunk.getNumRows(); if (limit && num_processed_rows >= limit) input.close(); diff --git a/src/Processors/QueryPlan/BufferChunksTransform.h b/src/Processors/QueryPlan/BufferChunksTransform.h index 752f9910734..fce79eeaef3 100644 --- a/src/Processors/QueryPlan/BufferChunksTransform.h +++ b/src/Processors/QueryPlan/BufferChunksTransform.h @@ -24,7 +24,7 @@ public: String getName() const override { return "BufferChunks"; } private: - Chunk pullChunk(); + Chunk pullChunk(bool & virtual_row); InputPort & input; OutputPort & output; diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index b507172597e..45dcb4616b1 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -661,7 +661,7 @@ Pipe ReadFromMergeTree::readInOrder( Pipe pipe(source); - if (enable_current_virtual_row) + if (enable_current_virtual_row && (read_type == ReadType::InOrder)) { pipe.addSimpleTransform([&](const Block & header) { diff --git a/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference b/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference index f7966645e8a..31462988c2d 100644 --- a/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference +++ b/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference @@ -35,11 +35,9 @@ ExpressionTransform ExpressionTransform × 2 (ReadFromMergeTree) ReverseTransform - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InReverseOrder) 0 → 1 - ReverseTransform - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InReverseOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InReverseOrder) 0 → 1 + ReverseTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InReverseOrder) 0 → 1 2020-10-01 9 2020-10-01 9 2020-10-01 9 From dd6503bb2ba0cb8bbedcc807df7ebe77fc0310c5 Mon Sep 17 00:00:00 2001 From: avogar Date: Wed, 18 Sep 2024 14:10:03 +0000 Subject: [PATCH 043/680] Don't allow Variant/Dynamic types in ORDER BY/GROUP BY/PARTITION BY/PRIMARY KEY by default --- docs/en/operations/settings/settings.md | 22 +++ docs/en/sql-reference/data-types/dynamic.md | 3 + docs/en/sql-reference/data-types/variant.md | 2 + src/Analyzer/Resolve/QueryAnalyzer.cpp | 52 ++++- src/Analyzer/Resolve/QueryAnalyzer.h | 4 + src/Core/Settings.h | 3 + src/Interpreters/ExpressionAnalyzer.cpp | 42 ++++ src/Interpreters/ExpressionAnalyzer.h | 2 + src/Storages/KeyDescription.cpp | 9 + ...mic_variant_in_order_by_group_by.reference | 184 ++++++++++++++++++ ...1_dynamic_variant_in_order_by_group_by.sql | 154 +++++++++++++++ 11 files changed, 472 insertions(+), 5 deletions(-) create mode 100644 tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference create mode 100644 tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql diff --git a/docs/en/operations/settings/settings.md b/docs/en/operations/settings/settings.md index b177ded3e32..302bc8da78f 100644 --- a/docs/en/operations/settings/settings.md +++ b/docs/en/operations/settings/settings.md @@ -5682,3 +5682,25 @@ Default value: `0`. Enable `IF NOT EXISTS` for `CREATE` statement by default. If either this setting or `IF NOT EXISTS` is specified and a table with the provided name already exists, no exception will be thrown. Default value: `false`. + +## allow_suspicious_types_in_group_by {#allow_suspicious_types_in_group_by} + +Allows or restricts using [Variant](../../sql-reference/data-types/variant.md) and [Dynamic](../../sql-reference/data-types/dynamic.md) types in GROUP BY keys. + +Possible values: + +- 1 — Usage of `Variant` and `Dynamic` types is not restricted. +- 0 — Usage of `Variant` and `Dynamic` types is restricted. + +Default value: 0. + +## allow_suspicious_types_in_group_by {#allow_suspicious_types_in_group_by} + +Allows or restricts using [Variant](../../sql-reference/data-types/variant.md) and [Dynamic](../../sql-reference/data-types/dynamic.md) types in GROUP BY keys. + +Possible values: + +- 1 — Usage of `Variant` and `Dynamic` types is not restricted. +- 0 — Usage of `Variant` and `Dynamic` types is restricted. + +Default value: 0. diff --git a/docs/en/sql-reference/data-types/dynamic.md b/docs/en/sql-reference/data-types/dynamic.md index f9befd166fe..4d0bf073535 100644 --- a/docs/en/sql-reference/data-types/dynamic.md +++ b/docs/en/sql-reference/data-types/dynamic.md @@ -411,6 +411,9 @@ SELECT d, dynamicType(d) FROM test ORDER by d; └─────┴────────────────┘ ``` +**Note** by default `Dynamic` type is not allowed in `GROUP BY`/`ORDER BY` keys, if you want to use it consider its special comparison rule and enable `allow_suspicious_types_in_group_by`/`allow_suspicious_types_in_order_by` settings. + + ## Reaching the limit in number of different data types stored inside Dynamic `Dynamic` data type can store only limited number of different data types as separate subcolumns. By default, this limit is 32, but you can change it in type declaration using syntax `Dynamic(max_types=N)` where N is between 0 and 254 (due to implementation details, it's impossible to have more than 254 different data types that can be stored as separate subcolumns inside Dynamic). diff --git a/docs/en/sql-reference/data-types/variant.md b/docs/en/sql-reference/data-types/variant.md index 3c2b6e0a362..7cb0f4ad4ea 100644 --- a/docs/en/sql-reference/data-types/variant.md +++ b/docs/en/sql-reference/data-types/variant.md @@ -441,6 +441,8 @@ SELECT v, variantType(v) FROM test ORDER by v; └─────┴────────────────┘ ``` +**Note** by default `Variant` type is not allowed in `GROUP BY`/`ORDER BY` keys, if you want to use it consider its special comparison rule and enable `allow_suspicious_types_in_group_by`/`allow_suspicious_types_in_order_by` settings. + ## JSONExtract functions with Variant All `JSONExtract*` functions support `Variant` type: diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index a18c2901a58..304338109c1 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -3962,6 +3962,8 @@ ProjectionNames QueryAnalyzer::resolveSortNodeList(QueryTreeNodePtr & sort_node_ sort_node.getExpression() = sort_column_list_node->getNodes().front(); } + validateSortingKeyType(sort_node.getExpression()->getResultType(), scope); + size_t sort_expression_projection_names_size = sort_expression_projection_names.size(); if (sort_expression_projection_names_size != 1) throw Exception(ErrorCodes::LOGICAL_ERROR, @@ -4047,6 +4049,24 @@ ProjectionNames QueryAnalyzer::resolveSortNodeList(QueryTreeNodePtr & sort_node_ return result_projection_names; } +void QueryAnalyzer::validateSortingKeyType(const DataTypePtr & sorting_key_type, const IdentifierResolveScope & scope) const +{ + if (scope.context->getSettingsRef().allow_suspicious_types_in_order_by) + return; + + auto check = [](const IDataType & type) + { + if (isDynamic(type) || isVariant(type)) + throw Exception( + ErrorCodes::ILLEGAL_COLUMN, + "Data types Variant/Dynamic are not allowed in ORDER BY keys, because it can lead to unexpected results. " + "Set setting allow_suspicious_types_in_order_by = 1 in order to allow it"); + }; + + check(*sorting_key_type); + sorting_key_type->forEachChild(check); +} + namespace { @@ -4086,11 +4106,12 @@ void QueryAnalyzer::resolveGroupByNode(QueryNode & query_node_typed, IdentifierR expandTuplesInList(group_by_list); } - if (scope.group_by_use_nulls) + for (const auto & grouping_set : query_node_typed.getGroupBy().getNodes()) { - for (const auto & grouping_set : query_node_typed.getGroupBy().getNodes()) + for (const auto & group_by_elem : grouping_set->as()->getNodes()) { - for (const auto & group_by_elem : grouping_set->as()->getNodes()) + validateGroupByKeyType(group_by_elem->getResultType(), scope); + if (scope.group_by_use_nulls) scope.nullable_group_by_keys.insert(group_by_elem); } } @@ -4106,14 +4127,35 @@ void QueryAnalyzer::resolveGroupByNode(QueryNode & query_node_typed, IdentifierR auto & group_by_list = query_node_typed.getGroupBy().getNodes(); expandTuplesInList(group_by_list); - if (scope.group_by_use_nulls) + for (const auto & group_by_elem : query_node_typed.getGroupBy().getNodes()) { - for (const auto & group_by_elem : query_node_typed.getGroupBy().getNodes()) + validateGroupByKeyType(group_by_elem->getResultType(), scope); + if (scope.group_by_use_nulls) scope.nullable_group_by_keys.insert(group_by_elem); } } } +/** Validate data types of GROUP BY key. + */ +void QueryAnalyzer::validateGroupByKeyType(const DataTypePtr & group_by_key_type, const IdentifierResolveScope & scope) const +{ + if (scope.context->getSettingsRef().allow_suspicious_types_in_group_by) + return; + + auto check = [](const IDataType & type) + { + if (isDynamic(type) || isVariant(type)) + throw Exception( + ErrorCodes::ILLEGAL_COLUMN, + "Data types Variant/Dynamic are not allowed in GROUP BY keys, because it can lead to unexpected results. " + "Set setting allow_suspicious_types_in_group_by = 1 in order to allow it"); + }; + + check(*group_by_key_type); + group_by_key_type->forEachChild(check); +} + /** Resolve interpolate columns nodes list. */ void QueryAnalyzer::resolveInterpolateColumnsNodeList(QueryTreeNodePtr & interpolate_node_list, IdentifierResolveScope & scope) diff --git a/src/Analyzer/Resolve/QueryAnalyzer.h b/src/Analyzer/Resolve/QueryAnalyzer.h index 7f9088b35e5..c90ded09876 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.h +++ b/src/Analyzer/Resolve/QueryAnalyzer.h @@ -217,8 +217,12 @@ private: ProjectionNames resolveSortNodeList(QueryTreeNodePtr & sort_node_list, IdentifierResolveScope & scope); + void validateSortingKeyType(const DataTypePtr & sorting_key_type, const IdentifierResolveScope & scope) const; + void resolveGroupByNode(QueryNode & query_node_typed, IdentifierResolveScope & scope); + void validateGroupByKeyType(const DataTypePtr & group_by_key_type, const IdentifierResolveScope & scope) const; + void resolveInterpolateColumnsNodeList(QueryTreeNodePtr & interpolate_node_list, IdentifierResolveScope & scope); void resolveWindowNodeList(QueryTreeNodePtr & window_node_list, IdentifierResolveScope & scope); diff --git a/src/Core/Settings.h b/src/Core/Settings.h index 23dc2a8fdc5..a3c58144fd0 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -389,6 +389,9 @@ class IColumn; M(Bool, prefer_global_in_and_join, false, "If enabled, all IN/JOIN operators will be rewritten as GLOBAL IN/JOIN. It's useful when the to-be-joined tables are only available on the initiator and we need to always scatter their data on-the-fly during distributed processing with the GLOBAL keyword. It's also useful to reduce the need to access the external sources joining external tables.", 0) \ M(Bool, enable_vertical_final, true, "If enable, remove duplicated rows during FINAL by marking rows as deleted and filtering them later instead of merging rows", 0) \ \ + M(Bool, allow_suspicious_types_in_group_by, false, "Allow suspicious types like Variant/Dynamic in GROUP BY clause", 0) \ + M(Bool, allow_suspicious_types_in_order_by, false, "Allow suspicious types like Variant/Dynamic in ORDER BY clause", 0) \ + \ \ /** Limits during query execution are part of the settings. \ * Used to provide a more safe execution of queries from the user interface. \ diff --git a/src/Interpreters/ExpressionAnalyzer.cpp b/src/Interpreters/ExpressionAnalyzer.cpp index 7063b2162a0..166b6619bdc 100644 --- a/src/Interpreters/ExpressionAnalyzer.cpp +++ b/src/Interpreters/ExpressionAnalyzer.cpp @@ -1367,6 +1367,9 @@ bool SelectQueryExpressionAnalyzer::appendGroupBy(ExpressionActionsChain & chain } } + for (const auto & result_column : step.getResultColumns()) + validateGroupByKeyType(result_column.type); + if (optimize_aggregation_in_order) { for (auto & child : asts) @@ -1381,6 +1384,24 @@ bool SelectQueryExpressionAnalyzer::appendGroupBy(ExpressionActionsChain & chain return true; } +void SelectQueryExpressionAnalyzer::validateGroupByKeyType(const DB::DataTypePtr & key_type) const +{ + if (getContext()->getSettingsRef().allow_suspicious_types_in_group_by) + return; + + auto check = [](const IDataType & type) + { + if (isDynamic(type) || isVariant(type)) + throw Exception( + ErrorCodes::ILLEGAL_COLUMN, + "Data types Variant/Dynamic are not allowed in GROUP BY keys, because it can lead to unexpected results. " + "Set setting allow_suspicious_types_in_group_by = 1 in order to allow it"); + }; + + check(*key_type); + key_type->forEachChild(check); +} + void SelectQueryExpressionAnalyzer::appendAggregateFunctionsArguments(ExpressionActionsChain & chain, bool only_types) { const auto * select_query = getAggregatingQuery(); @@ -1564,6 +1585,9 @@ ActionsAndProjectInputsFlagPtr SelectQueryExpressionAnalyzer::appendOrderBy(Expr getRootActions(select_query->orderBy(), only_types, step.actions()->dag); + for (const auto & result_column : step.getResultColumns()) + validateOrderByKeyType(result_column.type); + bool with_fill = false; for (auto & child : select_query->orderBy()->children) @@ -1643,6 +1667,24 @@ ActionsAndProjectInputsFlagPtr SelectQueryExpressionAnalyzer::appendOrderBy(Expr return actions; } +void SelectQueryExpressionAnalyzer::validateOrderByKeyType(const DataTypePtr & key_type) const +{ + if (getContext()->getSettingsRef().allow_suspicious_types_in_order_by) + return; + + auto check = [](const IDataType & type) + { + if (isDynamic(type) || isVariant(type)) + throw Exception( + ErrorCodes::ILLEGAL_COLUMN, + "Data types Variant/Dynamic are not allowed in ORDER BY keys, because it can lead to unexpected results. " + "Set setting allow_suspicious_types_in_order_by = 1 in order to allow it"); + }; + + check(*key_type); + key_type->forEachChild(check); +} + bool SelectQueryExpressionAnalyzer::appendLimitBy(ExpressionActionsChain & chain, bool only_types) { const auto * select_query = getSelectQuery(); diff --git a/src/Interpreters/ExpressionAnalyzer.h b/src/Interpreters/ExpressionAnalyzer.h index dc038e10594..3b006ee2106 100644 --- a/src/Interpreters/ExpressionAnalyzer.h +++ b/src/Interpreters/ExpressionAnalyzer.h @@ -397,6 +397,7 @@ private: ActionsAndProjectInputsFlagPtr appendPrewhere(ExpressionActionsChain & chain, bool only_types); bool appendWhere(ExpressionActionsChain & chain, bool only_types); bool appendGroupBy(ExpressionActionsChain & chain, bool only_types, bool optimize_aggregation_in_order, ManyExpressionActions &); + void validateGroupByKeyType(const DataTypePtr & key_type) const; void appendAggregateFunctionsArguments(ExpressionActionsChain & chain, bool only_types); void appendWindowFunctionsArguments(ExpressionActionsChain & chain, bool only_types); @@ -409,6 +410,7 @@ private: bool appendHaving(ExpressionActionsChain & chain, bool only_types); /// appendSelect ActionsAndProjectInputsFlagPtr appendOrderBy(ExpressionActionsChain & chain, bool only_types, bool optimize_read_in_order, ManyExpressionActions &); + void validateOrderByKeyType(const DataTypePtr & key_type) const; bool appendLimitBy(ExpressionActionsChain & chain, bool only_types); /// appendProjectResult }; diff --git a/src/Storages/KeyDescription.cpp b/src/Storages/KeyDescription.cpp index 7e43966556e..bb0b6d3542d 100644 --- a/src/Storages/KeyDescription.cpp +++ b/src/Storages/KeyDescription.cpp @@ -151,6 +151,15 @@ KeyDescription KeyDescription::getSortingKeyFromAST( throw Exception(ErrorCodes::DATA_TYPE_CANNOT_BE_USED_IN_KEY, "Column {} with type {} is not allowed in key expression, it's not comparable", backQuote(result.sample_block.getByPosition(i).name), result.data_types.back()->getName()); + + auto check = [&](const IDataType & type) + { + if (isDynamic(type) || isVariant(type)) + throw Exception(ErrorCodes::DATA_TYPE_CANNOT_BE_USED_IN_KEY, "Column with type Variant/Dynamic is not allowed in key expression"); + }; + + check(*result.data_types.back()); + result.data_types.back()->forEachChild(check); } return result; diff --git a/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference new file mode 100644 index 00000000000..a3eac1cf3fa --- /dev/null +++ b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference @@ -0,0 +1,184 @@ +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +4 +3 +2 +0 +1 +4 +3 +2 +[4] +[3] +[2] +[0] +[1] +{'str':0} +{'str':1} +{'str':4} +{'str':3} +{'str':2} +0 +1 +4 +3 +2 +\N +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +4 +3 +2 +0 +1 +4 +3 +2 +[4] +[3] +[2] +[0] +[1] +{'str':0} +{'str':1} +{'str':4} +{'str':3} +{'str':2} +\N +0 +1 +4 +3 +2 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +[4] +[0] +[1] +[2] +[3] +{'str':0} +{'str':1} +{'str':2} +{'str':3} +{'str':4} +0 +1 +2 +3 +4 +\N +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +[4] +[0] +[1] +[2] +[3] +{'str':0} +{'str':1} +{'str':2} +{'str':3} +{'str':4} +0 +1 +2 +3 +4 +\N diff --git a/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql new file mode 100644 index 00000000000..a4ea6425622 --- /dev/null +++ b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql @@ -0,0 +1,154 @@ +set allow_experimental_variant_type=1; +set allow_experimental_dynamic_type=1; + +drop table if exists test; + +create table test (d Dynamic) engine=MergeTree order by d; -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Dynamic) engine=MergeTree order by tuple(d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Dynamic) engine=MergeTree order by array(d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Dynamic) engine=MergeTree order by map('str', d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Dynamic) engine=MergeTree order by tuple() primary key d; -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Dynamic) engine=MergeTree order by tuple() partition by d; -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Dynamic) engine=MergeTree order by tuple() partition by tuple(d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Dynamic) engine=MergeTree order by tuple() partition by array(d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Dynamic) engine=MergeTree order by tuple() partition by map('str', d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} + +create table test (d Variant(UInt64)) engine=MergeTree order by d; -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Variant(UInt64)) engine=MergeTree order by tuple(d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Variant(UInt64)) engine=MergeTree order by array(d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Variant(UInt64)) engine=MergeTree order by map('str', d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Variant(UInt64)) engine=MergeTree order by tuple() primary key d; -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Variant(UInt64)) engine=MergeTree order by tuple() partition by d; -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Variant(UInt64)) engine=MergeTree order by tuple() partition by tuple(d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Variant(UInt64)) engine=MergeTree order by tuple() partition by array(d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} +create table test (d Variant(UInt64)) engine=MergeTree order by tuple() partition by map('str', d); -- {serverError DATA_TYPE_CANNOT_BE_USED_IN_KEY} + +create table test (d Dynamic) engine=Memory; +insert into test select * from numbers(5); + +set allow_experimental_analyzer=1; + +set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_order_by=0; + +select * from test order by d; -- {serverError ILLEGAL_COLUMN} +select * from test order by tuple(d); -- {serverError ILLEGAL_COLUMN} +select * from test order by array(d); -- {serverError ILLEGAL_COLUMN} +select * from test order by map('str', d); -- {serverError ILLEGAL_COLUMN} + +select * from test group by d; -- {serverError ILLEGAL_COLUMN} +select * from test group by tuple(d); -- {serverError ILLEGAL_COLUMN} +select array(d) from test group by array(d); -- {serverError ILLEGAL_COLUMN} +select map('str', d) from test group by map('str', d); -- {serverError ILLEGAL_COLUMN} +select * from test group by grouping sets ((d), ('str')); -- {serverError ILLEGAL_COLUMN} + +set allow_suspicious_types_in_group_by=1; +set allow_suspicious_types_in_order_by=1; + +select * from test order by d; +select * from test order by tuple(d); +select * from test order by array(d); +select * from test order by map('str', d); + +select * from test group by d; +select * from test group by tuple(d); +select array(d) from test group by array(d); +select map('str', d) from test group by map('str', d); +select * from test group by grouping sets ((d), ('str')); + +set allow_experimental_analyzer=0; + +set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_order_by=0; + +select * from test order by d; -- {serverError ILLEGAL_COLUMN} +select * from test order by tuple(d); -- {serverError ILLEGAL_COLUMN} +select * from test order by array(d); -- {serverError ILLEGAL_COLUMN} +select * from test order by map('str', d); -- {serverError ILLEGAL_COLUMN} + +select * from test group by d; -- {serverError ILLEGAL_COLUMN} +select * from test group by tuple(d); -- {serverError ILLEGAL_COLUMN} +select array(d) from test group by array(d); -- {serverError ILLEGAL_COLUMN} +select map('str', d) from test group by map('str', d); -- {serverError ILLEGAL_COLUMN} +select * from test group by grouping sets ((d), ('str')); -- {serverError ILLEGAL_COLUMN} + +set allow_suspicious_types_in_group_by=1; +set allow_suspicious_types_in_order_by=1; + +select * from test order by d; +select * from test order by tuple(d); +select * from test order by array(d); +select * from test order by map('str', d); + +select * from test group by d; +select * from test group by tuple(d); +select array(d) from test group by array(d); +select map('str', d) from test group by map('str', d); +select * from test group by grouping sets ((d), ('str')); + +drop table test; + +create table test (d Variant(UInt64)) engine=Memory; +insert into test select * from numbers(5); + +set allow_experimental_analyzer=1; + +set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_order_by=0; + +select * from test order by d; -- {serverError ILLEGAL_COLUMN} +select * from test order by tuple(d); -- {serverError ILLEGAL_COLUMN} +select * from test order by array(d); -- {serverError ILLEGAL_COLUMN} +select * from test order by map('str', d); -- {serverError ILLEGAL_COLUMN} + +select * from test group by d; -- {serverError ILLEGAL_COLUMN} +select * from test group by tuple(d); -- {serverError ILLEGAL_COLUMN} +select array(d) from test group by array(d); -- {serverError ILLEGAL_COLUMN} +select map('str', d) from test group by map('str', d); -- {serverError ILLEGAL_COLUMN} +select * from test group by grouping sets ((d), ('str')); -- {serverError ILLEGAL_COLUMN} + +set allow_suspicious_types_in_group_by=1; +set allow_suspicious_types_in_order_by=1; + +select * from test order by d; +select * from test order by tuple(d); +select * from test order by array(d); +select * from test order by map('str', d); + +select * from test group by d; +select * from test group by tuple(d); +select array(d) from test group by array(d); +select map('str', d) from test group by map('str', d); +select * from test group by grouping sets ((d), ('str')); + +set allow_experimental_analyzer=0; + +set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_order_by=0; + +select * from test order by d; -- {serverError ILLEGAL_COLUMN} +select * from test order by tuple(d); -- {serverError ILLEGAL_COLUMN} +select * from test order by array(d); -- {serverError ILLEGAL_COLUMN} +select * from test order by map('str', d); -- {serverError ILLEGAL_COLUMN} + +select * from test group by d; -- {serverError ILLEGAL_COLUMN} +select * from test group by tuple(d); -- {serverError ILLEGAL_COLUMN} +select array(d) from test group by array(d); -- {serverError ILLEGAL_COLUMN} +select map('str', d) from test group by map('str', d); -- {serverError ILLEGAL_COLUMN} +select * from test group by grouping sets ((d), ('str')); -- {serverError ILLEGAL_COLUMN} + +set allow_suspicious_types_in_group_by=1; +set allow_suspicious_types_in_order_by=1; + +select * from test order by d; +select * from test order by tuple(d); +select * from test order by array(d); +select * from test order by map('str', d); + +select * from test group by d; +select * from test group by tuple(d); +select array(d) from test group by array(d); +select map('str', d) from test group by map('str', d); +select * from test group by grouping sets ((d), ('str')); + +drop table test; From 3923efbabf2a3273a055e2889a0df19a517b0b6b Mon Sep 17 00:00:00 2001 From: avogar Date: Wed, 18 Sep 2024 14:11:07 +0000 Subject: [PATCH 044/680] Update settings changes history --- src/Core/SettingsChangesHistory.cpp | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 5e831c6301c..c2e5e51ab75 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -75,6 +75,8 @@ static std::initializer_list Date: Wed, 18 Sep 2024 19:54:37 +0200 Subject: [PATCH 045/680] Apply suggestions from code review MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Co-authored-by: János Benjamin Antal --- docs/en/operations/settings/settings.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/docs/en/operations/settings/settings.md b/docs/en/operations/settings/settings.md index 7dde006b14d..56341205bf7 100644 --- a/docs/en/operations/settings/settings.md +++ b/docs/en/operations/settings/settings.md @@ -5689,18 +5689,18 @@ Allows or restricts using [Variant](../../sql-reference/data-types/variant.md) a Possible values: -- 1 — Usage of `Variant` and `Dynamic` types is not restricted. - 0 — Usage of `Variant` and `Dynamic` types is restricted. +- 1 — Usage of `Variant` and `Dynamic` types is not restricted. Default value: 0. -## allow_suspicious_types_in_group_by {#allow_suspicious_types_in_group_by} +## allow_suspicious_types_in_order_by {#allow_suspicious_types_in_order_by} -Allows or restricts using [Variant](../../sql-reference/data-types/variant.md) and [Dynamic](../../sql-reference/data-types/dynamic.md) types in GROUP BY keys. +Allows or restricts using [Variant](../../sql-reference/data-types/variant.md) and [Dynamic](../../sql-reference/data-types/dynamic.md) types in ORDER BY keys. Possible values: -- 1 — Usage of `Variant` and `Dynamic` types is not restricted. - 0 — Usage of `Variant` and `Dynamic` types is restricted. +- 1 — Usage of `Variant` and `Dynamic` types is not restricted. Default value: 0. From c0c04eabbc20d5ab69066d0c0fb8c1339602f0b5 Mon Sep 17 00:00:00 2001 From: avogar Date: Wed, 18 Sep 2024 18:50:16 +0000 Subject: [PATCH 046/680] Update test --- ...mic_variant_in_order_by_group_by.reference | 10 +++---- ...1_dynamic_variant_in_order_by_group_by.sql | 28 +++++++++++++------ 2 files changed, 25 insertions(+), 13 deletions(-) diff --git a/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference index a3eac1cf3fa..5c7b4cb0bea 100644 --- a/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference +++ b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference @@ -40,9 +40,9 @@ {'str':2} 0 1 -4 -3 2 +3 +4 \N 0 1 @@ -84,12 +84,12 @@ {'str':4} {'str':3} {'str':2} -\N 0 1 -4 -3 2 +3 +4 +\N 0 1 2 diff --git a/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql index a4ea6425622..6e4a39c7234 100644 --- a/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql +++ b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql @@ -28,7 +28,7 @@ insert into test select * from numbers(5); set allow_experimental_analyzer=1; -set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_group_by=1; set allow_suspicious_types_in_order_by=0; select * from test order by d; -- {serverError ILLEGAL_COLUMN} @@ -36,6 +36,9 @@ select * from test order by tuple(d); -- {serverError ILLEGAL_COLUMN} select * from test order by array(d); -- {serverError ILLEGAL_COLUMN} select * from test order by map('str', d); -- {serverError ILLEGAL_COLUMN} +set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_order_by=1; + select * from test group by d; -- {serverError ILLEGAL_COLUMN} select * from test group by tuple(d); -- {serverError ILLEGAL_COLUMN} select array(d) from test group by array(d); -- {serverError ILLEGAL_COLUMN} @@ -54,11 +57,11 @@ select * from test group by d; select * from test group by tuple(d); select array(d) from test group by array(d); select map('str', d) from test group by map('str', d); -select * from test group by grouping sets ((d), ('str')); +select * from test group by grouping sets ((d), ('str')) order by all; set allow_experimental_analyzer=0; -set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_group_by=1; set allow_suspicious_types_in_order_by=0; select * from test order by d; -- {serverError ILLEGAL_COLUMN} @@ -66,6 +69,9 @@ select * from test order by tuple(d); -- {serverError ILLEGAL_COLUMN} select * from test order by array(d); -- {serverError ILLEGAL_COLUMN} select * from test order by map('str', d); -- {serverError ILLEGAL_COLUMN} +set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_order_by=1; + select * from test group by d; -- {serverError ILLEGAL_COLUMN} select * from test group by tuple(d); -- {serverError ILLEGAL_COLUMN} select array(d) from test group by array(d); -- {serverError ILLEGAL_COLUMN} @@ -84,7 +90,7 @@ select * from test group by d; select * from test group by tuple(d); select array(d) from test group by array(d); select map('str', d) from test group by map('str', d); -select * from test group by grouping sets ((d), ('str')); +select * from test group by grouping sets ((d), ('str')) order by all; drop table test; @@ -93,7 +99,7 @@ insert into test select * from numbers(5); set allow_experimental_analyzer=1; -set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_group_by=1; set allow_suspicious_types_in_order_by=0; select * from test order by d; -- {serverError ILLEGAL_COLUMN} @@ -101,6 +107,9 @@ select * from test order by tuple(d); -- {serverError ILLEGAL_COLUMN} select * from test order by array(d); -- {serverError ILLEGAL_COLUMN} select * from test order by map('str', d); -- {serverError ILLEGAL_COLUMN} +set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_order_by=1; + select * from test group by d; -- {serverError ILLEGAL_COLUMN} select * from test group by tuple(d); -- {serverError ILLEGAL_COLUMN} select array(d) from test group by array(d); -- {serverError ILLEGAL_COLUMN} @@ -119,11 +128,11 @@ select * from test group by d; select * from test group by tuple(d); select array(d) from test group by array(d); select map('str', d) from test group by map('str', d); -select * from test group by grouping sets ((d), ('str')); +select * from test group by grouping sets ((d), ('str')) order by all; set allow_experimental_analyzer=0; -set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_group_by=1; set allow_suspicious_types_in_order_by=0; select * from test order by d; -- {serverError ILLEGAL_COLUMN} @@ -131,6 +140,9 @@ select * from test order by tuple(d); -- {serverError ILLEGAL_COLUMN} select * from test order by array(d); -- {serverError ILLEGAL_COLUMN} select * from test order by map('str', d); -- {serverError ILLEGAL_COLUMN} +set allow_suspicious_types_in_group_by=0; +set allow_suspicious_types_in_order_by=1; + select * from test group by d; -- {serverError ILLEGAL_COLUMN} select * from test group by tuple(d); -- {serverError ILLEGAL_COLUMN} select array(d) from test group by array(d); -- {serverError ILLEGAL_COLUMN} @@ -149,6 +161,6 @@ select * from test group by d; select * from test group by tuple(d); select array(d) from test group by array(d); select map('str', d) from test group by map('str', d); -select * from test group by grouping sets ((d), ('str')); +select * from test group by grouping sets ((d), ('str')) order by all; drop table test; From cb488681eb43016e6b9af904e12243b8bb0aea27 Mon Sep 17 00:00:00 2001 From: avogar Date: Wed, 18 Sep 2024 18:51:46 +0000 Subject: [PATCH 047/680] Fix style --- src/Databases/enableAllExperimentalSettings.cpp | 2 ++ src/Interpreters/ExpressionAnalyzer.cpp | 1 + 2 files changed, 3 insertions(+) diff --git a/src/Databases/enableAllExperimentalSettings.cpp b/src/Databases/enableAllExperimentalSettings.cpp index 9abe05d7bce..01e989dc10b 100644 --- a/src/Databases/enableAllExperimentalSettings.cpp +++ b/src/Databases/enableAllExperimentalSettings.cpp @@ -32,6 +32,8 @@ void enableAllExperimentalSettings(ContextMutablePtr context) context->setSetting("allow_suspicious_low_cardinality_types", 1); context->setSetting("allow_suspicious_fixed_string_types", 1); + context->setSetting("allow_suspicious_types_in_group_by", 1); + context->setSetting("allow_suspicious_types_in_order_by", 1); context->setSetting("allow_suspicious_indices", 1); context->setSetting("allow_suspicious_codecs", 1); context->setSetting("allow_hyperscan", 1); diff --git a/src/Interpreters/ExpressionAnalyzer.cpp b/src/Interpreters/ExpressionAnalyzer.cpp index 9dcf4cd76e4..2df006aff9b 100644 --- a/src/Interpreters/ExpressionAnalyzer.cpp +++ b/src/Interpreters/ExpressionAnalyzer.cpp @@ -98,6 +98,7 @@ namespace ErrorCodes extern const int NOT_IMPLEMENTED; extern const int UNKNOWN_IDENTIFIER; extern const int UNKNOWN_TYPE_OF_AST_NODE; + extern const int ILLEGAL_COLUMN; } namespace From fd021f658df9ecef6804da3885067061f842e5b2 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Wed, 18 Sep 2024 20:01:11 +0000 Subject: [PATCH 048/680] check steps before mergesort --- .../Optimizations/optimizeReadInOrder.cpp | 17 ++++++++-- .../QueryPlan/ReadFromMergeTree.cpp | 17 +++++----- src/Processors/QueryPlan/ReadFromMergeTree.h | 14 ++++++-- ...er_optimization_with_virtual_row.reference | 2 ++ ...in_order_optimization_with_virtual_row.sql | 32 +++++++++++++++++++ 5 files changed, 70 insertions(+), 12 deletions(-) diff --git a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp index e7468a3a3f2..d3ecb3cac6b 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp @@ -94,6 +94,17 @@ static QueryPlan::Node * findReadingStep(QueryPlan::Node & node, StepStack & bac return nullptr; } +static bool checkVirtualRowSupport(const StepStack & backward_path) +{ + for (size_t i = 0; i < backward_path.size() - 1; i++) + { + IQueryPlanStep * step = backward_path[i]; + if (!typeid_cast(step) && !typeid_cast(step)) + return false; + } + return true; +} + void updateStepsDataStreams(StepStack & steps_to_update) { /// update data stream's sorting properties for found transforms @@ -825,8 +836,10 @@ InputOrderInfoPtr buildInputOrderInfo(SortingStep & sorting, QueryPlan::Node & n if (!can_read) return nullptr; - if (!order_info->first_prefix_fixed) - reading->enableVirtualRow(); + if (!checkVirtualRowSupport(backward_path)) + reading->setVirtualRowStatus(ReadFromMergeTree::VirtualRowStatus::No); + else if (!order_info->first_prefix_fixed) + reading->setVirtualRowStatus(ReadFromMergeTree::VirtualRowStatus::Possible); } return order_info; diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 45dcb4616b1..2ac663e0680 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -1109,17 +1109,18 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( } } - /// If enabled in the optimization stage, check whether there are more than one branch. - if (enable_virtual_row) - enable_virtual_row = splitted_parts_and_ranges.size() > 1 - || (splitted_parts_and_ranges.size() == 1 && splitted_parts_and_ranges[0].size() > 1); + /// If possible in the optimization stage, check whether there are more than one branch. + if (virtual_row_status == VirtualRowStatus::Possible) + virtual_row_status = splitted_parts_and_ranges.size() > 1 + || (splitted_parts_and_ranges.size() == 1 && splitted_parts_and_ranges[0].size() > 1) + ? VirtualRowStatus::Yes : VirtualRowStatus::NoConsiderInLogicalPlan; for (auto && item : splitted_parts_and_ranges) { - /// If not enabled before, try to enable it when conditions meet, as in the following section of preliminary merge, - /// only ExpressionTransform is added between MergingSortedTransform and readFromMergeTree. - bool enable_current_virtual_row = enable_virtual_row; - if (!enable_virtual_row) + bool enable_current_virtual_row = false; + if (virtual_row_status == VirtualRowStatus::Yes) + enable_current_virtual_row = true; + else if (virtual_row_status == VirtualRowStatus::NoConsiderInLogicalPlan) enable_current_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit, diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.h b/src/Processors/QueryPlan/ReadFromMergeTree.h index aa1b9dcfdcb..767fcf3b0f8 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.h +++ b/src/Processors/QueryPlan/ReadFromMergeTree.h @@ -108,6 +108,14 @@ public: using AnalysisResultPtr = std::shared_ptr; + enum class VirtualRowStatus + { + NoConsiderInLogicalPlan, + Possible, + No, + Yes, + }; + ReadFromMergeTree( MergeTreeData::DataPartsVector parts_, MergeTreeData::MutationsSnapshotPtr mutations_snapshot_, @@ -210,7 +218,7 @@ public: void applyFilters(ActionDAGNodes added_filter_nodes) override; - void enableVirtualRow() { enable_virtual_row = true; } + void setVirtualRowStatus(VirtualRowStatus virtual_row_status_) { virtual_row_status = virtual_row_status_; } private: int getSortDirection() const @@ -284,7 +292,9 @@ private: std::optional read_task_callback; bool enable_vertical_final = false; bool enable_remove_parts_from_snapshot_optimization = true; - bool enable_virtual_row = false; + + VirtualRowStatus virtual_row_status = VirtualRowStatus::NoConsiderInLogicalPlan; + std::optional number_of_current_replica; }; diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference index 08dabf3ee06..499ac19d374 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.reference @@ -38,3 +38,5 @@ 1 3 1 2 1 1 +-- test distinct ---- +0 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index 7e3af6c057a..4c7bc5d17c7 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -182,3 +182,35 @@ optimize_read_in_order = 1, read_in_order_two_level_merge_threshold = 0; --force preliminary merge DROP TABLE function_pk; + +-- modified from 02317_distinct_in_order_optimization +SELECT '-- test distinct ----'; + +DROP TABLE IF EXISTS distinct_in_order SYNC; + +CREATE TABLE distinct_in_order +( + `a` int, + `b` int, + `c` int +) +ENGINE = MergeTree +ORDER BY (a, b) +SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; + +SYSTEM STOP MERGES distinct_in_order; + +INSERT INTO distinct_in_order SELECT + number % number, + number % 5, + number % 10 +FROM numbers(1, 1000000); + +SELECT DISTINCT a +FROM distinct_in_order +ORDER BY a ASC +SETTINGS read_in_order_two_level_merge_threshold = 0, +optimize_read_in_order = 1, +max_threads = 2; + +DROP TABLE distinct_in_order; From 926e28e35cb1d17d0bb66c06b613671d3eeeeac2 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Thu, 19 Sep 2024 02:52:23 +0300 Subject: [PATCH 049/680] Rollback part rename if it was deduplicated --- .../MergeTree/ReplicatedMergeTreeSink.cpp | 26 +++++++++++-------- 1 file changed, 15 insertions(+), 11 deletions(-) diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp index fb2bc2fada7..98c46edda25 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp @@ -583,7 +583,7 @@ bool ReplicatedMergeTreeSinkImpl::writeExistingPart(MergeTreeData::Mutabl { error = ErrorCodes::INSERT_WAS_DEDUPLICATED; if (!endsWith(part->getDataPartStorage().getRelativePath(), "detached/attaching_" + part->name + "/")) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected relative path for a part: {}", part->getDataPartStorage().getRelativePath()); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected relative path for a deduplicated part: {}", part->getDataPartStorage().getRelativePath()); fs::path new_relative_path = fs::path("detached") / part->getNewName(part->info); part->renameTo(new_relative_path, false); } @@ -1013,16 +1013,6 @@ std::pair, bool> ReplicatedMergeTreeSinkImpl:: } } - transaction.rollback(); - - if (!Coordination::isUserError(multi_code)) - throw Exception( - ErrorCodes::UNEXPECTED_ZOOKEEPER_ERROR, - "Unexpected ZooKeeper error while adding block {} with ID '{}': {}", - block_number, - toString(block_id), - multi_code); - auto failed_op_idx = zkutil::getFailedOpIndex(multi_code, responses); String failed_op_path = ops[failed_op_idx]->getPath(); @@ -1032,6 +1022,10 @@ std::pair, bool> ReplicatedMergeTreeSinkImpl:: LOG_INFO(log, "Block with ID {} already exists (it was just appeared) for part {}. Ignore it.", toString(block_id), part->name); + transaction.rollbackPartsToTemporaryState(); + part->is_temp = true; + part->renameTo(temporary_part_relative_path, false); + if constexpr (async_insert) { retry_context.conflict_block_ids = std::vector({failed_op_path}); @@ -1043,6 +1037,16 @@ std::pair, bool> ReplicatedMergeTreeSinkImpl:: return CommitRetryContext::DUPLICATED_PART; } + transaction.rollback(); // Not in working set (data_parts) + + if (!Coordination::isUserError(multi_code)) + throw Exception( + ErrorCodes::UNEXPECTED_ZOOKEEPER_ERROR, + "Unexpected ZooKeeper error while adding block {} with ID '{}': {}", + block_number, + toString(block_id), + multi_code); + if (multi_code == Coordination::Error::ZNONODE && failed_op_idx == block_unlock_op_idx) throw Exception(ErrorCodes::QUERY_WAS_CANCELLED, "Insert query (for block {}) was canceled by concurrent ALTER PARTITION or TRUNCATE", From f570e8e2c0715001ac0f1633c898699700068edb Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Thu, 19 Sep 2024 13:34:51 +0300 Subject: [PATCH 050/680] Remove debug comment --- src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp index 98c46edda25..3f5c70adb64 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp @@ -1037,7 +1037,7 @@ std::pair, bool> ReplicatedMergeTreeSinkImpl:: return CommitRetryContext::DUPLICATED_PART; } - transaction.rollback(); // Not in working set (data_parts) + transaction.rollback(); if (!Coordination::isUserError(multi_code)) throw Exception( From e290745fe113efdba60cd5c807b92ae415c03d77 Mon Sep 17 00:00:00 2001 From: avogar Date: Thu, 19 Sep 2024 12:39:57 +0000 Subject: [PATCH 051/680] Fix tests --- tests/queries/0_stateless/02989_variant_comparison.sql | 1 + tests/queries/0_stateless/03035_dynamic_sorting.sql | 1 + .../03036_dynamic_read_shared_subcolumns_small.sql.j2 | 1 + .../0_stateless/03036_dynamic_read_subcolumns_small.sql.j2 | 1 + tests/queries/0_stateless/03096_variant_in_primary_key.sql | 1 + tests/queries/0_stateless/03150_dynamic_type_mv_insert.sql | 1 + .../queries/0_stateless/03151_dynamic_type_scale_max_types.sql | 2 +- tests/queries/0_stateless/03158_dynamic_type_from_variant.sql | 1 + tests/queries/0_stateless/03159_dynamic_type_all_types.sql | 2 +- tests/queries/0_stateless/03162_dynamic_type_nested.sql | 1 + tests/queries/0_stateless/03163_dynamic_as_supertype.sql | 1 + .../03228_dynamic_serializations_uninitialized_value.sql | 1 + .../queries/0_stateless/03231_dynamic_not_safe_primary_key.sql | 1 + 13 files changed, 13 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/02989_variant_comparison.sql b/tests/queries/0_stateless/02989_variant_comparison.sql index e0dcbc97c27..4d09933fb7b 100644 --- a/tests/queries/0_stateless/02989_variant_comparison.sql +++ b/tests/queries/0_stateless/02989_variant_comparison.sql @@ -1,4 +1,5 @@ set allow_experimental_variant_type=1; +set allow_suspicious_types_in_order_by=1; create table test (v1 Variant(String, UInt64, Array(UInt32)), v2 Variant(String, UInt64, Array(UInt32))) engine=Memory; diff --git a/tests/queries/0_stateless/03035_dynamic_sorting.sql b/tests/queries/0_stateless/03035_dynamic_sorting.sql index e0039a348c6..b2f36fed08e 100644 --- a/tests/queries/0_stateless/03035_dynamic_sorting.sql +++ b/tests/queries/0_stateless/03035_dynamic_sorting.sql @@ -1,4 +1,5 @@ set allow_experimental_dynamic_type = 1; +set allow_suspicious_types_in_order_by=1; drop table if exists test; create table test (d1 Dynamic(max_types=2), d2 Dynamic(max_types=2)) engine=Memory; diff --git a/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_small.sql.j2 b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_small.sql.j2 index dde4f3f53c3..d6732d91e74 100644 --- a/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_small.sql.j2 +++ b/tests/queries/0_stateless/03036_dynamic_read_shared_subcolumns_small.sql.j2 @@ -1,6 +1,7 @@ set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; set allow_experimental_dynamic_type = 1; +set allow_suspicious_types_in_order_by = 1; drop table if exists test; diff --git a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_small.sql.j2 b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_small.sql.j2 index 3253d7a6c68..daf85077160 100644 --- a/tests/queries/0_stateless/03036_dynamic_read_subcolumns_small.sql.j2 +++ b/tests/queries/0_stateless/03036_dynamic_read_subcolumns_small.sql.j2 @@ -1,6 +1,7 @@ set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; set allow_experimental_dynamic_type = 1; +set allow_suspicious_types_in_order_by = 1; drop table if exists test; diff --git a/tests/queries/0_stateless/03096_variant_in_primary_key.sql b/tests/queries/0_stateless/03096_variant_in_primary_key.sql index 48fbc821bcc..c422b4c3cc5 100644 --- a/tests/queries/0_stateless/03096_variant_in_primary_key.sql +++ b/tests/queries/0_stateless/03096_variant_in_primary_key.sql @@ -1,4 +1,5 @@ set allow_experimental_variant_type=1; +set allow_suspicious_types_in_order_by=1; drop table if exists test; create table test (id UInt64, v Variant(UInt64, String)) engine=MergeTree order by (id, v); insert into test values (1, 1), (1, 'str_1'), (1, 2), (1, 'str_2'); diff --git a/tests/queries/0_stateless/03150_dynamic_type_mv_insert.sql b/tests/queries/0_stateless/03150_dynamic_type_mv_insert.sql index 71d5dd4abd1..0e5119a38e0 100644 --- a/tests/queries/0_stateless/03150_dynamic_type_mv_insert.sql +++ b/tests/queries/0_stateless/03150_dynamic_type_mv_insert.sql @@ -1,4 +1,5 @@ SET allow_experimental_dynamic_type=1; +SET allow_suspicious_types_in_order_by=1; DROP TABLE IF EXISTS null_table; CREATE TABLE null_table diff --git a/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.sql b/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.sql index e476d34a1db..30a86dbc892 100644 --- a/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.sql +++ b/tests/queries/0_stateless/03151_dynamic_type_scale_max_types.sql @@ -1,5 +1,5 @@ SET allow_experimental_dynamic_type=1; -set min_compress_block_size = 585572, max_compress_block_size = 373374, max_block_size = 60768, max_joined_block_size_rows = 18966, max_insert_threads = 5, max_threads = 50, max_read_buffer_size = 708232, connect_timeout_with_failover_ms = 2000, connect_timeout_with_failover_secure_ms = 3000, idle_connection_timeout = 36000, use_uncompressed_cache = true, stream_like_engine_allow_direct_select = true, replication_wait_for_inactive_replica_timeout = 30, compile_aggregate_expressions = false, min_count_to_compile_aggregate_expression = 0, compile_sort_description = false, group_by_two_level_threshold = 1000000, group_by_two_level_threshold_bytes = 12610083, enable_memory_bound_merging_of_aggregation_results = false, min_chunk_bytes_for_parallel_parsing = 18769830, merge_tree_coarse_index_granularity = 12, min_bytes_to_use_direct_io = 10737418240, min_bytes_to_use_mmap_io = 10737418240, log_queries = true, insert_quorum_timeout = 60000, merge_tree_read_split_ranges_into_intersecting_and_non_intersecting_injection_probability = 0.05000000074505806, http_response_buffer_size = 294986, fsync_metadata = true, http_send_timeout = 60., http_receive_timeout = 60., opentelemetry_start_trace_probability = 0.10000000149011612, max_bytes_before_external_group_by = 1, max_bytes_before_external_sort = 10737418240, max_bytes_before_remerge_sort = 1326536545, max_untracked_memory = 1048576, memory_profiler_step = 1048576, log_comment = '03151_dynamic_type_scale_max_types.sql', send_logs_level = 'fatal', prefer_localhost_replica = false, optimize_read_in_order = false, optimize_aggregation_in_order = true, aggregation_in_order_max_block_bytes = 27069500, read_in_order_two_level_merge_threshold = 75, allow_introspection_functions = true, database_atomic_wait_for_drop_and_detach_synchronously = true, remote_filesystem_read_method = 'read', local_filesystem_read_prefetch = true, remote_filesystem_read_prefetch = false, merge_tree_compact_parts_min_granules_to_multibuffer_read = 119, async_insert_busy_timeout_max_ms = 5000, read_from_filesystem_cache_if_exists_otherwise_bypass_cache = true, filesystem_cache_segments_batch_size = 10, use_page_cache_for_disks_without_file_cache = true, page_cache_inject_eviction = true, allow_prefetched_read_pool_for_remote_filesystem = false, filesystem_prefetch_step_marks = 50, filesystem_prefetch_min_bytes_for_single_read_task = 16777216, filesystem_prefetch_max_memory_usage = 134217728, filesystem_prefetches_limit = 10, optimize_sorting_by_input_stream_properties = false, allow_experimental_dynamic_type = true, session_timezone = 'Africa/Khartoum', prefer_warmed_unmerged_parts_seconds = 2; +SET allow_suspicious_types_in_order_by=1; drop table if exists to_table; diff --git a/tests/queries/0_stateless/03158_dynamic_type_from_variant.sql b/tests/queries/0_stateless/03158_dynamic_type_from_variant.sql index a18f985f217..429ac21b5eb 100644 --- a/tests/queries/0_stateless/03158_dynamic_type_from_variant.sql +++ b/tests/queries/0_stateless/03158_dynamic_type_from_variant.sql @@ -1,5 +1,6 @@ SET allow_experimental_dynamic_type=1; SET allow_experimental_variant_type=1; +SET allow_suspicious_types_in_order_by=1; CREATE TABLE test_variable (v Variant(String, UInt32, IPv6, Bool, DateTime64)) ENGINE = Memory; CREATE TABLE test_dynamic (d Dynamic) ENGINE = Memory; diff --git a/tests/queries/0_stateless/03159_dynamic_type_all_types.sql b/tests/queries/0_stateless/03159_dynamic_type_all_types.sql index 28b679e2214..cf8ba687d3f 100644 --- a/tests/queries/0_stateless/03159_dynamic_type_all_types.sql +++ b/tests/queries/0_stateless/03159_dynamic_type_all_types.sql @@ -3,7 +3,7 @@ SET allow_experimental_dynamic_type=1; SET allow_experimental_variant_type=1; SET allow_suspicious_low_cardinality_types=1; - +SET allow_suspicious_types_in_order_by=1; CREATE TABLE t (d Dynamic(max_types=254)) ENGINE = Memory; -- Integer types: signed and unsigned integers (UInt8, UInt16, UInt32, UInt64, UInt128, UInt256, Int8, Int16, Int32, Int64, Int128, Int256) diff --git a/tests/queries/0_stateless/03162_dynamic_type_nested.sql b/tests/queries/0_stateless/03162_dynamic_type_nested.sql index 94007459a9e..59c22491957 100644 --- a/tests/queries/0_stateless/03162_dynamic_type_nested.sql +++ b/tests/queries/0_stateless/03162_dynamic_type_nested.sql @@ -1,4 +1,5 @@ SET allow_experimental_dynamic_type=1; +SET allow_suspicious_types_in_order_by=1; CREATE TABLE t (d Dynamic) ENGINE = Memory; diff --git a/tests/queries/0_stateless/03163_dynamic_as_supertype.sql b/tests/queries/0_stateless/03163_dynamic_as_supertype.sql index baba637eea4..e859fbd1815 100644 --- a/tests/queries/0_stateless/03163_dynamic_as_supertype.sql +++ b/tests/queries/0_stateless/03163_dynamic_as_supertype.sql @@ -1,4 +1,5 @@ SET allow_experimental_dynamic_type=1; +SET allow_suspicious_types_in_order_by=1; SELECT if(number % 2, number::Dynamic(max_types=3), ('str_' || toString(number))::Dynamic(max_types=2)) AS d, toTypeName(d), dynamicType(d) FROM numbers(4); CREATE TABLE dynamic_test_1 (d Dynamic(max_types=3)) ENGINE = Memory; INSERT INTO dynamic_test_1 VALUES ('str_1'), (42::UInt64); diff --git a/tests/queries/0_stateless/03228_dynamic_serializations_uninitialized_value.sql b/tests/queries/0_stateless/03228_dynamic_serializations_uninitialized_value.sql index 8a565fe36b9..60e2439d45f 100644 --- a/tests/queries/0_stateless/03228_dynamic_serializations_uninitialized_value.sql +++ b/tests/queries/0_stateless/03228_dynamic_serializations_uninitialized_value.sql @@ -1,4 +1,5 @@ set allow_experimental_dynamic_type=1; +set allow_suspicious_types_in_group_by=1; set cast_keep_nullable=1; SELECT toFixedString('str', 3), 3, CAST(if(1 = 0, toInt8(3), NULL), 'Int32') AS x from numbers(10) GROUP BY GROUPING SETS ((CAST(toInt32(1), 'Int32')), ('str', 3), (CAST(toFixedString('str', 3), 'Dynamic')), (CAST(toFixedString(toFixedString('str', 3), 3), 'Dynamic'))); diff --git a/tests/queries/0_stateless/03231_dynamic_not_safe_primary_key.sql b/tests/queries/0_stateless/03231_dynamic_not_safe_primary_key.sql index f207581f482..101c7cfe8fa 100644 --- a/tests/queries/0_stateless/03231_dynamic_not_safe_primary_key.sql +++ b/tests/queries/0_stateless/03231_dynamic_not_safe_primary_key.sql @@ -1,4 +1,5 @@ SET allow_experimental_dynamic_type = 1; +SET allow_suspicious_types_in_order_by = 1; DROP TABLE IF EXISTS t0; DROP TABLE IF EXISTS t1; CREATE TABLE t0 (c0 Int) ENGINE = AggregatingMergeTree() ORDER BY (c0); From 050b51799ce1e636f7806cb7af6d1bbb1cf481e5 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Thu, 19 Sep 2024 14:48:38 +0000 Subject: [PATCH 052/680] add inner and outer read-in-order virtual row test --- ...in_order_optimization_with_virtual_row.sql | 3 ++- ...ization_with_virtual_row_explain.reference | 25 ++++++++++++++++++ ..._optimization_with_virtual_row_explain.sql | 26 +++++++++++++++++++ 3 files changed, 53 insertions(+), 1 deletion(-) create mode 100644 tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.reference create mode 100644 tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index 4c7bc5d17c7..f66b4be2c69 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -196,7 +196,8 @@ CREATE TABLE distinct_in_order ) ENGINE = MergeTree ORDER BY (a, b) -SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'; +SETTINGS index_granularity = 8192, +index_granularity_bytes = '10Mi'; SYSTEM STOP MERGES distinct_in_order; diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.reference b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.reference new file mode 100644 index 00000000000..33ef6b19222 --- /dev/null +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.reference @@ -0,0 +1,25 @@ +(Expression) +ExpressionTransform + (Sorting) + MergingSortedTransform 4 → 1 + (Expression) + ExpressionTransform × 4 + (ReadFromMergeTree) + ExpressionTransform × 5 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + ExpressionTransform + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql new file mode 100644 index 00000000000..668b21275b4 --- /dev/null +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql @@ -0,0 +1,26 @@ +-- Tags: no-random-merge-tree-settings + +SET optimize_read_in_order = 1, merge_tree_min_rows_for_concurrent_read = 1000; + +DROP TABLE IF EXISTS tab; + +CREATE TABLE tab +( + `t` DateTime +) +ENGINE = MergeTree +ORDER BY t +SETTINGS index_granularity = 1; + +SYSTEM STOP MERGES tab; + +INSERT INTO tab SELECT toDateTime('2024-01-10') + number FROM numbers(10000); +INSERT INTO tab SELECT toDateTime('2024-01-30') + number FROM numbers(10000); +INSERT INTO tab SELECT toDateTime('2024-01-20') + number FROM numbers(10000); + +EXPLAIN PIPELINE +SELECT * +FROM tab +ORDER BY t ASC +SETTINGS read_in_order_two_level_merge_threshold = 0, max_threads = 4, read_in_order_use_buffering = 0 +FORMAT tsv; \ No newline at end of file From b4e5c11fd775cf915dff9e816673fb699c99a307 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Fri, 20 Sep 2024 02:11:29 +0000 Subject: [PATCH 053/680] fix --- ...3031_read_in_order_optimization_with_virtual_row_explain.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql index 668b21275b4..8cdcb4628ec 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql @@ -1,4 +1,4 @@ --- Tags: no-random-merge-tree-settings +-- Tags: no-random-merge-tree-settings, no-object-storage SET optimize_read_in_order = 1, merge_tree_min_rows_for_concurrent_read = 1000; From 82b4986ee35f974efb48f7ffbb6c698d4e363e43 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Sat, 21 Sep 2024 14:53:45 +0000 Subject: [PATCH 054/680] use empty chunk with pk block --- .../Merges/Algorithms/MergeTreeReadInfo.h | 41 ++++++++++++++++--- .../Algorithms/MergingSortedAlgorithm.cpp | 12 +++++- src/Processors/Merges/IMergingTransform.cpp | 4 +- .../QueryPlan/BufferChunksTransform.cpp | 2 +- .../QueryPlan/ReadFromMergeTree.cpp | 21 ++++++++-- .../Transforms/VirtualRowTransform.cpp | 33 ++++----------- .../Transforms/VirtualRowTransform.h | 11 +---- .../MergeTree/MergeTreeSelectProcessor.cpp | 2 +- .../MergeTree/MergeTreeSequentialSource.cpp | 2 +- 9 files changed, 77 insertions(+), 51 deletions(-) diff --git a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h index 862fa1b5e9a..425df2c24b9 100644 --- a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h +++ b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h @@ -1,6 +1,7 @@ #pragma once #include +#include namespace DB { @@ -10,13 +11,16 @@ class MergeTreeReadInfo : public ChunkInfoCloneable { public: MergeTreeReadInfo() = delete; - explicit MergeTreeReadInfo(size_t part_level, bool virtual_row_) : - origin_merge_tree_part_level(part_level), virtual_row(virtual_row_) { } + explicit MergeTreeReadInfo(size_t part_level) : + origin_merge_tree_part_level(part_level) {} + explicit MergeTreeReadInfo(size_t part_level, const Block & pk_block_) : + origin_merge_tree_part_level(part_level), pk_block(pk_block_) {} MergeTreeReadInfo(const MergeTreeReadInfo & other) = default; size_t origin_merge_tree_part_level = 0; - /// If virtual_row is true, the chunk must contain the virtual row only. - bool virtual_row = false; + + /// If is virtual_row, block should not be empty. + Block pk_block; }; inline size_t getPartLevelFromChunk(const Chunk & chunk) @@ -27,12 +31,37 @@ inline size_t getPartLevelFromChunk(const Chunk & chunk) return 0; } -inline bool getVirtualRowFromChunk(const Chunk & chunk) +inline bool isVirtualRow(const Chunk & chunk) { const auto read_info = chunk.getChunkInfos().get(); if (read_info) - return read_info->virtual_row; + return read_info->pk_block.columns() > 0; return false; } +inline void setVirtualRow(Chunk & chunk, const Block & header) +{ + const auto read_info = chunk.getChunkInfos().get(); + chassert(read_info); + + const Block & pk_block = read_info->pk_block; + + Columns ordered_columns; + ordered_columns.reserve(header.columns()); + + for (size_t i = 0; i < header.columns(); ++i) + { + const ColumnWithTypeAndName & type_and_name = header.getByPosition(i); + ColumnPtr current_column = type_and_name.type->createColumn(); + + size_t pos = type_and_name.name.find_last_of("."); + String column_name = (pos == String::npos) ? type_and_name.name : type_and_name.name.substr(pos + 1); + + const ColumnWithTypeAndName * column = pk_block.findByName(column_name, true); + ordered_columns.push_back(column ? column->column : current_column->cloneResized(1)); + } + + chunk.setColumns(ordered_columns, 1); +} + } diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp index 9476d46d939..75c04c8ddb2 100644 --- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp @@ -55,6 +55,14 @@ void MergingSortedAlgorithm::addInput() void MergingSortedAlgorithm::initialize(Inputs inputs) { + for (auto & input : inputs) + { + if (!isVirtualRow(input.chunk)) + continue; + + setVirtualRow(input.chunk, header); + } + removeConstAndSparse(inputs); merged_data.initialize(header, inputs); current_inputs = std::move(inputs); @@ -139,7 +147,7 @@ IMergingAlgorithm::Status MergingSortedAlgorithm::mergeImpl(TSortingHeap & queue auto current = queue.current(); - if (getVirtualRowFromChunk(current_inputs[current.impl->order].chunk)) + if (isVirtualRow(current_inputs[current.impl->order].chunk)) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Virtual row is not implemented for Non-batch mode."); if (current.impl->isLast() && current_inputs[current.impl->order].skip_last_row) @@ -238,7 +246,7 @@ IMergingAlgorithm::Status MergingSortedAlgorithm::mergeBatchImpl(TSortingQueue & auto [current_ptr, initial_batch_size] = queue.current(); auto current = *current_ptr; - if (getVirtualRowFromChunk(current_inputs[current.impl->order].chunk)) + if (isVirtualRow(current_inputs[current.impl->order].chunk)) { /// If virtual row is detected, there should be only one row as a single chunk, /// and always skip this chunk to pull the next one. diff --git a/src/Processors/Merges/IMergingTransform.cpp b/src/Processors/Merges/IMergingTransform.cpp index 7488cf4769e..68957cd55f9 100644 --- a/src/Processors/Merges/IMergingTransform.cpp +++ b/src/Processors/Merges/IMergingTransform.cpp @@ -104,14 +104,14 @@ IProcessor::Status IMergingTransformBase::prepareInitializeInputs() /// we won't have to read any chunks anymore; /// If virtual row exists, let it pass through, so don't read more chunks. auto chunk = input.pull(true); - bool virtual_row = getVirtualRowFromChunk(chunk); + bool virtual_row = isVirtualRow(chunk); if (limit_hint == 0 && !virtual_row) input.setNeeded(); if (!virtual_row && ((limit_hint && chunk.getNumRows() < limit_hint) || always_read_till_end)) input.setNeeded(); - if (!chunk.hasRows()) + if (!virtual_row && !chunk.hasRows()) { if (!input.isFinished()) { diff --git a/src/Processors/QueryPlan/BufferChunksTransform.cpp b/src/Processors/QueryPlan/BufferChunksTransform.cpp index 47e2c2ba0d5..75f5f91d981 100644 --- a/src/Processors/QueryPlan/BufferChunksTransform.cpp +++ b/src/Processors/QueryPlan/BufferChunksTransform.cpp @@ -88,7 +88,7 @@ IProcessor::Status BufferChunksTransform::prepare() Chunk BufferChunksTransform::pullChunk(bool & virtual_row) { auto chunk = input.pull(); - virtual_row = getVirtualRowFromChunk(chunk); + virtual_row = isVirtualRow(chunk); if (!virtual_row) num_processed_rows += chunk.getNumRows(); diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 2ac663e0680..4b5e33e8b07 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -663,12 +663,25 @@ Pipe ReadFromMergeTree::readInOrder( if (enable_current_virtual_row && (read_type == ReadType::InOrder)) { + const auto & index = part_with_ranges.data_part->getIndex(); + const auto & primary_key = storage_snapshot->metadata->primary_key; + size_t mark_range_begin = part_with_ranges.ranges.front().begin; + + ColumnsWithTypeAndName pk_columns; + pk_columns.reserve(index->size()); + + for (size_t j = 0; j < index->size(); ++j) + { + auto column = primary_key.data_types[j]->createColumn()->cloneEmpty(); + column->insert((*(*index)[j])[mark_range_begin]); + pk_columns.push_back({std::move(column), primary_key.data_types[j], primary_key.column_names[j]}); + } + + Block pk_block(std::move(pk_columns)); + pipe.addSimpleTransform([&](const Block & header) { - return std::make_shared(header, - storage_snapshot->metadata->primary_key, - part_with_ranges.data_part->getIndex(), - part_with_ranges.ranges.front().begin); + return std::make_shared(header, pk_block); }); } diff --git a/src/Processors/Transforms/VirtualRowTransform.cpp b/src/Processors/Transforms/VirtualRowTransform.cpp index 9b904fc4ae2..92bf5ce3064 100644 --- a/src/Processors/Transforms/VirtualRowTransform.cpp +++ b/src/Processors/Transforms/VirtualRowTransform.cpp @@ -9,14 +9,10 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; } -VirtualRowTransform::VirtualRowTransform(const Block & header_, - const KeyDescription & primary_key_, - const IMergeTreeDataPart::Index & index_, - size_t mark_range_begin_) +VirtualRowTransform::VirtualRowTransform(const Block & header_, const Block & pk_block_) : IProcessor({header_}, {header_}) , input(inputs.front()), output(outputs.front()) - , header(header_), primary_key(primary_key_) - , index(index_), mark_range_begin(mark_range_begin_) + , header(header_), pk_block(pk_block_) { } @@ -89,29 +85,16 @@ void VirtualRowTransform::work() is_first = false; - /// Reorder the columns according to result_header - Columns ordered_columns; - ordered_columns.reserve(header.columns()); - for (size_t i = 0, j = 0; i < header.columns(); ++i) + Columns empty_columns; + empty_columns.reserve(header.columns()); + for (size_t i = 0; i < header.columns(); ++i) { const ColumnWithTypeAndName & type_and_name = header.getByPosition(i); - ColumnPtr current_column = type_and_name.type->createColumn(); - // ordered_columns.push_back(current_column->cloneResized(1)); - - if (j < index->size() && type_and_name.name == primary_key.column_names[j] - && type_and_name.type == primary_key.data_types[j]) - { - auto column = current_column->cloneEmpty(); - column->insert((*(*index)[j])[mark_range_begin]); - ordered_columns.push_back(std::move(column)); - ++j; - } - else - ordered_columns.push_back(current_column->cloneResized(1)); + empty_columns.push_back(type_and_name.type->createColumn()->cloneEmpty()); } - current_chunk.setColumns(ordered_columns, 1); - current_chunk.getChunkInfos().add(std::make_shared(0, true)); + current_chunk.setColumns(empty_columns, 0); + current_chunk.getChunkInfos().add(std::make_shared(0, pk_block)); } else { diff --git a/src/Processors/Transforms/VirtualRowTransform.h b/src/Processors/Transforms/VirtualRowTransform.h index b9f0cb46242..e3215393ad1 100644 --- a/src/Processors/Transforms/VirtualRowTransform.h +++ b/src/Processors/Transforms/VirtualRowTransform.h @@ -11,10 +11,7 @@ namespace DB class VirtualRowTransform : public IProcessor { public: - explicit VirtualRowTransform(const Block & header_, - const KeyDescription & primary_key_, - const IMergeTreeDataPart::Index & index_, - size_t mark_range_begin_); + explicit VirtualRowTransform(const Block & header_, const Block & pk_block_); String getName() const override { return "VirtualRowTransform"; } @@ -32,11 +29,7 @@ private: bool is_first = true; Block header; - KeyDescription primary_key; - /// PK index used in virtual row. - IMergeTreeDataPart::Index index; - /// The first range that might contain the candidate. - size_t mark_range_begin; + Block pk_block; }; } diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index 85f545d2a51..cafe8dc3fbf 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -147,7 +147,7 @@ ChunkAndProgress MergeTreeSelectProcessor::read() auto chunk = Chunk(ordered_columns, res.row_count); if (add_part_level) - chunk.getChunkInfos().add(std::make_shared(task->getInfo().data_part->info.level, false)); + chunk.getChunkInfos().add(std::make_shared(task->getInfo().data_part->info.level)); return ChunkAndProgress{ .chunk = std::move(chunk), diff --git a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp index c62326f82dd..835045735fe 100644 --- a/src/Storages/MergeTree/MergeTreeSequentialSource.cpp +++ b/src/Storages/MergeTree/MergeTreeSequentialSource.cpp @@ -267,7 +267,7 @@ try auto result = Chunk(std::move(res_columns), rows_read); if (add_part_level) - result.getChunkInfos().add(std::make_shared(data_part->info.level, false)); + result.getChunkInfos().add(std::make_shared(data_part->info.level)); return result; } } From 10ed5a8521da3eb91e9d7caaf4bd0bdb32bca25c Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Wed, 25 Sep 2024 17:48:16 +0000 Subject: [PATCH 055/680] fix --- src/Processors/Merges/Algorithms/MergeTreeReadInfo.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h index 425df2c24b9..98cb414875b 100644 --- a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h +++ b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h @@ -54,7 +54,7 @@ inline void setVirtualRow(Chunk & chunk, const Block & header) const ColumnWithTypeAndName & type_and_name = header.getByPosition(i); ColumnPtr current_column = type_and_name.type->createColumn(); - size_t pos = type_and_name.name.find_last_of("."); + size_t pos = type_and_name.name.find_last_of('.'); String column_name = (pos == String::npos) ? type_and_name.name : type_and_name.name.substr(pos + 1); const ColumnWithTypeAndName * column = pk_block.findByName(column_name, true); From b4d7174ccc8b39458b5b9bc6984178437ddb345f Mon Sep 17 00:00:00 2001 From: vdimir Date: Wed, 21 Aug 2024 19:21:21 +0000 Subject: [PATCH 056/680] [wip] select inner table for hash join --- docs/en/operations/settings/settings.md | 4 + src/Core/Joins.h | 11 ++ src/Core/Settings.cpp | 1 + src/Core/SettingsChangesHistory.cpp | 2 + src/Core/SettingsEnums.cpp | 4 + src/Core/SettingsEnums.h | 2 +- src/Interpreters/HashJoin/HashJoin.cpp | 3 + .../HashJoin/HashJoinMethodsImpl.h | 18 +++- src/Interpreters/TableJoin.cpp | 55 +++++++++- src/Interpreters/TableJoin.h | 19 +++- src/Interpreters/TreeRewriter.cpp | 5 +- src/Parsers/CreateQueryUUIDs.cpp | 2 +- src/Planner/CollectColumnIdentifiers.cpp | 22 ++++ src/Planner/PlannerJoinTree.cpp | 25 +++-- src/Processors/QueryPlan/JoinStep.cpp | 13 ++- src/Processors/QueryPlan/JoinStep.h | 4 + .../QueryPlan/Optimizations/Optimizations.h | 1 + .../QueryPlan/Optimizations/optimizeJoin.cpp | 100 ++++++++++++++++++ .../QueryPlan/Optimizations/optimizeTree.cpp | 4 + .../QueryPlan/ReadFromMemoryStorageStep.h | 2 + tests/clickhouse-test | 4 + .../02962_join_using_bug_57894.reference | 1 + .../02962_join_using_bug_57894.sql | 2 + 23 files changed, 285 insertions(+), 19 deletions(-) create mode 100644 src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp diff --git a/docs/en/operations/settings/settings.md b/docs/en/operations/settings/settings.md index 392b1831ce3..fcb6f610894 100644 --- a/docs/en/operations/settings/settings.md +++ b/docs/en/operations/settings/settings.md @@ -5630,6 +5630,10 @@ Minimal size of block to compress in CROSS JOIN. Zero value means - disable this Default value: `1GiB`. +## query_plan_join_inner_table_selection + +Select the side of the join to be the inner table in the query plan. Possible values: 'auto', 'left', 'right'. In `auto` mode, ClickHouse will try to choose the table with the smallest number of rows. + ## use_json_alias_for_old_object_type When enabled, `JSON` data type alias will be used to create an old [Object('json')](../../sql-reference/data-types/json.md) type instead of the new [JSON](../../sql-reference/data-types/newjson.md) type. diff --git a/src/Core/Joins.h b/src/Core/Joins.h index 96d2b51325c..41e1de43702 100644 --- a/src/Core/Joins.h +++ b/src/Core/Joins.h @@ -119,4 +119,15 @@ enum class JoinTableSide : uint8_t const char * toString(JoinTableSide join_table_side); +/// Setting to choose which table to use as the inner table in hash join +enum class JoinInnerTableSelectionMode : uint8_t +{ + /// Use left table + Left, + /// Use right table + Right, + /// Use the table with the smallest number of rows + Auto, +}; + } diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 07b4ecd7a24..57dc297432a 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -345,6 +345,7 @@ namespace ErrorCodes M(Bool, any_join_distinct_right_table_keys, false, "Enable old ANY JOIN logic with many-to-one left-to-right table keys mapping for all ANY JOINs. It leads to confusing not equal results for 't1 ANY LEFT JOIN t2' and 't2 ANY RIGHT JOIN t1'. ANY RIGHT JOIN needs one-to-many keys mapping to be consistent with LEFT one.", IMPORTANT) \ M(Bool, single_join_prefer_left_table, true, "For single JOIN in case of identifier ambiguity prefer left table", IMPORTANT) \ \ + M(JoinInnerTableSelectionMode, query_plan_join_inner_table_selection, "auto", "Select the side of the join to be the inner table in the query plan. Possible values: 'auto', 'left', 'right'.", 0) \ M(UInt64, preferred_block_size_bytes, 1000000, "This setting adjusts the data block size for query processing and represents additional fine-tuning to the more rough 'max_block_size' setting. If the columns are large and with 'max_block_size' rows the block size is likely to be larger than the specified amount of bytes, its size will be lowered for better CPU cache locality.", 0) \ \ M(UInt64, max_replica_delay_for_distributed_queries, 300, "If set, distributed queries of Replicated tables will choose servers with replication delay in seconds less than the specified value (not inclusive). Zero means do not take delay into account.", 0) \ diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index c9723deaad8..8a79853c091 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -90,6 +90,8 @@ static std::initializer_listenableEnalyzer()) + left_columns_count = table_join->getOutputColumns(JoinTableSide::Left).size(); bool flag_per_row = needUsedFlagsForPerRightTableRow(table_join); if (!flag_per_row) diff --git a/src/Interpreters/HashJoin/HashJoinMethodsImpl.h b/src/Interpreters/HashJoin/HashJoinMethodsImpl.h index 320c8851ce4..5753e37ff88 100644 --- a/src/Interpreters/HashJoin/HashJoinMethodsImpl.h +++ b/src/Interpreters/HashJoin/HashJoinMethodsImpl.h @@ -56,7 +56,6 @@ Block HashJoinMethods::joinBlockImpl( const auto & key_names = !is_join_get ? onexprs[i].key_names_left : onexprs[i].key_names_right; join_on_keys.emplace_back(block, key_names, onexprs[i].condColumnNames().first, join.key_sizes[i]); } - size_t existing_columns = block.columns(); /** If you use FULL or RIGHT JOIN, then the columns from the "left" table must be materialized. * Because if they are constants, then in the "not joined" rows, they may have different values @@ -99,6 +98,23 @@ Block HashJoinMethods::joinBlockImpl( added_columns.buildJoinGetOutput(); else added_columns.buildOutput(); + + const auto & table_join = join.table_join; + if (table_join->enableEnalyzer()) + { + std::unordered_set left_output_columns; + for (const auto & out_column : table_join->getOutputColumns(JoinTableSide::Left)) + left_output_columns.insert(out_column.name); + std::set to_erase; + for (size_t i = 0; i < block.columns(); ++i) + { + if (!left_output_columns.contains(block.getByPosition(i).name)) + to_erase.insert(i); + } + block.erase(to_erase); + } + size_t existing_columns = block.columns(); + for (size_t i = 0; i < added_columns.size(); ++i) block.insert(added_columns.moveColumn(i)); diff --git a/src/Interpreters/TableJoin.cpp b/src/Interpreters/TableJoin.cpp index 2532dddba3c..d17300c229e 100644 --- a/src/Interpreters/TableJoin.cpp +++ b/src/Interpreters/TableJoin.cpp @@ -143,6 +143,7 @@ TableJoin::TableJoin(const Settings & settings, VolumePtr tmp_volume_, Temporary , max_memory_usage(settings[Setting::max_memory_usage]) , tmp_volume(tmp_volume_) , tmp_data(tmp_data_) + , enable_analyzer(settings.allow_experimental_analyzer) { } @@ -161,6 +162,8 @@ void TableJoin::resetCollected() clauses.clear(); columns_from_joined_table.clear(); columns_added_by_join.clear(); + columns_from_left_table.clear(); + result_columns_from_left_table.clear(); original_names.clear(); renames.clear(); left_type_map.clear(); @@ -203,6 +206,20 @@ size_t TableJoin::rightKeyInclusion(const String & name) const return count; } +void TableJoin::setInputColumns(NamesAndTypesList left_output_columns, NamesAndTypesList right_output_columns) +{ + columns_from_left_table = left_output_columns; + columns_from_joined_table = right_output_columns; +} + + +const NamesAndTypesList & TableJoin::getOutputColumns(JoinTableSide side) +{ + if (side == JoinTableSide::Left) + return result_columns_from_left_table; + return columns_added_by_join; +} + void TableJoin::deduplicateAndQualifyColumnNames(const NameSet & left_table_columns, const String & right_table_prefix) { NameSet joined_columns; @@ -351,9 +368,18 @@ bool TableJoin::rightBecomeNullable(const DataTypePtr & column_type) const return forceNullableRight() && JoinCommon::canBecomeNullable(column_type); } +void TableJoin::setUsedColumn(const NameAndTypePair & joined_column, JoinTableSide side) +{ + if (side == JoinTableSide::Left) + result_columns_from_left_table.push_back(joined_column); + else + columns_added_by_join.push_back(joined_column); + +} + void TableJoin::addJoinedColumn(const NameAndTypePair & joined_column) { - columns_added_by_join.emplace_back(joined_column); + setUsedColumn(joined_column, JoinTableSide::Right); } NamesAndTypesList TableJoin::correctedColumnsAddedByJoin() const @@ -995,5 +1021,32 @@ size_t TableJoin::getMaxMemoryUsage() const return max_memory_usage; } +void TableJoin::swapSides() +{ + assertEnableEnalyzer(); + + std::swap(key_asts_left, key_asts_right); + std::swap(left_type_map, right_type_map); + for (auto & clause : clauses) + { + std::swap(clause.key_names_left, clause.key_names_right); + std::swap(clause.on_filter_condition_left, clause.on_filter_condition_right); + std::swap(clause.analyzer_left_filter_condition_column_name, clause.analyzer_right_filter_condition_column_name); + } + + std::swap(columns_from_left_table, columns_from_joined_table); + std::swap(result_columns_from_left_table, columns_added_by_join); + + if (table_join.kind == JoinKind::Left) + table_join.kind = JoinKind::Right; + else if (table_join.kind == JoinKind::Right) + table_join.kind = JoinKind::Left; +} + +void TableJoin::assertEnableEnalyzer() const +{ + if (!enable_analyzer) + throw DB::Exception(ErrorCodes::NOT_IMPLEMENTED, "TableJoin: analyzer is disabled"); +} } diff --git a/src/Interpreters/TableJoin.h b/src/Interpreters/TableJoin.h index e1bae55a4ed..e0e1926fb12 100644 --- a/src/Interpreters/TableJoin.h +++ b/src/Interpreters/TableJoin.h @@ -167,6 +167,9 @@ private: ASOFJoinInequality asof_inequality = ASOFJoinInequality::GreaterOrEquals; + NamesAndTypesList columns_from_left_table; + NamesAndTypesList result_columns_from_left_table; + /// All columns which can be read from joined table. Duplicating names are qualified. NamesAndTypesList columns_from_joined_table; /// Columns will be added to block by JOIN. @@ -202,6 +205,8 @@ private: bool is_join_with_constant = false; + bool enable_analyzer = false; + Names requiredJoinedNames() const; /// Create converting actions and change key column names if required @@ -266,6 +271,8 @@ public: VolumePtr getGlobalTemporaryVolume() { return tmp_volume; } TemporaryDataOnDiskScopePtr getTempDataOnDisk() { return tmp_data; } + bool enableEnalyzer() const { return enable_analyzer; } + void assertEnableEnalyzer() const; ActionsDAG createJoinedBlockActions(ContextPtr context) const; @@ -282,6 +289,7 @@ public: } bool allowParallelHashJoin() const; + void swapSides(); bool joinUseNulls() const { return join_use_nulls; } @@ -372,6 +380,9 @@ public: bool leftBecomeNullable(const DataTypePtr & column_type) const; bool rightBecomeNullable(const DataTypePtr & column_type) const; void addJoinedColumn(const NameAndTypePair & joined_column); + + void setUsedColumn(const NameAndTypePair & joined_column, JoinTableSide side); + void setColumnsAddedByJoin(const NamesAndTypesList & columns_added_by_join_value) { columns_added_by_join = columns_added_by_join_value; @@ -397,11 +408,17 @@ public: ASTPtr leftKeysList() const; ASTPtr rightKeysList() const; /// For ON syntax only - void setColumnsFromJoinedTable(NamesAndTypesList columns_from_joined_table_value, const NameSet & left_table_columns, const String & right_table_prefix) + void setColumnsFromJoinedTable(NamesAndTypesList columns_from_joined_table_value, const NameSet & left_table_columns, const String & right_table_prefix, const NamesAndTypesList & columns_from_left_table_) { columns_from_joined_table = std::move(columns_from_joined_table_value); deduplicateAndQualifyColumnNames(left_table_columns, right_table_prefix); + result_columns_from_left_table = columns_from_left_table_; + columns_from_left_table = columns_from_left_table_; } + + void setInputColumns(NamesAndTypesList left_output_columns, NamesAndTypesList right_output_columns); + const NamesAndTypesList & getOutputColumns(JoinTableSide side); + const NamesAndTypesList & columnsFromJoinedTable() const { return columns_from_joined_table; } const NamesAndTypesList & columnsAddedByJoin() const { return columns_added_by_join; } diff --git a/src/Interpreters/TreeRewriter.cpp b/src/Interpreters/TreeRewriter.cpp index ea08fd92339..28e11166762 100644 --- a/src/Interpreters/TreeRewriter.cpp +++ b/src/Interpreters/TreeRewriter.cpp @@ -1353,12 +1353,15 @@ TreeRewriterResultPtr TreeRewriter::analyzeSelect( if (tables_with_columns.size() > 1) { + auto columns_from_left_table = tables_with_columns[0].columns; const auto & right_table = tables_with_columns[1]; auto columns_from_joined_table = right_table.columns; /// query can use materialized or aliased columns from right joined table, /// we want to request it for right table columns_from_joined_table.insert(columns_from_joined_table.end(), right_table.hidden_columns.begin(), right_table.hidden_columns.end()); - result.analyzed_join->setColumnsFromJoinedTable(std::move(columns_from_joined_table), source_columns_set, right_table.table.getQualifiedNamePrefix()); + columns_from_left_table.insert(columns_from_left_table.end(), tables_with_columns[0].hidden_columns.begin(), tables_with_columns[0].hidden_columns.end()); + result.analyzed_join->setColumnsFromJoinedTable( + std::move(columns_from_joined_table), source_columns_set, right_table.table.getQualifiedNamePrefix(), columns_from_left_table); } translateQualifiedNames(query, *select_query, source_columns_set, tables_with_columns); diff --git a/src/Parsers/CreateQueryUUIDs.cpp b/src/Parsers/CreateQueryUUIDs.cpp index fbdc6161408..1609ad43c69 100644 --- a/src/Parsers/CreateQueryUUIDs.cpp +++ b/src/Parsers/CreateQueryUUIDs.cpp @@ -31,7 +31,7 @@ CreateQueryUUIDs::CreateQueryUUIDs(const ASTCreateQuery & query, bool generate_r /// If we generate random UUIDs for already existing tables then those UUIDs will not be correct making those inner target table inaccessible. /// Thus it's not safe for example to replace /// "ATTACH MATERIALIZED VIEW mv AS SELECT a FROM b" with - /// "ATTACH MATERIALIZED VIEW mv TO INNER UUID "XXXX" AS SELECT a FROM b" + /// "ATTACH MATERIALIZED VIEW mv TO INNER UUID '123e4567-e89b-12d3-a456-426614174000' AS SELECT a FROM b" /// This replacement is safe only for CREATE queries when inner target tables don't exist yet. if (!query.attach) { diff --git a/src/Planner/CollectColumnIdentifiers.cpp b/src/Planner/CollectColumnIdentifiers.cpp index 95f1c7d53d8..ca468a353b2 100644 --- a/src/Planner/CollectColumnIdentifiers.cpp +++ b/src/Planner/CollectColumnIdentifiers.cpp @@ -2,6 +2,7 @@ #include #include +#include #include @@ -33,6 +34,27 @@ public: void visitImpl(const QueryTreeNodePtr & node) { + // if (node->getNodeType() == QueryTreeNodeType::QUERY) + // { + // const auto * join_node = node->as().getJoinTree()->as(); + // if (!join_node || !join_node->isUsingJoinExpression()) + // return; + + // const auto & using_list = join_node->getJoinExpression()->as(); + + // for (const auto & join_using_node : using_list.getNodes()) + // { + // const auto & join_using_expression = join_using_node->as().getExpression(); + // if (!join_using_expression) + // return; + // const auto & using_join_columns_list = join_using_expression->as().getNodes(); + // if (const auto * left_identifier = planner_context->getColumnNodeIdentifierOrNull(using_join_columns_list.at(0))) + // used_identifiers.insert(*left_identifier); + // if (const auto * right_identifier = planner_context->getColumnNodeIdentifierOrNull(using_join_columns_list.at(1))) + // used_identifiers.insert(*right_identifier); + // } + // } + if (node->getNodeType() != QueryTreeNodeType::COLUMN) return; diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp index 28789387d27..5a57d4e572d 100644 --- a/src/Planner/PlannerJoinTree.cpp +++ b/src/Planner/PlannerJoinTree.cpp @@ -1512,21 +1512,29 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ } const Block & left_header = left_plan.getCurrentDataStream().header; - auto left_table_names = left_header.getNames(); - NameSet left_table_names_set(left_table_names.begin(), left_table_names.end()); + const Block & right_header = right_plan.getCurrentDataStream().header; - auto columns_from_joined_table = right_plan.getCurrentDataStream().header.getNamesAndTypesList(); - table_join->setColumnsFromJoinedTable(columns_from_joined_table, left_table_names_set, ""); + auto columns_from_left_table = left_header.getNamesAndTypesList(); + auto columns_from_right_table = right_header.getNamesAndTypesList(); - for (auto & column_from_joined_table : columns_from_joined_table) + table_join->setInputColumns(columns_from_left_table, columns_from_right_table); + + for (auto & column_from_joined_table : columns_from_left_table) { - /// Add columns from joined table only if they are presented in outer scope, otherwise they can be dropped + /// Add columns to output only if they are presented in outer scope, otherwise they can be dropped if (planner_context->getGlobalPlannerContext()->hasColumnIdentifier(column_from_joined_table.name) && outer_scope_columns.contains(column_from_joined_table.name)) - table_join->addJoinedColumn(column_from_joined_table); + table_join->setUsedColumn(column_from_joined_table, JoinTableSide::Left); + } + + for (auto & column_from_joined_table : columns_from_right_table) + { + /// Add columns to output only if they are presented in outer scope, otherwise they can be dropped + if (planner_context->getGlobalPlannerContext()->hasColumnIdentifier(column_from_joined_table.name) && + outer_scope_columns.contains(column_from_joined_table.name)) + table_join->setUsedColumn(column_from_joined_table, JoinTableSide::Right); } - const Block & right_header = right_plan.getCurrentDataStream().header; auto join_algorithm = chooseJoinAlgorithm(table_join, join_node.getRightTableExpression(), left_header, right_header, planner_context); auto result_plan = QueryPlan(); @@ -1625,6 +1633,7 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ settings[Setting::max_block_size], settings[Setting::max_threads], false /*optimize_read_in_order*/); + join_step->inner_table_selection_mode = settings.query_plan_join_inner_table_selection; join_step->setStepDescription(fmt::format("JOIN {}", join_pipeline_type)); diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index 8fe2515e323..3f79a90149f 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -55,6 +55,9 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines if (pipelines.size() != 2) throw Exception(ErrorCodes::LOGICAL_ERROR, "JoinStep expect two input steps"); + if (swap_streams) + std::swap(pipelines[0], pipelines[1]); + if (join->pipelineType() == JoinPipelineType::YShaped) { auto joined_pipeline = QueryPipelineBuilder::joinPipelinesYShaped( @@ -63,7 +66,7 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines return joined_pipeline; } - return QueryPipelineBuilder::joinPipelinesRightLeft( + auto pipeline = QueryPipelineBuilder::joinPipelinesRightLeft( std::move(pipelines[0]), std::move(pipelines[1]), join, @@ -72,6 +75,7 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines max_streams, keep_left_read_in_order, &processors); + return pipeline; } bool JoinStep::allowPushDownToRight() const @@ -100,10 +104,9 @@ void JoinStep::describeActions(JSONBuilder::JSONMap & map) const void JoinStep::updateOutputStream() { - output_stream = DataStream - { - .header = JoiningTransform::transformHeader(input_streams[0].header, join), - }; + const auto & header = swap_streams ? input_streams[1].header : input_streams[0].header; + const auto & result_header = JoiningTransform::transformHeader(header, join); + output_stream = DataStream { .header = result_header }; } static ITransformingStep::Traits getStorageJoinTraits() diff --git a/src/Processors/QueryPlan/JoinStep.h b/src/Processors/QueryPlan/JoinStep.h index 51ea337b7c6..46fb49947ba 100644 --- a/src/Processors/QueryPlan/JoinStep.h +++ b/src/Processors/QueryPlan/JoinStep.h @@ -2,6 +2,7 @@ #include #include +#include namespace DB { @@ -36,6 +37,9 @@ public: bool canUpdateInputStream() const override { return true; } + JoinInnerTableSelectionMode inner_table_selection_mode = JoinInnerTableSelectionMode::Right; + bool swap_streams = false; + private: void updateOutputStream() override; diff --git a/src/Processors/QueryPlan/Optimizations/Optimizations.h b/src/Processors/QueryPlan/Optimizations/Optimizations.h index 43f07ced696..b81346e0fa1 100644 --- a/src/Processors/QueryPlan/Optimizations/Optimizations.h +++ b/src/Processors/QueryPlan/Optimizations/Optimizations.h @@ -116,6 +116,7 @@ void optimizePrimaryKeyConditionAndLimit(const Stack & stack); void optimizePrewhere(Stack & stack, QueryPlan::Nodes & nodes); void optimizeReadInOrder(QueryPlan::Node & node, QueryPlan::Nodes & nodes); void optimizeAggregationInOrder(QueryPlan::Node & node, QueryPlan::Nodes &); +void optimizeJoin(QueryPlan::Node & node, QueryPlan::Nodes &); /// Returns the name of used projection or nullopt if no projection is used. std::optional optimizeUseAggregateProjections(QueryPlan::Node & node, QueryPlan::Nodes & nodes, bool allow_implicit_projections); diff --git a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp new file mode 100644 index 00000000000..11e1c8d191c --- /dev/null +++ b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp @@ -0,0 +1,100 @@ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include +#include +#include + +namespace DB::QueryPlanOptimizations +{ + +static std::optional estimateReadRowsCount(QueryPlan::Node & node) +{ + IQueryPlanStep * step = node.step.get(); + if (const auto * reading = typeid_cast(step)) + { + if (auto analyzed_result = reading->getAnalyzedResult()) + return analyzed_result->selected_rows; + if (auto analyzed_result = reading->selectRangesToRead()) + return analyzed_result->selected_rows; + return {}; + } + + if (const auto * reading = typeid_cast(step)) + return reading->getStorage()->totalRows(Settings{}); + + if (node.children.size() != 1) + return {}; + + if (typeid_cast(step) || typeid_cast(step)) + return estimateReadRowsCount(*node.children.front()); + + return {}; +} + +void optimizeJoin(QueryPlan::Node & node, QueryPlan::Nodes &) +{ + auto * join_step = typeid_cast(node.step.get()); + if (!join_step || node.children.size() != 2) + return; + + const auto & join = join_step->getJoin(); + if (join->pipelineType() != JoinPipelineType::FillRightFirst || !join->isCloneSupported() || typeid_cast(join.get())) + return; + + const auto & table_join = join->getTableJoin(); + auto kind = table_join.kind(); + if (table_join.hasUsing() + || table_join.strictness() != JoinStrictness::All + || (kind != JoinKind::Inner && kind != JoinKind::Left + && kind != JoinKind::Right && kind != JoinKind::Full)) + return; + + bool need_swap = false; + if (join_step->inner_table_selection_mode == JoinInnerTableSelectionMode::Auto) + { + auto lhs_extimation = estimateReadRowsCount(*node.children[0]); + auto rhs_extimation = estimateReadRowsCount(*node.children[1]); + LOG_TRACE(getLogger("optimizeJoin"), "Left table estimation: {}, right table estimation: {}", + lhs_extimation.transform(toString).value_or("unknown"), + rhs_extimation.transform(toString).value_or("unknown")); + + if (lhs_extimation && rhs_extimation && *lhs_extimation < *rhs_extimation) + need_swap = true; + } + else if (join_step->inner_table_selection_mode == JoinInnerTableSelectionMode::Left) + { + need_swap = true; + } + + if (!need_swap) + return; + + const auto & streams = join_step->getInputStreams(); + if (streams.size() != 2) + return; + + const auto & left_stream_input_header = streams.front().header; + const auto & right_stream_input_header = streams.back().header; + join_step->swap_streams = true; + + auto updated_table_join = std::make_shared(table_join); + updated_table_join->swapSides(); + auto updated_join = join->clone(updated_table_join, right_stream_input_header, left_stream_input_header); + join_step->setJoin(std::move(updated_join)); +} + +} diff --git a/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp b/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp index f8504d84d12..a93f891eda2 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include @@ -226,6 +227,9 @@ void addStepsToBuildSets(QueryPlan & plan, QueryPlan::Node & root, QueryPlan::No /// NOTE: frame cannot be safely used after stack was modified. auto & frame = stack.back(); + if (frame.next_child == 0) + optimizeJoin(*frame.node, nodes); + /// Traverse all children first. if (frame.next_child < frame.node->children.size()) { diff --git a/src/Processors/QueryPlan/ReadFromMemoryStorageStep.h b/src/Processors/QueryPlan/ReadFromMemoryStorageStep.h index 238c1a3aad0..a9c2d2df2c4 100644 --- a/src/Processors/QueryPlan/ReadFromMemoryStorageStep.h +++ b/src/Processors/QueryPlan/ReadFromMemoryStorageStep.h @@ -35,6 +35,8 @@ public: void initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &) override; + const StoragePtr & getStorage() const { return storage; } + private: static constexpr auto name = "ReadFromMemoryStorage"; diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 810bae86cb0..be6a9a433a5 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -919,6 +919,10 @@ class SettingsRandomizer: "max_parsing_threads": lambda: random.choice([0, 1, 10]), "optimize_functions_to_subcolumns": lambda: random.randint(0, 1), "parallel_replicas_local_plan": lambda: random.randint(0, 1), + "query_plan_join_inner_table_selection": lambda: random.choice( + ["left", "auto"] + # ["left", "auto", "right"] + ), } @staticmethod diff --git a/tests/queries/0_stateless/02962_join_using_bug_57894.reference b/tests/queries/0_stateless/02962_join_using_bug_57894.reference index 454655081df..fc6fe462205 100644 --- a/tests/queries/0_stateless/02962_join_using_bug_57894.reference +++ b/tests/queries/0_stateless/02962_join_using_bug_57894.reference @@ -31,6 +31,7 @@ 8 9 \N +--- analyzer --- 0 1 2 diff --git a/tests/queries/0_stateless/02962_join_using_bug_57894.sql b/tests/queries/0_stateless/02962_join_using_bug_57894.sql index 96190241da5..e29347beb5e 100644 --- a/tests/queries/0_stateless/02962_join_using_bug_57894.sql +++ b/tests/queries/0_stateless/02962_join_using_bug_57894.sql @@ -21,6 +21,8 @@ SETTINGS join_algorithm = 'partial_merge'; SELECT x FROM t FULL JOIN r USING (x) ORDER BY ALL SETTINGS join_algorithm = 'full_sorting_merge'; +SELECT '--- analyzer ---'; + SET enable_analyzer = 1; SELECT x FROM t FULL JOIN r USING (x) ORDER BY ALL From 12e0b14d0ddbe58f4519c2cdcc877e7ca2818298 Mon Sep 17 00:00:00 2001 From: vdimir Date: Tue, 27 Aug 2024 10:37:41 +0000 Subject: [PATCH 057/680] fix column not found --- src/Interpreters/HashJoin/HashJoin.cpp | 2 +- src/Interpreters/HashJoin/HashJoinMethodsImpl.h | 8 ++++---- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/src/Interpreters/HashJoin/HashJoin.cpp b/src/Interpreters/HashJoin/HashJoin.cpp index 63edd837675..c6944313ee8 100644 --- a/src/Interpreters/HashJoin/HashJoin.cpp +++ b/src/Interpreters/HashJoin/HashJoin.cpp @@ -1230,7 +1230,7 @@ IBlocksStreamPtr HashJoin::getNonJoinedBlocks(const Block & left_sample_block, return {}; size_t left_columns_count = left_sample_block.columns(); - if (table_join->enableEnalyzer()) + if (table_join->enableEnalyzer() && !table_join->hasUsing()) left_columns_count = table_join->getOutputColumns(JoinTableSide::Left).size(); bool flag_per_row = needUsedFlagsForPerRightTableRow(table_join); diff --git a/src/Interpreters/HashJoin/HashJoinMethodsImpl.h b/src/Interpreters/HashJoin/HashJoinMethodsImpl.h index 5753e37ff88..2a7e029ab00 100644 --- a/src/Interpreters/HashJoin/HashJoinMethodsImpl.h +++ b/src/Interpreters/HashJoin/HashJoinMethodsImpl.h @@ -100,18 +100,17 @@ Block HashJoinMethods::joinBlockImpl( added_columns.buildOutput(); const auto & table_join = join.table_join; - if (table_join->enableEnalyzer()) + std::set block_columns_to_erase; + if (table_join->enableEnalyzer() && !table_join->hasUsing()) { std::unordered_set left_output_columns; for (const auto & out_column : table_join->getOutputColumns(JoinTableSide::Left)) left_output_columns.insert(out_column.name); - std::set to_erase; for (size_t i = 0; i < block.columns(); ++i) { if (!left_output_columns.contains(block.getByPosition(i).name)) - to_erase.insert(i); + block_columns_to_erase.insert(i); } - block.erase(to_erase); } size_t existing_columns = block.columns(); @@ -176,6 +175,7 @@ Block HashJoinMethods::joinBlockImpl( block.safeGetByPosition(pos).column = block.safeGetByPosition(pos).column->replicate(*offsets_to_replicate); } } + block.erase(block_columns_to_erase); return remaining_block; } From 7605a76a06c68dd0780af697c52531bb850cae06 Mon Sep 17 00:00:00 2001 From: vdimir Date: Tue, 27 Aug 2024 11:45:54 +0000 Subject: [PATCH 058/680] fix count() with query_plan_join_inner_table_selection --- src/Planner/PlannerJoinTree.cpp | 8 ++++++++ .../02514_analyzer_drop_join_on.reference | 10 ++-------- .../0_stateless/02514_analyzer_drop_join_on.sql | 1 + .../02835_join_step_explain.reference | 16 +++++++--------- .../0_stateless/02835_join_step_explain.sql | 2 ++ 5 files changed, 20 insertions(+), 17 deletions(-) diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp index 5a57d4e572d..1ffdf4e8c60 100644 --- a/src/Planner/PlannerJoinTree.cpp +++ b/src/Planner/PlannerJoinTree.cpp @@ -1535,6 +1535,14 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ table_join->setUsedColumn(column_from_joined_table, JoinTableSide::Right); } + if (table_join->getOutputColumns(JoinTableSide::Left).empty() && table_join->getOutputColumns(JoinTableSide::Right).empty()) + { + if (!columns_from_left_table.empty()) + table_join->setUsedColumn(columns_from_left_table.front(), JoinTableSide::Left); + else if (!columns_from_right_table.empty()) + table_join->setUsedColumn(columns_from_right_table.front(), JoinTableSide::Right); + } + auto join_algorithm = chooseJoinAlgorithm(table_join, join_node.getRightTableExpression(), left_header, right_header, planner_context); auto result_plan = QueryPlan(); diff --git a/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference b/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference index 2c62e278050..59983fff778 100644 --- a/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference +++ b/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference @@ -12,20 +12,17 @@ Header: count() UInt64 Header: __table1.a2 String Join (JOIN FillRightFirst) Header: __table1.a2 String - __table3.c1 UInt64 Expression ((JOIN actions + DROP unused columns after JOIN)) Header: __table1.a2 String __table3.c1 UInt64 Join (JOIN FillRightFirst) Header: __table1.a2 String - __table2.b1 UInt64 __table3.c1 UInt64 Expression ((JOIN actions + DROP unused columns after JOIN)) Header: __table1.a2 String __table2.b1 UInt64 Join (JOIN FillRightFirst) - Header: __table1.a1 UInt64 - __table1.a2 String + Header: __table1.a2 String __table2.b1 UInt64 Expression ((JOIN actions + Change column names to column identifiers)) Header: __table1.a1 UInt64 @@ -106,7 +103,6 @@ Header: bx String Header: __table1.a2 String __table2.bx String __table4.c2 String - __table4.c1 UInt64 Expression Header: __table1.a2 String __table2.bx String @@ -115,7 +111,6 @@ Header: bx String Join (JOIN FillRightFirst) Header: __table1.a2 String __table2.bx String - __table2.b1 UInt64 __table4.c2 String __table4.c1 UInt64 Expression ((JOIN actions + DROP unused columns after JOIN)) @@ -123,8 +118,7 @@ Header: bx String __table2.bx String __table2.b1 UInt64 Join (JOIN FillRightFirst) - Header: __table1.a1 UInt64 - __table1.a2 String + Header: __table1.a2 String __table2.bx String __table2.b1 UInt64 Expression ((JOIN actions + Change column names to column identifiers)) diff --git a/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql b/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql index df84e2f50b2..b10bf38e495 100644 --- a/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql +++ b/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql @@ -16,6 +16,7 @@ CREATE TABLE d (k UInt64, d1 UInt64, d2 String) ENGINE = Memory; INSERT INTO d VALUES (1, 1, 'a'), (2, 2, 'b'), (3, 3, 'c'); SET enable_analyzer = 1; +SET query_plan_join_inner_table_selection = 'right'; -- { echoOn } diff --git a/tests/queries/0_stateless/02835_join_step_explain.reference b/tests/queries/0_stateless/02835_join_step_explain.reference index 06f4a9cfc99..31205956662 100644 --- a/tests/queries/0_stateless/02835_join_step_explain.reference +++ b/tests/queries/0_stateless/02835_join_step_explain.reference @@ -57,19 +57,17 @@ Header: id UInt64 rhs.value_1 String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value_1 String : 1 - INPUT :: 2 -> __table1.value_2 UInt64 : 2 - INPUT : 3 -> __table2.value_1 String : 3 - INPUT :: 4 -> __table2.value_2 UInt64 : 4 - INPUT : 5 -> __table2.id UInt64 : 5 - ALIAS __table1.id :: 0 -> id UInt64 : 6 + INPUT : 2 -> __table2.value_1 String : 2 + INPUT :: 3 -> __table2.value_2 UInt64 : 3 + INPUT : 4 -> __table2.id UInt64 : 4 + ALIAS __table1.id :: 0 -> id UInt64 : 5 ALIAS __table1.value_1 :: 1 -> value_1 String : 0 - ALIAS __table2.value_1 :: 3 -> rhs.value_1 String : 1 - ALIAS __table2.id :: 5 -> rhs.id UInt64 : 3 -Positions: 6 0 3 1 + ALIAS __table2.value_1 :: 2 -> rhs.value_1 String : 1 + ALIAS __table2.id :: 4 -> rhs.id UInt64 : 2 +Positions: 5 0 2 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value_1 String - __table1.value_2 UInt64 __table2.value_1 String __table2.value_2 UInt64 __table2.id UInt64 diff --git a/tests/queries/0_stateless/02835_join_step_explain.sql b/tests/queries/0_stateless/02835_join_step_explain.sql index 1cdd3684a0b..b803ddbd911 100644 --- a/tests/queries/0_stateless/02835_join_step_explain.sql +++ b/tests/queries/0_stateless/02835_join_step_explain.sql @@ -19,6 +19,8 @@ CREATE TABLE test_table_2 INSERT INTO test_table_1 VALUES (0, 'Value', 0); INSERT INTO test_table_2 VALUES (0, 'Value', 0); +SET query_plan_join_inner_table_selection = 'right'; + EXPLAIN header = 1, actions = 1 SELECT lhs.id, lhs.value_1, rhs.id, rhs.value_1 FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id; From 0598419d5930054ac29030d62bb9c06823d53ae8 Mon Sep 17 00:00:00 2001 From: vdimir Date: Wed, 28 Aug 2024 11:40:12 +0000 Subject: [PATCH 059/680] Fix 'auto' join with inner table selection --- src/Interpreters/HashJoin/HashJoin.cpp | 15 ++++++++++++--- src/Interpreters/HashJoin/HashJoin.h | 3 +++ src/Interpreters/HashJoin/HashJoinMethodsImpl.h | 2 +- 3 files changed, 16 insertions(+), 4 deletions(-) diff --git a/src/Interpreters/HashJoin/HashJoin.cpp b/src/Interpreters/HashJoin/HashJoin.cpp index c6944313ee8..dad8a487745 100644 --- a/src/Interpreters/HashJoin/HashJoin.cpp +++ b/src/Interpreters/HashJoin/HashJoin.cpp @@ -383,6 +383,16 @@ size_t HashJoin::getTotalByteCount() const return res; } +bool HashJoin::isUsedByAnotherAlgorithm() const +{ + return table_join->isEnabledAlgorithm(JoinAlgorithm::AUTO) || table_join->isEnabledAlgorithm(JoinAlgorithm::GRACE_HASH); +} + +bool HashJoin::canRemoveColumnsFromLeftBlock() const +{ + return table_join->enableEnalyzer() && !table_join->hasUsing() && !isUsedByAnotherAlgorithm(); +} + void HashJoin::initRightBlockStructure(Block & saved_block_sample) { if (isCrossOrComma(kind)) @@ -394,8 +404,7 @@ void HashJoin::initRightBlockStructure(Block & saved_block_sample) bool multiple_disjuncts = !table_join->oneDisjunct(); /// We could remove key columns for LEFT | INNER HashJoin but we should keep them for JoinSwitcher (if any). - bool save_key_columns = table_join->isEnabledAlgorithm(JoinAlgorithm::AUTO) || - table_join->isEnabledAlgorithm(JoinAlgorithm::GRACE_HASH) || + bool save_key_columns = isUsedByAnotherAlgorithm() || isRightOrFull(kind) || multiple_disjuncts || table_join->getMixedJoinExpression(); @@ -1230,7 +1239,7 @@ IBlocksStreamPtr HashJoin::getNonJoinedBlocks(const Block & left_sample_block, return {}; size_t left_columns_count = left_sample_block.columns(); - if (table_join->enableEnalyzer() && !table_join->hasUsing()) + if (canRemoveColumnsFromLeftBlock()) left_columns_count = table_join->getOutputColumns(JoinTableSide::Left).size(); bool flag_per_row = needUsedFlagsForPerRightTableRow(table_join); diff --git a/src/Interpreters/HashJoin/HashJoin.h b/src/Interpreters/HashJoin/HashJoin.h index 4c1ebbcdc66..d5abdc2ddb8 100644 --- a/src/Interpreters/HashJoin/HashJoin.h +++ b/src/Interpreters/HashJoin/HashJoin.h @@ -464,6 +464,9 @@ private: bool empty() const; + bool isUsedByAnotherAlgorithm() const; + bool canRemoveColumnsFromLeftBlock() const; + void validateAdditionalFilterExpression(std::shared_ptr additional_filter_expression); bool needUsedFlagsForPerRightTableRow(std::shared_ptr table_join_) const; diff --git a/src/Interpreters/HashJoin/HashJoinMethodsImpl.h b/src/Interpreters/HashJoin/HashJoinMethodsImpl.h index 2a7e029ab00..ab522d94e37 100644 --- a/src/Interpreters/HashJoin/HashJoinMethodsImpl.h +++ b/src/Interpreters/HashJoin/HashJoinMethodsImpl.h @@ -101,7 +101,7 @@ Block HashJoinMethods::joinBlockImpl( const auto & table_join = join.table_join; std::set block_columns_to_erase; - if (table_join->enableEnalyzer() && !table_join->hasUsing()) + if (join.canRemoveColumnsFromLeftBlock()) { std::unordered_set left_output_columns; for (const auto & out_column : table_join->getOutputColumns(JoinTableSide::Left)) From 2b82db289386181f2e73c63eee7e98002e9e49fa Mon Sep 17 00:00:00 2001 From: vdimir Date: Wed, 25 Sep 2024 09:25:38 +0000 Subject: [PATCH 060/680] setting --- src/Core/Settings.cpp | 2 +- src/Core/Settings.h | 1 + src/Core/SettingsChangesHistory.cpp | 2 +- src/Interpreters/TableJoin.cpp | 3 ++- src/Planner/PlannerJoinTree.cpp | 4 +++- tests/clickhouse-test | 2 +- tests/integration/helpers/random_settings.py | 2 ++ 7 files changed, 11 insertions(+), 5 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 57dc297432a..4e63c3ae957 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -345,7 +345,7 @@ namespace ErrorCodes M(Bool, any_join_distinct_right_table_keys, false, "Enable old ANY JOIN logic with many-to-one left-to-right table keys mapping for all ANY JOINs. It leads to confusing not equal results for 't1 ANY LEFT JOIN t2' and 't2 ANY RIGHT JOIN t1'. ANY RIGHT JOIN needs one-to-many keys mapping to be consistent with LEFT one.", IMPORTANT) \ M(Bool, single_join_prefer_left_table, true, "For single JOIN in case of identifier ambiguity prefer left table", IMPORTANT) \ \ - M(JoinInnerTableSelectionMode, query_plan_join_inner_table_selection, "auto", "Select the side of the join to be the inner table in the query plan. Possible values: 'auto', 'left', 'right'.", 0) \ + M(JoinInnerTableSelectionMode, query_plan_join_inner_table_selection, JoinInnerTableSelectionMode::Auto, "Select the side of the join to be the inner table in the query plan. Possible values: 'auto', 'left', 'right'.", 0) \ M(UInt64, preferred_block_size_bytes, 1000000, "This setting adjusts the data block size for query processing and represents additional fine-tuning to the more rough 'max_block_size' setting. If the columns are large and with 'max_block_size' rows the block size is likely to be larger than the specified amount of bytes, its size will be lowered for better CPU cache locality.", 0) \ \ M(UInt64, max_replica_delay_for_distributed_queries, 300, "If set, distributed queries of Replicated tables will choose servers with replication delay in seconds less than the specified value (not inclusive). Zero means do not take delay into account.", 0) \ diff --git a/src/Core/Settings.h b/src/Core/Settings.h index 6bb66039afb..c413d285ba1 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -65,6 +65,7 @@ class WriteBuffer; M(CLASS_NAME, IntervalOutputFormat) \ M(CLASS_NAME, JoinAlgorithm) \ M(CLASS_NAME, JoinStrictness) \ + M(CLASS_NAME, JoinInnerTableSelectionMode) \ M(CLASS_NAME, LightweightMutationProjectionMode) \ M(CLASS_NAME, LoadBalancing) \ M(CLASS_NAME, LocalFSReadMethod) \ diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 8a79853c091..25954dc544c 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -70,6 +70,7 @@ static std::initializer_listinner_table_selection_mode = settings.query_plan_join_inner_table_selection; + if (settings[Setting::query_plan_join_inner_table_selection]) + join_step->inner_table_selection_mode = JoinInnerTableSelectionMode::Auto; join_step->setStepDescription(fmt::format("JOIN {}", join_pipeline_type)); diff --git a/tests/clickhouse-test b/tests/clickhouse-test index be6a9a433a5..1c606bea228 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -788,7 +788,7 @@ def threshold_generator(always_on_prob, always_off_prob, min_val, max_val): def get_localzone(): return os.getenv("TZ", "/".join(os.readlink("/etc/localtime").split("/")[-2:])) - +# Refer to `tests/integration/helpers/random_settings.py` for integration test random settings class SettingsRandomizer: settings = { "max_insert_threads": lambda: ( diff --git a/tests/integration/helpers/random_settings.py b/tests/integration/helpers/random_settings.py index b2319561fd7..49498b9f778 100644 --- a/tests/integration/helpers/random_settings.py +++ b/tests/integration/helpers/random_settings.py @@ -5,6 +5,8 @@ def randomize_settings(): yield "max_joined_block_size_rows", random.randint(8000, 100000) if random.random() < 0.5: yield "max_block_size", random.randint(8000, 100000) + if random.random() < 0.5: + yield "query_plan_join_inner_table_selection", random.choice(["auto", "left", "right"]) def write_random_settings_config(destination): From da2e6aeb32822416136195eb5d98f831fcbdb921 Mon Sep 17 00:00:00 2001 From: vdimir Date: Wed, 25 Sep 2024 11:50:27 +0000 Subject: [PATCH 061/680] join step swap header --- src/Interpreters/ConcurrentHashJoin.h | 7 ++ src/Interpreters/TableJoin.cpp | 4 +- src/Planner/CollectColumnIdentifiers.cpp | 21 ------ src/Processors/QueryPlan/JoinStep.cpp | 69 ++++++++++++++++++- .../QueryPlan/Optimizations/optimizeJoin.cpp | 2 +- .../QueryPlan/Optimizations/optimizeTree.cpp | 1 - 6 files changed, 78 insertions(+), 26 deletions(-) diff --git a/src/Interpreters/ConcurrentHashJoin.h b/src/Interpreters/ConcurrentHashJoin.h index a911edaccc3..355218554ce 100644 --- a/src/Interpreters/ConcurrentHashJoin.h +++ b/src/Interpreters/ConcurrentHashJoin.h @@ -60,6 +60,13 @@ public: IBlocksStreamPtr getNonJoinedBlocks(const Block & left_sample_block, const Block & result_sample_block, UInt64 max_block_size) const override; + + bool isCloneSupported() const override { return true; } + std::shared_ptr clone(const std::shared_ptr & table_join_, const Block &, const Block & right_sample_block_) const override + { + return std::make_shared(context, table_join_, slots, right_sample_block_, stats_collecting_params); + } + private: struct InternalHashJoin { diff --git a/src/Interpreters/TableJoin.cpp b/src/Interpreters/TableJoin.cpp index d4304df313c..555aaff2e06 100644 --- a/src/Interpreters/TableJoin.cpp +++ b/src/Interpreters/TableJoin.cpp @@ -209,8 +209,8 @@ size_t TableJoin::rightKeyInclusion(const String & name) const void TableJoin::setInputColumns(NamesAndTypesList left_output_columns, NamesAndTypesList right_output_columns) { - columns_from_left_table = left_output_columns; - columns_from_joined_table = right_output_columns; + columns_from_left_table = std::move(left_output_columns); + columns_from_joined_table = std::move(right_output_columns); } diff --git a/src/Planner/CollectColumnIdentifiers.cpp b/src/Planner/CollectColumnIdentifiers.cpp index ca468a353b2..dd5bdd4d141 100644 --- a/src/Planner/CollectColumnIdentifiers.cpp +++ b/src/Planner/CollectColumnIdentifiers.cpp @@ -34,27 +34,6 @@ public: void visitImpl(const QueryTreeNodePtr & node) { - // if (node->getNodeType() == QueryTreeNodeType::QUERY) - // { - // const auto * join_node = node->as().getJoinTree()->as(); - // if (!join_node || !join_node->isUsingJoinExpression()) - // return; - - // const auto & using_list = join_node->getJoinExpression()->as(); - - // for (const auto & join_using_node : using_list.getNodes()) - // { - // const auto & join_using_expression = join_using_node->as().getExpression(); - // if (!join_using_expression) - // return; - // const auto & using_join_columns_list = join_using_expression->as().getNodes(); - // if (const auto * left_identifier = planner_context->getColumnNodeIdentifierOrNull(using_join_columns_list.at(0))) - // used_identifiers.insert(*left_identifier); - // if (const auto * right_identifier = planner_context->getColumnNodeIdentifierOrNull(using_join_columns_list.at(1))) - // used_identifiers.insert(*right_identifier); - // } - // } - if (node->getNodeType() != QueryTreeNodeType::COLUMN) return; diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index 3f79a90149f..0e9332c186e 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -6,6 +6,7 @@ #include #include #include +#include namespace DB { @@ -36,6 +37,53 @@ std::vector> describeJoinActions(const JoinPtr & join) return description; } +size_t getPrefixLength(const NameSet & prefix, const Names & names) +{ + size_t i = 0; + for (; i < names.size(); ++i) + { + if (!prefix.contains(names[i])) + break; + } + LOG_DEBUG(&Poco::Logger::get("XXXX"), "{}:{}: [{}] [{}] -> {}", __FILE__, __LINE__, fmt::join(names, ", "), fmt::join(prefix, ", "), i); + return i; +} + +std::vector getPermutationToRotate(size_t prefix_size, size_t total_size) +{ + std::vector permutation(total_size); + size_t i = prefix_size; + for (auto & elem : permutation) + { + elem = i; + i = (i + 1) % total_size; + } + return permutation; +} + +Block rotateBlock(const Block & block, size_t prefix_size) +{ + auto columns = block.getColumnsWithTypeAndName(); + std::rotate(columns.begin(), columns.begin() + prefix_size, columns.end()); + auto res = Block(std::move(columns)); + return res; +} + +NameSet getNameSetFromBlock(const Block & block) +{ + NameSet names; + for (const auto & column : block) + names.insert(column.name); + return names; +} + +Block rotateBlock(const Block & block, const Block & prefix_block) +{ + NameSet prefix_names_set = getNameSetFromBlock(prefix_block); + size_t prefix_size = getPrefixLength(prefix_names_set, block.getNames()); + return rotateBlock(block, prefix_size); +} + } JoinStep::JoinStep( @@ -55,6 +103,8 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines if (pipelines.size() != 2) throw Exception(ErrorCodes::LOGICAL_ERROR, "JoinStep expect two input steps"); + NameSet rhs_names = getNameSetFromBlock(pipelines[1]->getHeader()); + if (swap_streams) std::swap(pipelines[0], pipelines[1]); @@ -75,6 +125,18 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines max_streams, keep_left_read_in_order, &processors); + + const auto & result_names = pipeline->getHeader().getNames(); + size_t prefix_size = getPrefixLength(rhs_names, result_names); + if (0 < prefix_size && prefix_size < result_names.size()) + { + auto column_permutation = getPermutationToRotate(prefix_size, result_names.size()); + pipeline->addSimpleTransform([column_perm = std::move(column_permutation)](const Block & header) + { + return std::make_shared(header, std::move(column_perm)); + }); + } + return pipeline; } @@ -105,7 +167,12 @@ void JoinStep::describeActions(JSONBuilder::JSONMap & map) const void JoinStep::updateOutputStream() { const auto & header = swap_streams ? input_streams[1].header : input_streams[0].header; - const auto & result_header = JoiningTransform::transformHeader(header, join); + + Block result_header = JoiningTransform::transformHeader(header, join); + + if (swap_streams) + result_header = rotateBlock(result_header, input_streams[1].header); + output_stream = DataStream { .header = result_header }; } diff --git a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp index 11e1c8d191c..8074304de52 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp @@ -52,7 +52,7 @@ void optimizeJoin(QueryPlan::Node & node, QueryPlan::Nodes &) return; const auto & join = join_step->getJoin(); - if (join->pipelineType() != JoinPipelineType::FillRightFirst || !join->isCloneSupported() || typeid_cast(join.get())) + if (join->pipelineType() != JoinPipelineType::FillRightFirst || !join->isCloneSupported()) return; const auto & table_join = join->getTableJoin(); diff --git a/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp b/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp index a93f891eda2..d58720268a6 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp @@ -4,7 +4,6 @@ #include #include #include -#include #include From bf591fa12b27f16411bf2441b06d1173616d34ba Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Thu, 26 Sep 2024 12:20:51 +0000 Subject: [PATCH 062/680] Introduce virtual row conversions. --- .../Merges/Algorithms/MergeTreeReadInfo.h | 40 +++-- .../Algorithms/MergingSortedAlgorithm.cpp | 7 +- .../Algorithms/MergingSortedAlgorithm.h | 5 +- .../Merges/MergingSortedTransform.cpp | 4 +- .../Merges/MergingSortedTransform.h | 1 + .../Optimizations/actionsDAGUtils.cpp | 2 + .../QueryPlan/Optimizations/actionsDAGUtils.h | 4 + .../Optimizations/distinctReadInOrder.cpp | 2 +- .../Optimizations/optimizeReadInOrder.cpp | 168 +++++++++++++----- .../QueryPlan/ReadFromMergeTree.cpp | 73 ++++---- src/Processors/QueryPlan/ReadFromMergeTree.h | 16 +- .../Transforms/MergeSortingTransform.cpp | 2 + .../Transforms/VirtualRowTransform.cpp | 8 +- .../Transforms/VirtualRowTransform.h | 4 +- src/Storages/ReadInOrderOptimizer.cpp | 2 +- src/Storages/SelectQueryInfo.h | 11 +- src/Storages/StorageMerge.cpp | 2 +- 17 files changed, 226 insertions(+), 125 deletions(-) diff --git a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h index 98cb414875b..62bbe3eac6e 100644 --- a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h +++ b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h @@ -2,6 +2,7 @@ #include #include +#include namespace DB { @@ -13,14 +14,15 @@ public: MergeTreeReadInfo() = delete; explicit MergeTreeReadInfo(size_t part_level) : origin_merge_tree_part_level(part_level) {} - explicit MergeTreeReadInfo(size_t part_level, const Block & pk_block_) : - origin_merge_tree_part_level(part_level), pk_block(pk_block_) {} + explicit MergeTreeReadInfo(size_t part_level, const Block & pk_block_, ExpressionActionsPtr virtual_row_conversions_) : + origin_merge_tree_part_level(part_level), pk_block(pk_block_), virtual_row_conversions(std::move(virtual_row_conversions_)) {} MergeTreeReadInfo(const MergeTreeReadInfo & other) = default; size_t origin_merge_tree_part_level = 0; /// If is virtual_row, block should not be empty. Block pk_block; + ExpressionActionsPtr virtual_row_conversions; }; inline size_t getPartLevelFromChunk(const Chunk & chunk) @@ -39,29 +41,33 @@ inline bool isVirtualRow(const Chunk & chunk) return false; } -inline void setVirtualRow(Chunk & chunk, const Block & header) +inline void setVirtualRow(Chunk & chunk, bool apply_virtual_row_conversions) { - const auto read_info = chunk.getChunkInfos().get(); + auto read_info = chunk.getChunkInfos().extract(); chassert(read_info); - const Block & pk_block = read_info->pk_block; + Block & pk_block = read_info->pk_block; + if (apply_virtual_row_conversions) + read_info->virtual_row_conversions->execute(pk_block); - Columns ordered_columns; - ordered_columns.reserve(header.columns()); + chunk.setColumns(pk_block.getColumns(), 1); - for (size_t i = 0; i < header.columns(); ++i) - { - const ColumnWithTypeAndName & type_and_name = header.getByPosition(i); - ColumnPtr current_column = type_and_name.type->createColumn(); + // Columns ordered_columns; + // ordered_columns.reserve(pk_block.columns()); - size_t pos = type_and_name.name.find_last_of('.'); - String column_name = (pos == String::npos) ? type_and_name.name : type_and_name.name.substr(pos + 1); + // for (size_t i = 0; i < header.columns(); ++i) + // { + // const ColumnWithTypeAndName & type_and_name = header.getByPosition(i); + // ColumnPtr current_column = type_and_name.type->createColumn(); - const ColumnWithTypeAndName * column = pk_block.findByName(column_name, true); - ordered_columns.push_back(column ? column->column : current_column->cloneResized(1)); - } + // size_t pos = type_and_name.name.find_last_of('.'); + // String column_name = (pos == String::npos) ? type_and_name.name : type_and_name.name.substr(pos + 1); - chunk.setColumns(ordered_columns, 1); + // const ColumnWithTypeAndName * column = pk_block.findByName(column_name, true); + // ordered_columns.push_back(column ? column->column : current_column->cloneResized(1)); + // } + + // chunk.setColumns(ordered_columns, 1); } } diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp index 75c04c8ddb2..0dd95729ba3 100644 --- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp @@ -22,12 +22,14 @@ MergingSortedAlgorithm::MergingSortedAlgorithm( SortingQueueStrategy sorting_queue_strategy_, UInt64 limit_, WriteBuffer * out_row_sources_buf_, - bool use_average_block_sizes) + bool use_average_block_sizes, + bool apply_virtual_row_conversions_) : header(std::move(header_)) , merged_data(use_average_block_sizes, max_block_size_, max_block_size_bytes_) , description(description_) , limit(limit_) , out_row_sources_buf(out_row_sources_buf_) + , apply_virtual_row_conversions(apply_virtual_row_conversions_) , current_inputs(num_inputs) , sorting_queue_strategy(sorting_queue_strategy_) , cursors(num_inputs) @@ -60,7 +62,8 @@ void MergingSortedAlgorithm::initialize(Inputs inputs) if (!isVirtualRow(input.chunk)) continue; - setVirtualRow(input.chunk, header); + setVirtualRow(input.chunk, apply_virtual_row_conversions); + input.skip_last_row = true; } removeConstAndSparse(inputs); diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h index c889668a38e..0a99b1bd8a6 100644 --- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h +++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.h @@ -22,7 +22,8 @@ public: SortingQueueStrategy sorting_queue_strategy_, UInt64 limit_ = 0, WriteBuffer * out_row_sources_buf_ = nullptr, - bool use_average_block_sizes = false); + bool use_average_block_sizes = false, + bool apply_virtual_row_conversions_ = true); void addInput(); @@ -47,6 +48,8 @@ private: /// If it is not nullptr then it should be populated during execution WriteBuffer * out_row_sources_buf = nullptr; + bool apply_virtual_row_conversions; + /// Chunks currently being merged. Inputs current_inputs; diff --git a/src/Processors/Merges/MergingSortedTransform.cpp b/src/Processors/Merges/MergingSortedTransform.cpp index d2895a2a2e9..760108facb6 100644 --- a/src/Processors/Merges/MergingSortedTransform.cpp +++ b/src/Processors/Merges/MergingSortedTransform.cpp @@ -22,6 +22,7 @@ MergingSortedTransform::MergingSortedTransform( bool always_read_till_end_, WriteBuffer * out_row_sources_buf_, bool use_average_block_sizes, + bool apply_virtual_row_conversions, bool have_all_inputs_) : IMergingTransform( num_inputs, @@ -38,7 +39,8 @@ MergingSortedTransform::MergingSortedTransform( sorting_queue_strategy, limit_, out_row_sources_buf_, - use_average_block_sizes) + use_average_block_sizes, + apply_virtual_row_conversions) { } diff --git a/src/Processors/Merges/MergingSortedTransform.h b/src/Processors/Merges/MergingSortedTransform.h index 6e52450efa7..220ecf0902a 100644 --- a/src/Processors/Merges/MergingSortedTransform.h +++ b/src/Processors/Merges/MergingSortedTransform.h @@ -22,6 +22,7 @@ public: bool always_read_till_end_ = false, WriteBuffer * out_row_sources_buf_ = nullptr, bool use_average_block_sizes = false, + bool apply_virtual_row_conversions = true, bool have_all_inputs_ = true); String getName() const override { return "MergingSortedTransform"; } diff --git a/src/Processors/QueryPlan/Optimizations/actionsDAGUtils.cpp b/src/Processors/QueryPlan/Optimizations/actionsDAGUtils.cpp index 2f1618ea6e1..b8216d6c4c4 100644 --- a/src/Processors/QueryPlan/Optimizations/actionsDAGUtils.cpp +++ b/src/Processors/QueryPlan/Optimizations/actionsDAGUtils.cpp @@ -210,6 +210,8 @@ MatchedTrees::Matches matchTrees(const ActionsDAG::NodeRawConstPtrs & inner_dag, MatchedTrees::Monotonicity monotonicity; monotonicity.direction *= info.is_positive ? 1 : -1; monotonicity.strict = info.is_strict; + monotonicity.child_match = &child_match; + monotonicity.child_node = monotonic_child; if (child_match.monotonicity) { diff --git a/src/Processors/QueryPlan/Optimizations/actionsDAGUtils.h b/src/Processors/QueryPlan/Optimizations/actionsDAGUtils.h index e78d658978e..82f0962f799 100644 --- a/src/Processors/QueryPlan/Optimizations/actionsDAGUtils.h +++ b/src/Processors/QueryPlan/Optimizations/actionsDAGUtils.h @@ -22,12 +22,16 @@ namespace DB /// DAG for PK does not contain aliases and ambiguous nodes. struct MatchedTrees { + struct Match; + /// Monotonicity is calculated for monotonic functions chain. /// Chain is not strict if there is any non-strict monotonic function. struct Monotonicity { int direction = 1; bool strict = true; + const Match * child_match = nullptr; + const ActionsDAG::Node * child_node = nullptr; }; struct Match diff --git a/src/Processors/QueryPlan/Optimizations/distinctReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/distinctReadInOrder.cpp index 37e61a6c388..5af680b42b8 100644 --- a/src/Processors/QueryPlan/Optimizations/distinctReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/distinctReadInOrder.cpp @@ -129,7 +129,7 @@ size_t tryDistinctReadInOrder(QueryPlan::Node * parent_node) /// update input order info in read_from_merge_tree step const int direction = 0; /// for DISTINCT direction doesn't matter, ReadFromMergeTree will choose proper one - bool can_read = read_from_merge_tree->requestReadingInOrder(number_of_sorted_distinct_columns, direction, pre_distinct->getLimitHint()); + bool can_read = read_from_merge_tree->requestReadingInOrder(number_of_sorted_distinct_columns, direction, pre_distinct->getLimitHint(), {}); if (!can_read) return 0; diff --git a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp index d3ecb3cac6b..8cd0a634a1e 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp @@ -94,17 +94,6 @@ static QueryPlan::Node * findReadingStep(QueryPlan::Node & node, StepStack & bac return nullptr; } -static bool checkVirtualRowSupport(const StepStack & backward_path) -{ - for (size_t i = 0; i < backward_path.size() - 1; i++) - { - IQueryPlanStep * step = backward_path[i]; - if (!typeid_cast(step) && !typeid_cast(step)) - return false; - } - return true; -} - void updateStepsDataStreams(StepStack & steps_to_update) { /// update data stream's sorting properties for found transforms @@ -338,11 +327,42 @@ void enrichFixedColumns(const ActionsDAG & dag, FixedColumns & fixed_columns) } } -InputOrderInfoPtr buildInputOrderInfo( +static const ActionsDAG::Node * addMonotonicChain(ActionsDAG & dag, const ActionsDAG::Node * node, const MatchedTrees::Match * match) +{ + if (!match->monotonicity) + return &dag.addInput(node->result_name, node->result_type); + + if (node->type == ActionsDAG::ActionType::ALIAS) + return &dag.addAlias(*addMonotonicChain(dag, node->children.front(), match), node->result_name); + + ActionsDAG::NodeRawConstPtrs args; + args.reserve(node->children.size()); + for (const auto * child : node->children) + { + if (child == match->monotonicity->child_node) + args.push_back(addMonotonicChain(dag, match->monotonicity->child_node, match->monotonicity->child_match)); + else + args.push_back(&dag.addColumn({child->column, child->result_type, child->result_name})); + } + + return &dag.addFunction(node->function_base, std::move(args), {}); +} + +struct SortingInputOrder +{ + InputOrderInfoPtr input_order{}; + /// This is needed for virtual row optimization. + /// Convert the PR values to ORDER BY key. + /// If empty, the optimization cannot be applied. + std::optional virtual_row_conversion{}; +}; + +SortingInputOrder buildInputOrderInfo( const FixedColumns & fixed_columns, const std::optional & dag, const SortDescription & description, const KeyDescription & sorting_key, + const Names & pk_column_names, size_t limit) { //std::cerr << "------- buildInputOrderInfo " << std::endl; @@ -381,7 +401,18 @@ InputOrderInfoPtr buildInputOrderInfo( int read_direction = 0; size_t next_description_column = 0; size_t next_sort_key = 0; - bool first_prefix_fixed = false; + + bool can_optimize_virtual_row = true; + + struct MatchInfo + { + const ActionsDAG::Node * source = nullptr; + const ActionsDAG::Node * fixed_column = nullptr; + const MatchedTrees::Match * monotonic = nullptr; + }; + + std::vector match_infos; + match_infos.reserve(description.size()); while (next_description_column < description.size() && next_sort_key < sorting_key.column_names.size()) { @@ -424,6 +455,7 @@ InputOrderInfoPtr buildInputOrderInfo( //std::cerr << "====== (no dag) Found direct match" << std::endl; + match_infos.push_back({.source = sort_column_node}); ++next_description_column; ++next_sort_key; } @@ -452,27 +484,46 @@ InputOrderInfoPtr buildInputOrderInfo( { current_direction *= match.monotonicity->direction; strict_monotonic = match.monotonicity->strict; + match_infos.push_back({.source = sort_node, .monotonic = &match}); } + else + match_infos.push_back({.source = sort_node}); ++next_description_column; ++next_sort_key; } else if (fixed_key_columns.contains(sort_column_node)) { + if (next_sort_key == 0) - first_prefix_fixed = true; + { + // Disable virtual row optimization. + // For example, when pk is (a,b), a = 1, order by b, virtual row should be + // disabled in the following case: + // 1st part (0, 100), (1, 2), (1, 3), (1, 4) + // 2nd part (0, 100), (1, 2), (1, 3), (1, 4). + + can_optimize_virtual_row = true; + } //std::cerr << "+++++++++ Found fixed key by match" << std::endl; ++next_sort_key; } else { - //std::cerr << "====== Check for fixed const : " << bool(sort_node->column) << " fixed : " << fixed_columns.contains(sort_node) << std::endl; bool is_fixed_column = sort_node->column || fixed_columns.contains(sort_node); if (!is_fixed_column) break; + if (!sort_node->column) + /// Virtual row for fixed column from order by is not supported now. + /// TODO: we can do it for the simple case, + /// But it's better to remove fixed columns from ORDER BY completely, e.g: + /// WHERE x = 42 ORDER BY x, y => WHERE x = 42 ORDER BY y + can_optimize_virtual_row = false; + + match_infos.push_back({.source = sort_node, .fixed_column = sort_node}); order_key_prefix_descr.push_back(sort_column_description); ++next_description_column; } @@ -494,9 +545,36 @@ InputOrderInfoPtr buildInputOrderInfo( } if (read_direction == 0 || order_key_prefix_descr.empty()) - return nullptr; + return {}; - return std::make_shared(order_key_prefix_descr, next_sort_key, read_direction, limit, first_prefix_fixed); + /// If the prefix description is used, we can't restore the full description from PK value. + /// TODO: partial sort description can be used as well. Implement support later. + if (order_key_prefix_descr.size() < description.size() || pk_column_names.size() < next_sort_key) + can_optimize_virtual_row = false; + + auto order_info = std::make_shared(order_key_prefix_descr, next_sort_key, read_direction, limit); + + std::optional virtual_row_conversion; + if (can_optimize_virtual_row) + { + ActionsDAG virtual_row_dag; + virtual_row_dag.getOutputs().reserve(match_infos.size()); + for (const auto & info : match_infos) + { + const ActionsDAG::Node * output; + if (info.fixed_column) + output = &virtual_row_dag.addColumn({info.fixed_column->column, info.fixed_column->result_type, info.fixed_column->result_name}); + else if (info.monotonic) + output = addMonotonicChain(virtual_row_dag, info.source, info.monotonic); + else + output = &virtual_row_dag.addInput(info.source->result_name, info.source->result_type); + + virtual_row_dag.getOutputs().push_back(output); + } + virtual_row_conversion = std::move(virtual_row_dag); + } + + return {std::move(order_info), std::move(virtual_row_conversion)}; } /// We really need three different sort descriptions here. @@ -700,11 +778,11 @@ AggregationInputOrder buildInputOrderInfo( for (const auto & key : not_matched_group_by_keys) group_by_sort_description.emplace_back(SortColumnDescription(std::string(key))); - auto input_order = std::make_shared(order_key_prefix_descr, next_sort_key, /*read_direction*/ 1, /* limit */ 0, false); + auto input_order = std::make_shared(order_key_prefix_descr, next_sort_key, /*read_direction*/ 1, /* limit */ 0); return { std::move(input_order), std::move(sort_description_for_merging), std::move(group_by_sort_description) }; } -InputOrderInfoPtr buildInputOrderInfo( +SortingInputOrder buildInputOrderInfo( const ReadFromMergeTree * reading, const FixedColumns & fixed_columns, const std::optional & dag, @@ -712,15 +790,17 @@ InputOrderInfoPtr buildInputOrderInfo( size_t limit) { const auto & sorting_key = reading->getStorageMetadata()->getSortingKey(); + const auto & pk_column_names = reading->getStorageMetadata()->getPrimaryKey().column_names; return buildInputOrderInfo( fixed_columns, dag, description, sorting_key, + pk_column_names, limit); } -InputOrderInfoPtr buildInputOrderInfo( +SortingInputOrder buildInputOrderInfo( ReadFromMerge * merge, const FixedColumns & fixed_columns, const std::optional & dag, @@ -729,28 +809,31 @@ InputOrderInfoPtr buildInputOrderInfo( { const auto & tables = merge->getSelectedTables(); - InputOrderInfoPtr order_info; + SortingInputOrder order_info; for (const auto & table : tables) { auto storage = std::get(table); - const auto & sorting_key = storage->getInMemoryMetadataPtr()->getSortingKey(); + auto metadata = storage->getInMemoryMetadataPtr(); + const auto & sorting_key = metadata->getSortingKey(); + // const auto & pk_column_names = metadata->getPrimaryKey().column_names; if (sorting_key.column_names.empty()) - return nullptr; + return {}; auto table_order_info = buildInputOrderInfo( fixed_columns, dag, description, sorting_key, + {}, limit); - if (!table_order_info) - return nullptr; + if (!table_order_info.input_order) + return {}; - if (!order_info) - order_info = table_order_info; - else if (*order_info != *table_order_info) - return nullptr; + if (!order_info.input_order) + order_info = std::move(table_order_info); + else if (*order_info.input_order != *table_order_info.input_order) + return {}; } return order_info; @@ -830,19 +913,19 @@ InputOrderInfoPtr buildInputOrderInfo(SortingStep & sorting, QueryPlan::Node & n dag, description, limit); - if (order_info) + if (order_info.input_order) { - bool can_read = reading->requestReadingInOrder(order_info->used_prefix_of_sorting_key_size, order_info->direction, order_info->limit); + bool can_read = reading->requestReadingInOrder( + order_info.input_order->used_prefix_of_sorting_key_size, + order_info.input_order->direction, + order_info.input_order->limit, + std::move(order_info.virtual_row_conversion)); + if (!can_read) return nullptr; - - if (!checkVirtualRowSupport(backward_path)) - reading->setVirtualRowStatus(ReadFromMergeTree::VirtualRowStatus::No); - else if (!order_info->first_prefix_fixed) - reading->setVirtualRowStatus(ReadFromMergeTree::VirtualRowStatus::Possible); } - return order_info; + return order_info.input_order; } else if (auto * merge = typeid_cast(reading_node->step.get())) { @@ -852,14 +935,14 @@ InputOrderInfoPtr buildInputOrderInfo(SortingStep & sorting, QueryPlan::Node & n dag, description, limit); - if (order_info) + if (order_info.input_order) { - bool can_read = merge->requestReadingInOrder(order_info); + bool can_read = merge->requestReadingInOrder(order_info.input_order); if (!can_read) return nullptr; } - return order_info; + return order_info.input_order; } return nullptr; @@ -893,7 +976,8 @@ AggregationInputOrder buildInputOrderInfo(AggregatingStep & aggregating, QueryPl bool can_read = reading->requestReadingInOrder( order_info.input_order->used_prefix_of_sorting_key_size, order_info.input_order->direction, - order_info.input_order->limit); + order_info.input_order->limit, + {}); if (!can_read) return {}; } @@ -1139,7 +1223,7 @@ size_t tryReuseStorageOrderingForWindowFunctions(QueryPlan::Node * parent_node, if (order_info) { - bool can_read = read_from_merge_tree->requestReadingInOrder(order_info->used_prefix_of_sorting_key_size, order_info->direction, order_info->limit); + bool can_read = read_from_merge_tree->requestReadingInOrder(order_info->used_prefix_of_sorting_key_size, order_info->direction, order_info->limit, {}); if (!can_read) return 0; sorting->convertToFinishSorting(order_info->sort_description_for_merging, false); diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 4b5e33e8b07..f4783862a50 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -549,8 +549,7 @@ Pipe ReadFromMergeTree::readInOrder( Names required_columns, PoolSettings pool_settings, ReadType read_type, - UInt64 read_limit, - bool enable_current_virtual_row) + UInt64 read_limit) { /// For reading in order it makes sense to read only /// one range per task to reduce number of read rows. @@ -661,7 +660,7 @@ Pipe ReadFromMergeTree::readInOrder( Pipe pipe(source); - if (enable_current_virtual_row && (read_type == ReadType::InOrder)) + if (virtual_row_conversion && (read_type == ReadType::InOrder)) { const auto & index = part_with_ranges.data_part->getIndex(); const auto & primary_key = storage_snapshot->metadata->primary_key; @@ -681,7 +680,7 @@ Pipe ReadFromMergeTree::readInOrder( pipe.addSimpleTransform([&](const Block & header) { - return std::make_shared(header, pk_block); + return std::make_shared(header, pk_block, virtual_row_conversion); }); } @@ -729,7 +728,7 @@ Pipe ReadFromMergeTree::read( if (read_type == ReadType::Default && (max_streams > 1 || checkAllPartsOnRemoteFS(parts_with_range))) return readFromPool(std::move(parts_with_range), std::move(required_columns), std::move(pool_settings)); - auto pipe = readInOrder(parts_with_range, required_columns, pool_settings, read_type, /*limit=*/ 0, false); + auto pipe = readInOrder(parts_with_range, required_columns, pool_settings, read_type, /*limit=*/ 0); /// Use ConcatProcessor to concat sources together. /// It is needed to read in parts order (and so in PK order) if single thread is used. @@ -1038,7 +1037,7 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( /// For parallel replicas the split will be performed on the initiator side. if (is_parallel_reading_from_replicas) { - pipes.emplace_back(readInOrder(std::move(parts_with_ranges), column_names, pool_settings, read_type, input_order_info->limit, false)); + pipes.emplace_back(readInOrder(std::move(parts_with_ranges), column_names, pool_settings, read_type, input_order_info->limit)); } else { @@ -1111,33 +1110,32 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( splitted_parts_and_ranges.emplace_back(std::move(new_parts)); } - bool primary_key_type_supports_virtual_row = true; - const auto & actions = storage_snapshot->metadata->getPrimaryKey().expression->getActions(); - for (const auto & action : actions) - { - if (action.node->type != ActionsDAG::ActionType::INPUT) - { - primary_key_type_supports_virtual_row = false; - break; - } - } + // bool primary_key_type_supports_virtual_row = true; + // const auto & actions = storage_snapshot->metadata->getPrimaryKey().expression->getActions(); + // for (const auto & action : actions) + // { + // if (action.node->type != ActionsDAG::ActionType::INPUT) + // { + // primary_key_type_supports_virtual_row = false; + // break; + // } + // } - /// If possible in the optimization stage, check whether there are more than one branch. - if (virtual_row_status == VirtualRowStatus::Possible) - virtual_row_status = splitted_parts_and_ranges.size() > 1 - || (splitted_parts_and_ranges.size() == 1 && splitted_parts_and_ranges[0].size() > 1) - ? VirtualRowStatus::Yes : VirtualRowStatus::NoConsiderInLogicalPlan; + // /// If possible in the optimization stage, check whether there are more than one branch. + // if (virtual_row_status == VirtualRowStatus::Possible) + // virtual_row_status = splitted_parts_and_ranges.size() > 1 + // || (splitted_parts_and_ranges.size() == 1 && splitted_parts_and_ranges[0].size() > 1) + // ? VirtualRowStatus::Yes : VirtualRowStatus::NoConsiderInLogicalPlan; for (auto && item : splitted_parts_and_ranges) { - bool enable_current_virtual_row = false; - if (virtual_row_status == VirtualRowStatus::Yes) - enable_current_virtual_row = true; - else if (virtual_row_status == VirtualRowStatus::NoConsiderInLogicalPlan) - enable_current_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; + // bool enable_current_virtual_row = false; + // if (virtual_row_status == VirtualRowStatus::Yes) + // enable_current_virtual_row = true; + // else if (virtual_row_status == VirtualRowStatus::NoConsiderInLogicalPlan) + // enable_current_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; - pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit, - enable_current_virtual_row && primary_key_type_supports_virtual_row)); + pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit)); } } @@ -1172,7 +1170,8 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( if (pipe.numOutputPorts() > 1) { auto transform = std::make_shared( - pipe.getHeader(), pipe.numOutputPorts(), sort_description, block_size.max_block_size_rows, /*max_block_size_bytes=*/0, SortingQueueStrategy::Batch); + pipe.getHeader(), pipe.numOutputPorts(), sort_description, block_size.max_block_size_rows, /*max_block_size_bytes=*/0, SortingQueueStrategy::Batch, + 0, false, nullptr, false, /*apply_virtual_row_conversions*/ false); pipe.addTransform(std::move(transform)); } @@ -1811,7 +1810,7 @@ ReadFromMergeTree::AnalysisResultPtr ReadFromMergeTree::selectRangesToRead( return std::make_shared(std::move(result)); } -bool ReadFromMergeTree::requestReadingInOrder(size_t prefix_size, int direction, size_t read_limit) +bool ReadFromMergeTree::requestReadingInOrder(size_t prefix_size, int direction, size_t read_limit, std::optional virtual_row_conversion_) { /// if dirction is not set, use current one if (!direction) @@ -1822,7 +1821,7 @@ bool ReadFromMergeTree::requestReadingInOrder(size_t prefix_size, int direction, if (direction != 1 && query_info.isFinal()) return false; - query_info.input_order_info = std::make_shared(SortDescription{}, prefix_size, direction, read_limit, false); + query_info.input_order_info = std::make_shared(SortDescription{}, prefix_size, direction, read_limit); reader_settings.read_in_order = true; /// In case or read-in-order, don't create too many reading streams. @@ -1855,6 +1854,9 @@ bool ReadFromMergeTree::requestReadingInOrder(size_t prefix_size, int direction, /// Let prefer in-order optimization over vertical FINAL for now enable_vertical_final = false; + if (virtual_row_conversion_) + virtual_row_conversion = std::make_shared(std::move(*virtual_row_conversion_)); + return true; } @@ -2305,6 +2307,12 @@ void ReadFromMergeTree::describeActions(FormatSettings & format_settings) const expression->describeActions(format_settings.out, prefix); } } + + if (virtual_row_conversion) + { + format_settings.out << prefix << "Virtual row conversions" << '\n'; + virtual_row_conversion->describeActions(format_settings.out, prefix); + } } void ReadFromMergeTree::describeActions(JSONBuilder::JSONMap & map) const @@ -2344,6 +2352,9 @@ void ReadFromMergeTree::describeActions(JSONBuilder::JSONMap & map) const map.add("Prewhere info", std::move(prewhere_info_map)); } + + if (virtual_row_conversion) + map.add("Virtual row conversions", virtual_row_conversion->toTree()); } void ReadFromMergeTree::describeIndexes(FormatSettings & format_settings) const diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.h b/src/Processors/QueryPlan/ReadFromMergeTree.h index 767fcf3b0f8..e20c06aeb53 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.h +++ b/src/Processors/QueryPlan/ReadFromMergeTree.h @@ -108,14 +108,6 @@ public: using AnalysisResultPtr = std::shared_ptr; - enum class VirtualRowStatus - { - NoConsiderInLogicalPlan, - Possible, - No, - Yes, - }; - ReadFromMergeTree( MergeTreeData::DataPartsVector parts_, MergeTreeData::MutationsSnapshotPtr mutations_snapshot_, @@ -195,7 +187,7 @@ public: StorageMetadataPtr getStorageMetadata() const { return storage_snapshot->metadata; } /// Returns `false` if requested reading cannot be performed. - bool requestReadingInOrder(size_t prefix_size, int direction, size_t limit); + bool requestReadingInOrder(size_t prefix_size, int direction, size_t limit, std::optional virtual_row_conversion_); bool readsInOrder() const; void updatePrewhereInfo(const PrewhereInfoPtr & prewhere_info_value) override; @@ -218,8 +210,6 @@ public: void applyFilters(ActionDAGNodes added_filter_nodes) override; - void setVirtualRowStatus(VirtualRowStatus virtual_row_status_) { virtual_row_status = virtual_row_status_; } - private: int getSortDirection() const { @@ -262,7 +252,7 @@ private: Pipe read(RangesInDataParts parts_with_range, Names required_columns, ReadType read_type, size_t max_streams, size_t min_marks_for_concurrent_read, bool use_uncompressed_cache); Pipe readFromPool(RangesInDataParts parts_with_range, Names required_columns, PoolSettings pool_settings); Pipe readFromPoolParallelReplicas(RangesInDataParts parts_with_range, Names required_columns, PoolSettings pool_settings); - Pipe readInOrder(RangesInDataParts parts_with_ranges, Names required_columns, PoolSettings pool_settings, ReadType read_type, UInt64 limit, bool enable_current_virtual_row); + Pipe readInOrder(RangesInDataParts parts_with_ranges, Names required_columns, PoolSettings pool_settings, ReadType read_type, UInt64 limit); Pipe spreadMarkRanges(RangesInDataParts && parts_with_ranges, size_t num_streams, AnalysisResult & result, std::optional & result_projection); @@ -293,7 +283,7 @@ private: bool enable_vertical_final = false; bool enable_remove_parts_from_snapshot_optimization = true; - VirtualRowStatus virtual_row_status = VirtualRowStatus::NoConsiderInLogicalPlan; + ExpressionActionsPtr virtual_row_conversion; std::optional number_of_current_replica; }; diff --git a/src/Processors/Transforms/MergeSortingTransform.cpp b/src/Processors/Transforms/MergeSortingTransform.cpp index c45192e7118..6121a847ca8 100644 --- a/src/Processors/Transforms/MergeSortingTransform.cpp +++ b/src/Processors/Transforms/MergeSortingTransform.cpp @@ -187,6 +187,7 @@ void MergeSortingTransform::consume(Chunk chunk) { bool have_all_inputs = false; bool use_average_block_sizes = false; + bool apply_virtual_row = false; external_merging_sorted = std::make_shared( header_without_constants, @@ -199,6 +200,7 @@ void MergeSortingTransform::consume(Chunk chunk) /*always_read_till_end_=*/ false, nullptr, use_average_block_sizes, + apply_virtual_row, have_all_inputs); processors.emplace_back(external_merging_sorted); diff --git a/src/Processors/Transforms/VirtualRowTransform.cpp b/src/Processors/Transforms/VirtualRowTransform.cpp index 92bf5ce3064..5f2bf0b0788 100644 --- a/src/Processors/Transforms/VirtualRowTransform.cpp +++ b/src/Processors/Transforms/VirtualRowTransform.cpp @@ -9,10 +9,11 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; } -VirtualRowTransform::VirtualRowTransform(const Block & header_, const Block & pk_block_) +VirtualRowTransform::VirtualRowTransform(const Block & header_, const Block & pk_block_, ExpressionActionsPtr virtual_row_conversions_) : IProcessor({header_}, {header_}) , input(inputs.front()), output(outputs.front()) - , header(header_), pk_block(pk_block_) + , pk_block(pk_block_) + , virtual_row_conversions(std::move(virtual_row_conversions_)) { } @@ -86,6 +87,7 @@ void VirtualRowTransform::work() is_first = false; Columns empty_columns; + const auto & header = getOutputs().front().getHeader(); empty_columns.reserve(header.columns()); for (size_t i = 0; i < header.columns(); ++i) { @@ -94,7 +96,7 @@ void VirtualRowTransform::work() } current_chunk.setColumns(empty_columns, 0); - current_chunk.getChunkInfos().add(std::make_shared(0, pk_block)); + current_chunk.getChunkInfos().add(std::make_shared(0, pk_block, virtual_row_conversions)); } else { diff --git a/src/Processors/Transforms/VirtualRowTransform.h b/src/Processors/Transforms/VirtualRowTransform.h index e3215393ad1..efc54419a6e 100644 --- a/src/Processors/Transforms/VirtualRowTransform.h +++ b/src/Processors/Transforms/VirtualRowTransform.h @@ -11,7 +11,7 @@ namespace DB class VirtualRowTransform : public IProcessor { public: - explicit VirtualRowTransform(const Block & header_, const Block & pk_block_); + explicit VirtualRowTransform(const Block & header_, const Block & pk_block_, ExpressionActionsPtr virtual_row_conversions_); String getName() const override { return "VirtualRowTransform"; } @@ -28,8 +28,8 @@ private: bool can_generate = true; bool is_first = true; - Block header; Block pk_block; + ExpressionActionsPtr virtual_row_conversions; }; } diff --git a/src/Storages/ReadInOrderOptimizer.cpp b/src/Storages/ReadInOrderOptimizer.cpp index ea7ea218feb..9c8c4c2fe79 100644 --- a/src/Storages/ReadInOrderOptimizer.cpp +++ b/src/Storages/ReadInOrderOptimizer.cpp @@ -249,7 +249,7 @@ InputOrderInfoPtr ReadInOrderOptimizer::getInputOrderImpl( if (sort_description_for_merging.empty()) return {}; - return std::make_shared(std::move(sort_description_for_merging), key_pos, read_direction, limit, false); + return std::make_shared(std::move(sort_description_for_merging), key_pos, read_direction, limit); } InputOrderInfoPtr ReadInOrderOptimizer::getInputOrder( diff --git a/src/Storages/SelectQueryInfo.h b/src/Storages/SelectQueryInfo.h index bf1229f7a3a..7ad6a733c6f 100644 --- a/src/Storages/SelectQueryInfo.h +++ b/src/Storages/SelectQueryInfo.h @@ -119,22 +119,13 @@ struct InputOrderInfo const int direction; const UInt64 limit; - /** For virtual row optimization only - * for example, when pk is (a,b), a = 1, order by b, virtual row should be - * disabled in the following case: - * 1st part (0, 100), (1, 2), (1, 3), (1, 4) - * 2nd part (0, 100), (1, 2), (1, 3), (1, 4). - */ - bool first_prefix_fixed; - InputOrderInfo( const SortDescription & sort_description_for_merging_, size_t used_prefix_of_sorting_key_size_, - int direction_, UInt64 limit_, bool first_prefix_fixed_) + int direction_, UInt64 limit_) : sort_description_for_merging(sort_description_for_merging_) , used_prefix_of_sorting_key_size(used_prefix_of_sorting_key_size_) , direction(direction_), limit(limit_) - , first_prefix_fixed(first_prefix_fixed_) { } diff --git a/src/Storages/StorageMerge.cpp b/src/Storages/StorageMerge.cpp index f40aa8ae4e8..40713a89f30 100644 --- a/src/Storages/StorageMerge.cpp +++ b/src/Storages/StorageMerge.cpp @@ -1555,7 +1555,7 @@ bool ReadFromMerge::requestReadingInOrder(InputOrderInfoPtr order_info_) auto request_read_in_order = [order_info_](ReadFromMergeTree & read_from_merge_tree) { return read_from_merge_tree.requestReadingInOrder( - order_info_->used_prefix_of_sorting_key_size, order_info_->direction, order_info_->limit); + order_info_->used_prefix_of_sorting_key_size, order_info_->direction, order_info_->limit, {}); }; bool ok = true; From 35cf3e8b91ce20bcfd6218d8d34a5f0a96fdd03e Mon Sep 17 00:00:00 2001 From: vdimir Date: Thu, 26 Sep 2024 13:34:25 +0000 Subject: [PATCH 063/680] fix stylecheck Signed-off-by: vdimir --- src/Planner/PlannerJoinTree.cpp | 5 ++--- tests/integration/helpers/random_settings.py | 4 +++- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp index 1ee0024f053..543dc1a88f6 100644 --- a/src/Planner/PlannerJoinTree.cpp +++ b/src/Planner/PlannerJoinTree.cpp @@ -104,7 +104,7 @@ namespace Setting extern const SettingsBool optimize_move_to_prewhere; extern const SettingsBool optimize_move_to_prewhere_if_final; extern const SettingsBool use_concurrency_control; - extern const SettingsBool query_plan_join_inner_table_selection; + extern const SettingsJoinInnerTableSelectionMode query_plan_join_inner_table_selection; } namespace ErrorCodes @@ -1642,8 +1642,7 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ settings[Setting::max_block_size], settings[Setting::max_threads], false /*optimize_read_in_order*/); - if (settings[Setting::query_plan_join_inner_table_selection]) - join_step->inner_table_selection_mode = JoinInnerTableSelectionMode::Auto; + join_step->inner_table_selection_mode = settings[Setting::query_plan_join_inner_table_selection]; join_step->setStepDescription(fmt::format("JOIN {}", join_pipeline_type)); diff --git a/tests/integration/helpers/random_settings.py b/tests/integration/helpers/random_settings.py index 49498b9f778..a34d8e93c47 100644 --- a/tests/integration/helpers/random_settings.py +++ b/tests/integration/helpers/random_settings.py @@ -6,7 +6,9 @@ def randomize_settings(): if random.random() < 0.5: yield "max_block_size", random.randint(8000, 100000) if random.random() < 0.5: - yield "query_plan_join_inner_table_selection", random.choice(["auto", "left", "right"]) + yield "query_plan_join_inner_table_selection", random.choice( + ["auto", "left", "right"] + ) def write_random_settings_config(destination): From 3ee6fd9b059d4afc67bc154c90746a1e0ec51bd9 Mon Sep 17 00:00:00 2001 From: vdimir Date: Thu, 26 Sep 2024 14:19:20 +0000 Subject: [PATCH 064/680] Fix header --- src/Processors/QueryPlan/JoinStep.cpp | 5 +++-- src/Processors/QueryPlan/JoinStep.h | 2 ++ 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index 0e9332c186e..fefb193827f 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -111,7 +111,7 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines if (join->pipelineType() == JoinPipelineType::YShaped) { auto joined_pipeline = QueryPipelineBuilder::joinPipelinesYShaped( - std::move(pipelines[0]), std::move(pipelines[1]), join, output_stream->header, max_block_size, &processors); + std::move(pipelines[0]), std::move(pipelines[1]), join, join_algorithm_header, max_block_size, &processors); joined_pipeline->resize(max_streams); return joined_pipeline; } @@ -120,7 +120,7 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines std::move(pipelines[0]), std::move(pipelines[1]), join, - output_stream->header, + join_algorithm_header, max_block_size, max_streams, keep_left_read_in_order, @@ -170,6 +170,7 @@ void JoinStep::updateOutputStream() Block result_header = JoiningTransform::transformHeader(header, join); + join_algorithm_header = result_header; if (swap_streams) result_header = rotateBlock(result_header, input_streams[1].header); diff --git a/src/Processors/QueryPlan/JoinStep.h b/src/Processors/QueryPlan/JoinStep.h index 46fb49947ba..96c02f9fd19 100644 --- a/src/Processors/QueryPlan/JoinStep.h +++ b/src/Processors/QueryPlan/JoinStep.h @@ -42,6 +42,8 @@ public: private: void updateOutputStream() override; + /// Header that expected to be returned from IJoin + Block join_algorithm_header; JoinPtr join; size_t max_block_size; From fca592a31fc0c233fb971deff211e1ce7c040cfa Mon Sep 17 00:00:00 2001 From: vdimir Date: Thu, 26 Sep 2024 14:20:45 +0000 Subject: [PATCH 065/680] fix stylecheck --- tests/clickhouse-test | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 1c606bea228..06a044eef32 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -788,6 +788,7 @@ def threshold_generator(always_on_prob, always_off_prob, min_val, max_val): def get_localzone(): return os.getenv("TZ", "/".join(os.readlink("/etc/localtime").split("/")[-2:])) + # Refer to `tests/integration/helpers/random_settings.py` for integration test random settings class SettingsRandomizer: settings = { @@ -2154,9 +2155,9 @@ class TestSuite: ) ) self.all_tags: Dict[str, Set[str]] = all_tags_and_random_settings_limits[0] - self.all_random_settings_limits: Dict[str, Dict[str, (int, int)]] = ( - all_tags_and_random_settings_limits[1] - ) + self.all_random_settings_limits: Dict[ + str, Dict[str, (int, int)] + ] = all_tags_and_random_settings_limits[1] self.sequential_tests = [] self.parallel_tests = [] for test_name in self.all_tests: From d39d9a876537c603626a6fbe478d32e0a208275b Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 26 Sep 2024 14:30:06 +0000 Subject: [PATCH 066/680] Automatic style fix --- tests/clickhouse-test | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 06a044eef32..c1615d039cb 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -2155,9 +2155,9 @@ class TestSuite: ) ) self.all_tags: Dict[str, Set[str]] = all_tags_and_random_settings_limits[0] - self.all_random_settings_limits: Dict[ - str, Dict[str, (int, int)] - ] = all_tags_and_random_settings_limits[1] + self.all_random_settings_limits: Dict[str, Dict[str, (int, int)]] = ( + all_tags_and_random_settings_limits[1] + ) self.sequential_tests = [] self.parallel_tests = [] for test_name in self.all_tests: From 9642e6cdcc4d2e700efa43ebbf523ed25f728cd8 Mon Sep 17 00:00:00 2001 From: vdimir Date: Thu, 26 Sep 2024 15:01:59 +0000 Subject: [PATCH 067/680] add ColumnPermuteTransform --- .../Transforms/ColumnPermuteTransform.cpp | 49 +++++++++++++++++++ .../Transforms/ColumnPermuteTransform.h | 28 +++++++++++ 2 files changed, 77 insertions(+) create mode 100644 src/Processors/Transforms/ColumnPermuteTransform.cpp create mode 100644 src/Processors/Transforms/ColumnPermuteTransform.h diff --git a/src/Processors/Transforms/ColumnPermuteTransform.cpp b/src/Processors/Transforms/ColumnPermuteTransform.cpp new file mode 100644 index 00000000000..ac7793bd136 --- /dev/null +++ b/src/Processors/Transforms/ColumnPermuteTransform.cpp @@ -0,0 +1,49 @@ +#include + +namespace DB +{ + +namespace +{ + +template +void applyPermutation(std::vector & data, const std::vector & permutation) +{ + std::vector res; + res.reserve(data.size()); + for (size_t i = 0; i < data.size(); ++i) + res.emplace_back(std::move(data[permutation[i]])); + data = std::move(res); +} + +Block permuteBlock(const Block & block, const std::vector & permutation) +{ + auto columns = block.getColumnsWithTypeAndName(); + applyPermutation(columns, permutation); + return Block(columns); +} + +void permuteChunk(Chunk & chunk, const std::vector & permutation) +{ + size_t num_rows = chunk.getNumRows(); + auto columns = chunk.detachColumns(); + applyPermutation(columns, permutation); + chunk.setColumns(std::move(columns), num_rows); +} + +} + +ColumnPermuteTransform::ColumnPermuteTransform(const Block & header_, std::vector permutation_) + : ISimpleTransform(header_, permuteBlock(header_, permutation_), false) + , permutation(std::move(permutation_)) +{ +} + + +void ColumnPermuteTransform::transform(Chunk & chunk) +{ + permuteChunk(chunk, permutation); +} + + +} diff --git a/src/Processors/Transforms/ColumnPermuteTransform.h b/src/Processors/Transforms/ColumnPermuteTransform.h new file mode 100644 index 00000000000..b2e3c469833 --- /dev/null +++ b/src/Processors/Transforms/ColumnPermuteTransform.h @@ -0,0 +1,28 @@ +#pragma once + +#include +#include +#include +#include +#include +#include + +namespace DB +{ + +class ColumnPermuteTransform : public ISimpleTransform +{ +public: + ColumnPermuteTransform(const Block & header_, std::vector permutation_); + + String getName() const override { return "ColumnPermuteTransform"; } + + void transform(Chunk & chunk) override; + +private: + Names column_names; + std::vector permutation; +}; + + +} From 7feda9a05413fedf681d3fe1e229bf9e5ab434ef Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Thu, 26 Sep 2024 15:27:57 +0000 Subject: [PATCH 068/680] Fix 03031_read_in_order_optimization_with_virtual_row --- .../Merges/Algorithms/MergeTreeReadInfo.h | 51 +++++++++++++------ .../Algorithms/MergingSortedAlgorithm.cpp | 3 +- .../Optimizations/optimizeReadInOrder.cpp | 24 ++++++--- .../QueryPlan/ReadFromMergeTree.cpp | 5 +- 4 files changed, 57 insertions(+), 26 deletions(-) diff --git a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h index 62bbe3eac6e..a4baaca215b 100644 --- a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h +++ b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h @@ -7,6 +7,11 @@ namespace DB { +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + /// To carry part level and virtual row if chunk is produced by a merge tree source class MergeTreeReadInfo : public ChunkInfoCloneable { @@ -41,33 +46,49 @@ inline bool isVirtualRow(const Chunk & chunk) return false; } -inline void setVirtualRow(Chunk & chunk, bool apply_virtual_row_conversions) +inline void setVirtualRow(Chunk & chunk, const Block & header, bool apply_virtual_row_conversions) { - auto read_info = chunk.getChunkInfos().extract(); + auto read_info = chunk.getChunkInfos().get(); chassert(read_info); Block & pk_block = read_info->pk_block; + + // std::cerr << apply_virtual_row_conversions << std::endl; + // std::cerr << read_info->virtual_row_conversions->dumpActions() << std::endl; + if (apply_virtual_row_conversions) read_info->virtual_row_conversions->execute(pk_block); - chunk.setColumns(pk_block.getColumns(), 1); + // std::cerr << "++++" << pk_block.dumpStructure() << std::endl; - // Columns ordered_columns; - // ordered_columns.reserve(pk_block.columns()); + Columns ordered_columns; + ordered_columns.reserve(pk_block.columns()); - // for (size_t i = 0; i < header.columns(); ++i) - // { - // const ColumnWithTypeAndName & type_and_name = header.getByPosition(i); - // ColumnPtr current_column = type_and_name.type->createColumn(); + for (size_t i = 0; i < header.columns(); ++i) + { + const ColumnWithTypeAndName & col = header.getByPosition(i); + if (const auto * pk_col = pk_block.findByName(col.name)) + { + if (!col.type->equals(*pk_col->type)) + throw Exception(ErrorCodes::LOGICAL_ERROR, + "Virtual row has different tupe for {}. Expected {}, got {}", + col.name, col.dumpStructure(), pk_col->dumpStructure()); - // size_t pos = type_and_name.name.find_last_of('.'); - // String column_name = (pos == String::npos) ? type_and_name.name : type_and_name.name.substr(pos + 1); + ordered_columns.push_back(pk_col->column); + } + else + ordered_columns.push_back(col.type->createColumnConstWithDefaultValue(1)); - // const ColumnWithTypeAndName * column = pk_block.findByName(column_name, true); - // ordered_columns.push_back(column ? column->column : current_column->cloneResized(1)); - // } + // ColumnPtr current_column = type_and_name.type->createColumn(); - // chunk.setColumns(ordered_columns, 1); + // size_t pos = type_and_name.name.find_last_of('.'); + // String column_name = (pos == String::npos) ? type_and_name.name : type_and_name.name.substr(pos + 1); + + // const ColumnWithTypeAndName * column = pk_block.findByName(column_name, true); + // ordered_columns.push_back(column ? column->column : current_column->cloneResized(1)); + } + + chunk.setColumns(ordered_columns, 1); } } diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp index 0dd95729ba3..011f713744b 100644 --- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp @@ -62,8 +62,7 @@ void MergingSortedAlgorithm::initialize(Inputs inputs) if (!isVirtualRow(input.chunk)) continue; - setVirtualRow(input.chunk, apply_virtual_row_conversions); - input.skip_last_row = true; + setVirtualRow(input.chunk, header, apply_virtual_row_conversions); } removeConstAndSparse(inputs); diff --git a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp index 49ce9a0280d..5396cced6c1 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp @@ -339,20 +339,20 @@ void enrichFixedColumns(const ActionsDAG & dag, FixedColumns & fixed_columns) } } -static const ActionsDAG::Node * addMonotonicChain(ActionsDAG & dag, const ActionsDAG::Node * node, const MatchedTrees::Match * match) +static const ActionsDAG::Node * addMonotonicChain(ActionsDAG & dag, const ActionsDAG::Node * node, const MatchedTrees::Match * match, const std::string & input_name) { if (!match->monotonicity) - return &dag.addInput(node->result_name, node->result_type); + return &dag.addInput(input_name, node->result_type); if (node->type == ActionsDAG::ActionType::ALIAS) - return &dag.addAlias(*addMonotonicChain(dag, node->children.front(), match), node->result_name); + return &dag.addAlias(*addMonotonicChain(dag, node->children.front(), match, input_name), node->result_name); ActionsDAG::NodeRawConstPtrs args; args.reserve(node->children.size()); for (const auto * child : node->children) { if (child == match->monotonicity->child_node) - args.push_back(addMonotonicChain(dag, match->monotonicity->child_node, match->monotonicity->child_match)); + args.push_back(addMonotonicChain(dag, match->monotonicity->child_node, match->monotonicity->child_match, input_name)); else args.push_back(&dag.addColumn({child->column, child->result_type, child->result_name})); } @@ -571,15 +571,25 @@ SortingInputOrder buildInputOrderInfo( { ActionsDAG virtual_row_dag; virtual_row_dag.getOutputs().reserve(match_infos.size()); + size_t next_pk_name = 0; for (const auto & info : match_infos) { const ActionsDAG::Node * output; if (info.fixed_column) output = &virtual_row_dag.addColumn({info.fixed_column->column, info.fixed_column->result_type, info.fixed_column->result_name}); - else if (info.monotonic) - output = addMonotonicChain(virtual_row_dag, info.source, info.monotonic); else - output = &virtual_row_dag.addInput(info.source->result_name, info.source->result_type); + { + if (info.monotonic) + output = addMonotonicChain(virtual_row_dag, info.source, info.monotonic, pk_column_names[next_pk_name]); + else + { + output = &virtual_row_dag.addInput(pk_column_names[next_pk_name], info.source->result_type); + if (pk_column_names[next_pk_name] != info.source->result_name) + output = &virtual_row_dag.addAlias(*output, info.source->result_name); + } + + ++next_pk_name; + } virtual_row_dag.getOutputs().push_back(output); } diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 38f018d34ee..c6fc924d7a7 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -701,9 +701,10 @@ Pipe ReadFromMergeTree::readInOrder( size_t mark_range_begin = part_with_ranges.ranges.front().begin; ColumnsWithTypeAndName pk_columns; - pk_columns.reserve(index->size()); + size_t num_columns = virtual_row_conversion->getSampleBlock().columns(); + pk_columns.reserve(num_columns); - for (size_t j = 0; j < index->size(); ++j) + for (size_t j = 0; j < num_columns; ++j) { auto column = primary_key.data_types[j]->createColumn()->cloneEmpty(); column->insert((*(*index)[j])[mark_range_begin]); From d5c0c499df1c80d94c3248394f8d8e271003d8fc Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Thu, 26 Sep 2024 16:01:47 +0000 Subject: [PATCH 069/680] Fix PK size. --- src/Processors/QueryPlan/ReadFromMergeTree.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index c6fc924d7a7..b16a460ec68 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -701,7 +701,7 @@ Pipe ReadFromMergeTree::readInOrder( size_t mark_range_begin = part_with_ranges.ranges.front().begin; ColumnsWithTypeAndName pk_columns; - size_t num_columns = virtual_row_conversion->getSampleBlock().columns(); + size_t num_columns = virtual_row_conversion->getRequiredColumnsWithTypes().size(); pk_columns.reserve(num_columns); for (size_t j = 0; j < num_columns; ++j) From d6b444dac9328ea0b64fda1005f78e2164fbab1b Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Thu, 26 Sep 2024 16:12:18 +0000 Subject: [PATCH 070/680] Skip virtual row chunk by skipping last row. --- .../Algorithms/MergingSortedAlgorithm.cpp | 21 ++++++++++--------- 1 file changed, 11 insertions(+), 10 deletions(-) diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp index 011f713744b..331b67066be 100644 --- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp @@ -63,6 +63,7 @@ void MergingSortedAlgorithm::initialize(Inputs inputs) continue; setVirtualRow(input.chunk, header, apply_virtual_row_conversions); + input.skip_last_row = true; } removeConstAndSparse(inputs); @@ -149,8 +150,8 @@ IMergingAlgorithm::Status MergingSortedAlgorithm::mergeImpl(TSortingHeap & queue auto current = queue.current(); - if (isVirtualRow(current_inputs[current.impl->order].chunk)) - throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Virtual row is not implemented for Non-batch mode."); + // if (isVirtualRow(current_inputs[current.impl->order].chunk)) + // throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Virtual row is not implemented for Non-batch mode."); if (current.impl->isLast() && current_inputs[current.impl->order].skip_last_row) { @@ -248,14 +249,14 @@ IMergingAlgorithm::Status MergingSortedAlgorithm::mergeBatchImpl(TSortingQueue & auto [current_ptr, initial_batch_size] = queue.current(); auto current = *current_ptr; - if (isVirtualRow(current_inputs[current.impl->order].chunk)) - { - /// If virtual row is detected, there should be only one row as a single chunk, - /// and always skip this chunk to pull the next one. - chassert(initial_batch_size == 1); - queue.removeTop(); - return Status(current.impl->order); - } + // if (isVirtualRow(current_inputs[current.impl->order].chunk)) + // { + // /// If virtual row is detected, there should be only one row as a single chunk, + // /// and always skip this chunk to pull the next one. + // chassert(initial_batch_size == 1); + // queue.removeTop(); + // return Status(current.impl->order); + // } bool batch_skip_last_row = false; if (current.impl->isLast(initial_batch_size) && current_inputs[current.impl->order].skip_last_row) From fb0b46adbf40dfbf7feaf36796c88d8e82da6633 Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Fri, 27 Sep 2024 09:24:54 +0000 Subject: [PATCH 071/680] DIsable virtual row for FINAL. --- src/Processors/QueryPlan/ReadFromMergeTree.cpp | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index b16a460ec68..37a159dd865 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -1890,7 +1890,8 @@ bool ReadFromMergeTree::requestReadingInOrder(size_t prefix_size, int direction, /// Let prefer in-order optimization over vertical FINAL for now enable_vertical_final = false; - if (virtual_row_conversion_) + /// Disable virtual row for FINAL. + if (virtual_row_conversion_ && !isQueryWithFinal()) virtual_row_conversion = std::make_shared(std::move(*virtual_row_conversion_)); return true; From 63c89ded04c36c572e9280c356a4bf5570c65bf7 Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Fri, 27 Sep 2024 10:56:28 +0000 Subject: [PATCH 072/680] Fixing other tests. --- .../Optimizations/optimizeReadInOrder.cpp | 2 +- .../01786_explain_merge_tree.reference | 14 ++++++++++++++ .../02149_read_in_order_fixed_prefix.reference | 16 ++++++---------- 3 files changed, 21 insertions(+), 11 deletions(-) diff --git a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp index 5396cced6c1..cac4cf69054 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp @@ -515,7 +515,7 @@ SortingInputOrder buildInputOrderInfo( // 1st part (0, 100), (1, 2), (1, 3), (1, 4) // 2nd part (0, 100), (1, 2), (1, 3), (1, 4). - can_optimize_virtual_row = true; + can_optimize_virtual_row = false; } //std::cerr << "+++++++++ Found fixed key by match" << std::endl; diff --git a/tests/queries/0_stateless/01786_explain_merge_tree.reference b/tests/queries/0_stateless/01786_explain_merge_tree.reference index 3a015d32539..f02dbcb59c9 100644 --- a/tests/queries/0_stateless/01786_explain_merge_tree.reference +++ b/tests/queries/0_stateless/01786_explain_merge_tree.reference @@ -86,11 +86,17 @@ ReadType: InOrder Parts: 1 Granules: 3 + Virtual row conversions + Actions: INPUT :: 0 -> x UInt32 : 0 + Positions: 0 ----------------- ReadFromMergeTree (default.test_index) ReadType: InReverseOrder Parts: 1 Granules: 3 + Virtual row conversions + Actions: INPUT :: 0 -> x UInt32 : 0 + Positions: 0 ReadFromMergeTree (default.idx) Indexes: PrimaryKey @@ -174,11 +180,19 @@ ReadType: InOrder Parts: 1 Granules: 3 + Virtual row conversions + Actions: INPUT : 0 -> x UInt32 : 0 + ALIAS x :: 0 -> __table1.x UInt32 : 1 + Positions: 1 ----------------- ReadFromMergeTree (default.test_index) ReadType: InReverseOrder Parts: 1 Granules: 3 + Virtual row conversions + Actions: INPUT : 0 -> x UInt32 : 0 + ALIAS x :: 0 -> __table1.x UInt32 : 1 + Positions: 1 ReadFromMergeTree (default.idx) Indexes: PrimaryKey diff --git a/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference b/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference index 31462988c2d..cb96a7167da 100644 --- a/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference +++ b/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.reference @@ -14,10 +14,7 @@ ExpressionTransform (Expression) ExpressionTransform × 2 (ReadFromMergeTree) - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 2020-10-01 9 2020-10-01 9 2020-10-01 9 @@ -54,10 +51,7 @@ ExpressionTransform (Expression) ExpressionTransform × 2 (ReadFromMergeTree) - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 2020-10-11 0 2020-10-11 0 2020-10-11 0 @@ -178,7 +172,8 @@ ExpressionTransform (Expression) ExpressionTransform (ReadFromMergeTree) - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 2020-10-10 00:00:00 0.01 2020-10-10 00:00:00 0.01 2020-10-10 00:00:00 0.01 @@ -192,7 +187,8 @@ ExpressionTransform (Expression) ExpressionTransform (ReadFromMergeTree) - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + VirtualRowTransform + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 2020-10-10 00:00:00 0.01 2020-10-10 00:00:00 0.01 2020-10-10 00:00:00 0.01 From 3c8594d401d7c625a810a61776e689083d79912a Mon Sep 17 00:00:00 2001 From: divanik Date: Fri, 27 Sep 2024 14:30:07 +0000 Subject: [PATCH 073/680] Remove unnecessary changes --- .../DataLakes/DataLakeConfiguration.h | 86 +++++++++ .../DataLakes/DeltaLakeMetadata.cpp | 40 ++-- .../DataLakes/DeltaLakeMetadata.h | 12 +- .../ObjectStorage/DataLakes/HudiMetadata.cpp | 12 +- .../ObjectStorage/DataLakes/HudiMetadata.h | 8 +- .../DataLakes/IStorageDataLake.h | 172 ------------------ .../DataLakes/IcebergMetadata.cpp | 24 +-- .../ObjectStorage/DataLakes/IcebergMetadata.h | 8 +- .../DataLakes/registerDataLakeStorages.cpp | 132 -------------- .../ObjectStorage/StorageObjectStorage.cpp | 18 +- .../ObjectStorage/StorageObjectStorage.h | 21 ++- .../registerStorageObjectStorage.cpp | 105 +++++++++++ src/TableFunctions/ITableFunctionDataLake.h | 120 ------------ .../TableFunctionObjectStorage.cpp | 90 +++++++++ .../TableFunctionObjectStorage.h | 55 ++++++ .../registerDataLakeTableFunctions.cpp | 88 --------- 16 files changed, 407 insertions(+), 584 deletions(-) create mode 100644 src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h delete mode 100644 src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h delete mode 100644 src/Storages/ObjectStorage/DataLakes/registerDataLakeStorages.cpp delete mode 100644 src/TableFunctions/ITableFunctionDataLake.h delete mode 100644 src/TableFunctions/registerDataLakeTableFunctions.cpp diff --git a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h new file mode 100644 index 00000000000..6d8e64aa3b7 --- /dev/null +++ b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h @@ -0,0 +1,86 @@ +#pragma once + +#include "config.h" + +#if USE_AVRO + +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include +# include + +# include + + +namespace DB +{ + +template +concept StorageConfiguration = std::derived_from; + +template +class DataLakeConfiguration : public BaseStorageConfiguration, public std::enable_shared_from_this +{ +public: + using Configuration = StorageObjectStorage::Configuration; + + bool isDataLakeConfiguration() const override { return true; } + + std::string getEngineName() const override { return DataLakeMetadata::name; } + + void update(ObjectStoragePtr object_storage, ContextPtr local_context) override + { + auto new_metadata = DataLakeMetadata::create(object_storage, weak_from_this(), local_context); + if (current_metadata && *current_metadata == *new_metadata) + return; + + current_metadata = std::move(new_metadata); + BaseStorageConfiguration::setPaths(current_metadata->getDataFiles()); + BaseStorageConfiguration::setPartitionColumns(current_metadata->getPartitionColumns()); + } + +private: + DataLakeMetadataPtr current_metadata; + + ReadFromFormatInfo prepareReadingFromFormat( + ObjectStoragePtr object_storage, + const Strings & requested_columns, + const StorageSnapshotPtr & storage_snapshot, + bool supports_subset_of_columns, + ContextPtr local_context) override + { + auto info = DB::prepareReadingFromFormat(requested_columns, storage_snapshot, supports_subset_of_columns); + if (!current_metadata) + { + current_metadata = DataLakeMetadata::create(object_storage, weak_from_this(), local_context); + } + auto column_mapping = current_metadata->getColumnNameToPhysicalNameMapping(); + if (!column_mapping.empty()) + { + for (const auto & [column_name, physical_name] : column_mapping) + { + auto & column = info.format_header.getByName(column_name); + column.name = physical_name; + } + } + return info; + } +}; + +using StorageS3IcebergConfiguration = DataLakeConfiguration; +using StorageAzureIcebergConfiguration = DataLakeConfiguration; +using StorageLocalIcebergConfiguration = DataLakeConfiguration; +using StorageS3DeltaLakeConfiguration = DataLakeConfiguration; +using StorageS3HudiConfiguration = DataLakeConfiguration; + + +} + +#endif diff --git a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp index f04e868ee5a..f437faa2e90 100644 --- a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp +++ b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.cpp @@ -55,22 +55,18 @@ namespace ErrorCodes struct DeltaLakeMetadataImpl { - using ConfigurationPtr = DeltaLakeMetadata::ConfigurationPtr; + using ConfigurationObservePtr = DeltaLakeMetadata::ConfigurationObservePtr; ObjectStoragePtr object_storage; - ConfigurationPtr configuration; + ConfigurationObservePtr configuration; ContextPtr context; /** * Useful links: * - https://github.com/delta-io/delta/blob/master/PROTOCOL.md#data-files */ - DeltaLakeMetadataImpl(ObjectStoragePtr object_storage_, - ConfigurationPtr configuration_, - ContextPtr context_) - : object_storage(object_storage_) - , configuration(configuration_) - , context(context_) + DeltaLakeMetadataImpl(ObjectStoragePtr object_storage_, ConfigurationObservePtr configuration_, ContextPtr context_) + : object_storage(object_storage_), configuration(configuration_), context(context_) { } @@ -110,6 +106,7 @@ struct DeltaLakeMetadataImpl }; DeltaLakeMetadata processMetadataFiles() { + auto configuration_ptr = configuration.lock(); std::set result_files; NamesAndTypesList current_schema; DataLakePartitionColumns current_partition_columns; @@ -121,7 +118,7 @@ struct DeltaLakeMetadataImpl while (true) { const auto filename = withPadding(++current_version) + metadata_file_suffix; - const auto file_path = std::filesystem::path(configuration->getPath()) / deltalake_metadata_directory / filename; + const auto file_path = std::filesystem::path(configuration_ptr->getPath()) / deltalake_metadata_directory / filename; if (!object_storage->exists(StoredObject(file_path))) break; @@ -135,7 +132,7 @@ struct DeltaLakeMetadataImpl } else { - const auto keys = listFiles(*object_storage, *configuration, deltalake_metadata_directory, metadata_file_suffix); + const auto keys = listFiles(*object_storage, *configuration_ptr, deltalake_metadata_directory, metadata_file_suffix); for (const String & key : keys) processMetadataFile(key, current_schema, current_partition_columns, result_files); } @@ -244,6 +241,8 @@ struct DeltaLakeMetadataImpl } } + auto configuration_ptr = configuration.lock(); + if (object->has("add")) { auto add_object = object->get("add").extract(); @@ -251,7 +250,7 @@ struct DeltaLakeMetadataImpl throw Exception(ErrorCodes::LOGICAL_ERROR, "Failed to extract `add` field"); auto path = add_object->getValue("path"); - result.insert(fs::path(configuration->getPath()) / path); + result.insert(fs::path(configuration_ptr->getPath()) / path); auto filename = fs::path(path).filename().string(); auto it = file_partition_columns.find(filename); @@ -295,7 +294,7 @@ struct DeltaLakeMetadataImpl throw Exception(ErrorCodes::LOGICAL_ERROR, "Failed to extract `remove` field"); auto path = remove_object->getValue("path"); - result.erase(fs::path(configuration->getPath()) / path); + result.erase(fs::path(configuration_ptr->getPath()) / path); } } } @@ -486,7 +485,9 @@ struct DeltaLakeMetadataImpl */ size_t readLastCheckpointIfExists() const { - const auto last_checkpoint_file = std::filesystem::path(configuration->getPath()) / deltalake_metadata_directory / "_last_checkpoint"; + auto configuration_ptr = configuration.lock(); + const auto last_checkpoint_file + = std::filesystem::path(configuration_ptr->getPath()) / deltalake_metadata_directory / "_last_checkpoint"; if (!object_storage->exists(StoredObject(last_checkpoint_file))) return 0; @@ -552,7 +553,11 @@ struct DeltaLakeMetadataImpl return 0; const auto checkpoint_filename = withPadding(version) + ".checkpoint.parquet"; - const auto checkpoint_path = std::filesystem::path(configuration->getPath()) / deltalake_metadata_directory / checkpoint_filename; + + auto configuration_ptr = configuration.lock(); + + const auto checkpoint_path + = std::filesystem::path(configuration_ptr->getPath()) / deltalake_metadata_directory / checkpoint_filename; LOG_TRACE(log, "Using checkpoint file: {}", checkpoint_path.string()); @@ -667,7 +672,7 @@ struct DeltaLakeMetadataImpl } LOG_TEST(log, "Adding {}", path); - const auto [_, inserted] = result.insert(std::filesystem::path(configuration->getPath()) / path); + const auto [_, inserted] = result.insert(std::filesystem::path(configuration_ptr->getPath()) / path); if (!inserted) throw Exception(ErrorCodes::INCORRECT_DATA, "File already exists {}", path); } @@ -678,10 +683,7 @@ struct DeltaLakeMetadataImpl LoggerPtr log = getLogger("DeltaLakeMetadataParser"); }; -DeltaLakeMetadata::DeltaLakeMetadata( - ObjectStoragePtr object_storage_, - ConfigurationPtr configuration_, - ContextPtr context_) +DeltaLakeMetadata::DeltaLakeMetadata(ObjectStoragePtr object_storage_, ConfigurationObservePtr configuration_, ContextPtr context_) { auto impl = DeltaLakeMetadataImpl(object_storage_, configuration_, context_); auto result = impl.processMetadataFiles(); diff --git a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h index a479a3dd293..549443f115e 100644 --- a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h +++ b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h @@ -12,13 +12,10 @@ namespace DB class DeltaLakeMetadata final : public IDataLakeMetadata { public: - using ConfigurationPtr = StorageObjectStorage::ConfigurationPtr; + using ConfigurationObservePtr = StorageObjectStorage::ConfigurationObservePtr; static constexpr auto name = "DeltaLake"; - DeltaLakeMetadata( - ObjectStoragePtr object_storage_, - ConfigurationPtr configuration_, - ContextPtr context_); + DeltaLakeMetadata(ObjectStoragePtr object_storage_, ConfigurationObservePtr configuration_, ContextPtr context_); Strings getDataFiles() const override { return data_files; } @@ -36,10 +33,7 @@ public: && data_files == deltalake_metadata->data_files; } - static DataLakeMetadataPtr create( - ObjectStoragePtr object_storage, - ConfigurationPtr configuration, - ContextPtr local_context) + static DataLakeMetadataPtr create(ObjectStoragePtr object_storage, ConfigurationObservePtr configuration, ContextPtr local_context) { return std::make_unique(object_storage, configuration, local_context); } diff --git a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp index 91a586ccbf9..8a93a0ea6d3 100644 --- a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp +++ b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp @@ -43,8 +43,9 @@ namespace ErrorCodes */ Strings HudiMetadata::getDataFilesImpl() const { + auto configuration_ptr = configuration.lock(); auto log = getLogger("HudiMetadata"); - const auto keys = listFiles(*object_storage, *configuration, "", Poco::toLower(configuration->format)); + const auto keys = listFiles(*object_storage, *configuration_ptr, "", Poco::toLower(configuration_ptr->format)); using Partition = std::string; using FileID = std::string; @@ -86,13 +87,8 @@ Strings HudiMetadata::getDataFilesImpl() const return result; } -HudiMetadata::HudiMetadata( - ObjectStoragePtr object_storage_, - ConfigurationPtr configuration_, - ContextPtr context_) - : WithContext(context_) - , object_storage(object_storage_) - , configuration(configuration_) +HudiMetadata::HudiMetadata(ObjectStoragePtr object_storage_, ConfigurationObservePtr configuration_, ContextPtr context_) + : WithContext(context_), object_storage(object_storage_), configuration(configuration_) { } diff --git a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h index b060b1b0d39..b22dfacb0ad 100644 --- a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h +++ b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h @@ -13,13 +13,13 @@ namespace DB class HudiMetadata final : public IDataLakeMetadata, private WithContext { public: - using ConfigurationPtr = StorageObjectStorage::ConfigurationPtr; + using ConfigurationObservePtr = StorageObjectStorage::ConfigurationObservePtr; static constexpr auto name = "Hudi"; HudiMetadata( ObjectStoragePtr object_storage_, - ConfigurationPtr configuration_, + ConfigurationObservePtr configuration_, ContextPtr context_); Strings getDataFiles() const override; @@ -40,7 +40,7 @@ public: static DataLakeMetadataPtr create( ObjectStoragePtr object_storage, - ConfigurationPtr configuration, + ConfigurationObservePtr configuration, ContextPtr local_context) { return std::make_unique(object_storage, configuration, local_context); @@ -48,7 +48,7 @@ public: private: const ObjectStoragePtr object_storage; - const ConfigurationPtr configuration; + const ConfigurationObservePtr configuration; mutable Strings data_files; std::unordered_map column_name_to_physical_name; DataLakePartitionColumns partition_columns; diff --git a/src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h b/src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h deleted file mode 100644 index a17fd163253..00000000000 --- a/src/Storages/ObjectStorage/DataLakes/IStorageDataLake.h +++ /dev/null @@ -1,172 +0,0 @@ -#pragma once - -#include "config.h" - -#if USE_AVRO - -#include -#include -#include -#include -#include -#include -#include -#include - - -namespace DB -{ - -/// Storage for read-only integration with Apache Iceberg tables in Amazon S3 (see https://iceberg.apache.org/) -/// Right now it's implemented on top of StorageS3 and right now it doesn't support -/// many Iceberg features like schema evolution, partitioning, positional and equality deletes. -template -class IStorageDataLake final : public StorageObjectStorage -{ -public: - using Storage = StorageObjectStorage; - using ConfigurationPtr = Storage::ConfigurationPtr; - - static StoragePtr create( - ConfigurationPtr base_configuration, - ContextPtr context, - const StorageID & table_id_, - const ColumnsDescription & columns_, - const ConstraintsDescription & constraints_, - const String & comment_, - std::optional format_settings_, - LoadingStrictnessLevel mode) - { - auto object_storage = base_configuration->createObjectStorage(context, /* is_readonly */true); - DataLakeMetadataPtr metadata; - NamesAndTypesList schema_from_metadata; - const bool use_schema_from_metadata = columns_.empty(); - - if (base_configuration->format == "auto") - base_configuration->format = "Parquet"; - - ConfigurationPtr configuration = base_configuration->clone(); - - try - { - metadata = DataLakeMetadata::create(object_storage, base_configuration, context); - configuration->setPaths(metadata->getDataFiles()); - if (use_schema_from_metadata) - schema_from_metadata = metadata->getTableSchema(); - } - catch (...) - { - if (mode <= LoadingStrictnessLevel::CREATE) - throw; - - metadata.reset(); - configuration->setPaths({}); - tryLogCurrentException(__PRETTY_FUNCTION__); - } - - return std::make_shared>( - base_configuration, std::move(metadata), configuration, object_storage, - context, table_id_, - use_schema_from_metadata ? ColumnsDescription(schema_from_metadata) : columns_, - constraints_, comment_, format_settings_); - } - - String getName() const override { return DataLakeMetadata::name; } - - static ColumnsDescription getTableStructureFromData( - ObjectStoragePtr object_storage_, - ConfigurationPtr base_configuration, - const std::optional & format_settings_, - ContextPtr local_context) - { - auto metadata = DataLakeMetadata::create(object_storage_, base_configuration, local_context); - - auto schema_from_metadata = metadata->getTableSchema(); - if (!schema_from_metadata.empty()) - { - return ColumnsDescription(std::move(schema_from_metadata)); - } - else - { - ConfigurationPtr configuration = base_configuration->clone(); - configuration->setPaths(metadata->getDataFiles()); - std::string sample_path; - return Storage::resolveSchemaFromData( - object_storage_, configuration, format_settings_, sample_path, local_context); - } - } - - void updateConfiguration(ContextPtr local_context) override - { - Storage::updateConfiguration(local_context); - - auto new_metadata = DataLakeMetadata::create(Storage::object_storage, base_configuration, local_context); - if (current_metadata && *current_metadata == *new_metadata) - return; - - current_metadata = std::move(new_metadata); - auto updated_configuration = base_configuration->clone(); - updated_configuration->setPaths(current_metadata->getDataFiles()); - updated_configuration->setPartitionColumns(current_metadata->getPartitionColumns()); - - Storage::configuration = updated_configuration; - } - - template - IStorageDataLake( - ConfigurationPtr base_configuration_, - DataLakeMetadataPtr metadata_, - Args &&... args) - : Storage(std::forward(args)...) - , base_configuration(base_configuration_) - , current_metadata(std::move(metadata_)) - { - if (base_configuration->format == "auto") - { - base_configuration->format = Storage::configuration->format; - } - - if (current_metadata) - { - const auto & columns = current_metadata->getPartitionColumns(); - base_configuration->setPartitionColumns(columns); - Storage::configuration->setPartitionColumns(columns); - } - } - -private: - ConfigurationPtr base_configuration; - DataLakeMetadataPtr current_metadata; - - ReadFromFormatInfo prepareReadingFromFormat( - const Strings & requested_columns, - const StorageSnapshotPtr & storage_snapshot, - bool supports_subset_of_columns, - ContextPtr local_context) override - { - auto info = DB::prepareReadingFromFormat(requested_columns, storage_snapshot, supports_subset_of_columns); - if (!current_metadata) - { - Storage::updateConfiguration(local_context); - current_metadata = DataLakeMetadata::create(Storage::object_storage, base_configuration, local_context); - } - auto column_mapping = current_metadata->getColumnNameToPhysicalNameMapping(); - if (!column_mapping.empty()) - { - for (const auto & [column_name, physical_name] : column_mapping) - { - auto & column = info.format_header.getByName(column_name); - column.name = physical_name; - } - } - return info; - } -}; - -using StorageIceberg = IStorageDataLake; -using StorageDeltaLake = IStorageDataLake; -using StorageHudi = IStorageDataLake; - -} - -#endif diff --git a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp index ffc4dd09a3a..11ff749fd9d 100644 --- a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp +++ b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp @@ -50,7 +50,7 @@ extern const int UNSUPPORTED_METHOD; IcebergMetadata::IcebergMetadata( ObjectStoragePtr object_storage_, - ConfigurationPtr configuration_, + ConfigurationObservePtr configuration_, DB::ContextPtr context_, Int32 metadata_version_, Int32 format_version_, @@ -381,12 +381,12 @@ std::pair getMetadataFileAndVersion( } -DataLakeMetadataPtr IcebergMetadata::create( - ObjectStoragePtr object_storage, - ConfigurationPtr configuration, - ContextPtr local_context) +DataLakeMetadataPtr +IcebergMetadata::create(ObjectStoragePtr object_storage, ConfigurationObservePtr configuration, ContextPtr local_context) { - const auto [metadata_version, metadata_file_path] = getMetadataFileAndVersion(object_storage, *configuration); + auto configuration_ptr = configuration.lock(); + + const auto [metadata_version, metadata_file_path] = getMetadataFileAndVersion(object_storage, *configuration_ptr); LOG_DEBUG(getLogger("IcebergMetadata"), "Parse metadata {}", metadata_file_path); auto read_settings = local_context->getReadSettings(); auto buf = object_storage->readObject(StoredObject(metadata_file_path), read_settings); @@ -411,12 +411,13 @@ DataLakeMetadataPtr IcebergMetadata::create( if (snapshot->getValue("snapshot-id") == current_snapshot_id) { const auto path = snapshot->getValue("manifest-list"); - manifest_list_file = std::filesystem::path(configuration->getPath()) / "metadata" / std::filesystem::path(path).filename(); + manifest_list_file = std::filesystem::path(configuration_ptr->getPath()) / "metadata" / std::filesystem::path(path).filename(); break; } } - return std::make_unique(object_storage, configuration, local_context, metadata_version, format_version, manifest_list_file, schema_id, schema); + return std::make_unique( + object_storage, configuration_ptr, local_context, metadata_version, format_version, manifest_list_file, schema_id, schema); } /** @@ -446,6 +447,7 @@ DataLakeMetadataPtr IcebergMetadata::create( */ Strings IcebergMetadata::getDataFiles() const { + auto configuration_ptr = configuration.lock(); if (!data_files.empty()) return data_files; @@ -478,7 +480,7 @@ Strings IcebergMetadata::getDataFiles() const { const auto file_path = col_str->getDataAt(i).toView(); const auto filename = std::filesystem::path(file_path).filename(); - manifest_files.emplace_back(std::filesystem::path(configuration->getPath()) / "metadata" / filename); + manifest_files.emplace_back(std::filesystem::path(configuration_ptr->getPath()) / "metadata" / filename); } NameSet files; @@ -612,9 +614,9 @@ Strings IcebergMetadata::getDataFiles() const const auto status = status_int_column->getInt(i); const auto data_path = std::string(file_path_string_column->getDataAt(i).toView()); - const auto pos = data_path.find(configuration->getPath()); + const auto pos = data_path.find(configuration_ptr->getPath()); if (pos == std::string::npos) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected to find {} in data path: {}", configuration->getPath(), data_path); + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected to find {} in data path: {}", configuration_ptr->getPath(), data_path); const auto file_path = data_path.substr(pos); diff --git a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h index 7b0deab91c3..7811bcd8b4b 100644 --- a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h +++ b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h @@ -61,13 +61,13 @@ namespace DB class IcebergMetadata : public IDataLakeMetadata, private WithContext { public: - using ConfigurationPtr = StorageObjectStorage::ConfigurationPtr; + using ConfigurationObservePtr = StorageObjectStorage::ConfigurationObservePtr; static constexpr auto name = "Iceberg"; IcebergMetadata( ObjectStoragePtr object_storage_, - ConfigurationPtr configuration_, + ConfigurationObservePtr configuration_, ContextPtr context_, Int32 metadata_version_, Int32 format_version_, @@ -94,14 +94,14 @@ public: static DataLakeMetadataPtr create( ObjectStoragePtr object_storage, - ConfigurationPtr configuration, + ConfigurationObservePtr configuration, ContextPtr local_context); private: size_t getVersion() const { return metadata_version; } const ObjectStoragePtr object_storage; - const ConfigurationPtr configuration; + const ConfigurationObservePtr configuration; Int32 metadata_version; Int32 format_version; String manifest_list_file; diff --git a/src/Storages/ObjectStorage/DataLakes/registerDataLakeStorages.cpp b/src/Storages/ObjectStorage/DataLakes/registerDataLakeStorages.cpp deleted file mode 100644 index f0bd51de375..00000000000 --- a/src/Storages/ObjectStorage/DataLakes/registerDataLakeStorages.cpp +++ /dev/null @@ -1,132 +0,0 @@ -#include "config.h" - -#if USE_AWS_S3 - -# include -# include -# include -# include -# include -# include - - -namespace DB -{ - -#if USE_AVRO /// StorageIceberg depending on Avro to parse metadata with Avro format. - -void registerStorageIceberg(StorageFactory & factory) -{ - factory.registerStorage( - "Iceberg", - [&](const StorageFactory::Arguments & args) - { - auto configuration = std::make_shared(); - StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); - - return StorageIceberg::create( - configuration, args.getContext(), args.table_id, args.columns, args.constraints, args.comment, std::nullopt, args.mode); - }, - { - .supports_settings = false, - .supports_schema_inference = true, - .source_access_type = AccessType::S3, - }); - - factory.registerStorage( - "IcebergS3", - [&](const StorageFactory::Arguments & args) - { - auto configuration = std::make_shared(); - StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); - - return StorageIceberg::create( - configuration, args.getContext(), args.table_id, args.columns, args.constraints, args.comment, std::nullopt, args.mode); - }, - { - .supports_settings = false, - .supports_schema_inference = true, - .source_access_type = AccessType::S3, - }); - - factory.registerStorage( - "IcebergAzure", - [&](const StorageFactory::Arguments & args) - { - auto configuration = std::make_shared(); - StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), true); - - return StorageIceberg::create( - configuration, args.getContext(), args.table_id, args.columns, args.constraints, args.comment, std::nullopt, args.mode); - }, - { - .supports_settings = false, - .supports_schema_inference = true, - .source_access_type = AccessType::AZURE, - }); - - factory.registerStorage( - "IcebergLocal", - [&](const StorageFactory::Arguments & args) - { - auto configuration = std::make_shared(); - StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); - - return StorageIceberg::create( - configuration, args.getContext(), args.table_id, args.columns, - args.constraints, args.comment, std::nullopt, args.mode); - }, - { - .supports_settings = false, - .supports_schema_inference = true, - .source_access_type = AccessType::FILE, - }); -} - -#endif - -#if USE_PARQUET -void registerStorageDeltaLake(StorageFactory & factory) -{ - factory.registerStorage( - "DeltaLake", - [&](const StorageFactory::Arguments & args) - { - auto configuration = std::make_shared(); - StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); - - return StorageDeltaLake::create( - configuration, args.getContext(), args.table_id, args.columns, - args.constraints, args.comment, std::nullopt, args.mode); - }, - { - .supports_settings = false, - .supports_schema_inference = true, - .source_access_type = AccessType::S3, - }); -} -#endif - -void registerStorageHudi(StorageFactory & factory) -{ - factory.registerStorage( - "Hudi", - [&](const StorageFactory::Arguments & args) - { - auto configuration = std::make_shared(); - StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); - - return StorageHudi::create( - configuration, args.getContext(), args.table_id, args.columns, - args.constraints, args.comment, std::nullopt, args.mode); - }, - { - .supports_settings = false, - .supports_schema_inference = true, - .source_access_type = AccessType::S3, - }); -} - -} - -#endif diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index bc27820707c..f62e0fe20dc 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -124,12 +124,11 @@ bool StorageObjectStorage::supportsSubsetOfColumns(const ContextPtr & context) c return FormatFactory::instance().checkIfFormatSupportsSubsetOfColumns(configuration->format, context, format_settings); } -void StorageObjectStorage::updateConfiguration(ContextPtr context) +void StorageObjectStorage::Configuration::update(ObjectStoragePtr object_storage_ptr, ContextPtr context) { - IObjectStorage::ApplyNewSettingsOptions options{ .allow_client_change = !configuration->isStaticConfiguration() }; - object_storage->applyNewSettings(context->getConfigRef(), configuration->getTypeName() + ".", context, options); + IObjectStorage::ApplyNewSettingsOptions options{.allow_client_change = !isStaticConfiguration()}; + object_storage_ptr->applyNewSettings(context->getConfigRef(), getTypeName() + ".", context, options); } - namespace { class ReadFromObjectStorageStep : public SourceStepWithFilter @@ -243,7 +242,8 @@ private: }; } -ReadFromFormatInfo StorageObjectStorage::prepareReadingFromFormat( +ReadFromFormatInfo StorageObjectStorage::Configuration::prepareReadingFromFormat( + ObjectStoragePtr, const Strings & requested_columns, const StorageSnapshotPtr & storage_snapshot, bool supports_subset_of_columns, @@ -262,7 +262,7 @@ void StorageObjectStorage::read( size_t max_block_size, size_t num_streams) { - updateConfiguration(local_context); + configuration->update(object_storage, local_context); if (partition_by && configuration->withPartitionWildcard()) { throw Exception(ErrorCodes::NOT_IMPLEMENTED, @@ -270,8 +270,8 @@ void StorageObjectStorage::read( getName()); } - const auto read_from_format_info = prepareReadingFromFormat( - column_names, storage_snapshot, supportsSubsetOfColumns(local_context), local_context); + const auto read_from_format_info = configuration->prepareReadingFromFormat( + object_storage, column_names, storage_snapshot, supportsSubsetOfColumns(local_context), local_context); const bool need_only_count = (query_info.optimize_trivial_count || read_from_format_info.requested_columns.empty()) && local_context->getSettingsRef()[Setting::optimize_count_from_files]; @@ -300,7 +300,7 @@ SinkToStoragePtr StorageObjectStorage::write( ContextPtr local_context, bool /* async_insert */) { - updateConfiguration(local_context); + configuration->update(object_storage, local_context); const auto sample_block = metadata_snapshot->getSampleBlock(); const auto & settings = configuration->getQuerySettings(local_context); diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.h b/src/Storages/ObjectStorage/StorageObjectStorage.h index f39586c23b4..9781d5dbe6e 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.h +++ b/src/Storages/ObjectStorage/StorageObjectStorage.h @@ -25,6 +25,7 @@ class StorageObjectStorage : public IStorage public: class Configuration; using ConfigurationPtr = std::shared_ptr; + using ConfigurationObservePtr = std::weak_ptr; using ObjectInfo = RelativePathWithMetadata; using ObjectInfoPtr = std::shared_ptr; using ObjectInfos = std::vector; @@ -120,16 +121,8 @@ public: const ContextPtr & context); protected: - virtual void updateConfiguration(ContextPtr local_context); - String getPathSample(StorageInMemoryMetadata metadata, ContextPtr context); - virtual ReadFromFormatInfo prepareReadingFromFormat( - const Strings & requested_columns, - const StorageSnapshotPtr & storage_snapshot, - bool supports_subset_of_columns, - ContextPtr local_context); - static std::unique_ptr createReadBufferIterator( const ObjectStoragePtr & object_storage, const ConfigurationPtr & configuration, @@ -206,14 +199,26 @@ public: void setPartitionColumns(const DataLakePartitionColumns & columns) { partition_columns = columns; } const DataLakePartitionColumns & getPartitionColumns() const { return partition_columns; } + virtual bool isDataLakeConfiguration() const { return false; } + + virtual ReadFromFormatInfo prepareReadingFromFormat( + ObjectStoragePtr object_storage, + const Strings & requested_columns, + const StorageSnapshotPtr & storage_snapshot, + bool supports_subset_of_columns, + ContextPtr local_context); + String format = "auto"; String compression_method = "auto"; String structure = "auto"; + virtual void update(ObjectStoragePtr object_storage, ContextPtr local_context); + protected: virtual void fromNamedCollection(const NamedCollection & collection, ContextPtr context) = 0; virtual void fromAST(ASTs & args, ContextPtr context, bool with_structure) = 0; + void assertInitialized() const; bool initialized = false; diff --git a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp index d0cacc29adf..570e888da91 100644 --- a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp @@ -2,6 +2,7 @@ #include #include #include +#include #include #include #include @@ -148,4 +149,108 @@ void registerStorageObjectStorage(StorageFactory & factory) UNUSED(factory); } +#if USE_AVRO /// StorageIceberg depending on Avro to parse metadata with Avro format. + +void registerStorageIceberg(StorageFactory & factory) +{ + factory.registerStorage( + "Iceberg", + [&](const StorageFactory::Arguments & args) + { + auto configuration = std::make_shared(); + StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); + + return createStorageObjectStorage(args, configuration, args.getLocalContext()); + }, + { + .supports_settings = false, + .supports_schema_inference = true, + .source_access_type = AccessType::S3, + }); + + factory.registerStorage( + "IcebergS3", + [&](const StorageFactory::Arguments & args) + { + auto configuration = std::make_shared(); + StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); + + return createStorageObjectStorage(args, configuration, args.getLocalContext()); + }, + { + .supports_settings = false, + .supports_schema_inference = true, + .source_access_type = AccessType::S3, + }); + + factory.registerStorage( + "IcebergAzure", + [&](const StorageFactory::Arguments & args) + { + auto configuration = std::make_shared(); + StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), true); + + return createStorageObjectStorage(args, configuration, args.getLocalContext()); + }, + { + .supports_settings = false, + .supports_schema_inference = true, + .source_access_type = AccessType::AZURE, + }); + + factory.registerStorage( + "IcebergLocal", + [&](const StorageFactory::Arguments & args) + { + auto configuration = std::make_shared(); + StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); + + return createStorageObjectStorage(args, configuration, args.getLocalContext()); + }, + { + .supports_settings = false, + .supports_schema_inference = true, + .source_access_type = AccessType::FILE, + }); +} + +#endif + +#if USE_PARQUET +void registerStorageDeltaLake(StorageFactory & factory) +{ + factory.registerStorage( + "DeltaLake", + [&](const StorageFactory::Arguments & args) + { + auto configuration = std::make_shared(); + StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); + + return createStorageObjectStorage(args, configuration, args.getLocalContext()); + }, + { + .supports_settings = false, + .supports_schema_inference = true, + .source_access_type = AccessType::S3, + }); +} +#endif + +void registerStorageHudi(StorageFactory & factory) +{ + factory.registerStorage( + "Hudi", + [&](const StorageFactory::Arguments & args) + { + auto configuration = std::make_shared(); + StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); + + return createStorageObjectStorage(args, configuration, args.getLocalContext()); + }, + { + .supports_settings = false, + .supports_schema_inference = true, + .source_access_type = AccessType::S3, + }); +} } diff --git a/src/TableFunctions/ITableFunctionDataLake.h b/src/TableFunctions/ITableFunctionDataLake.h deleted file mode 100644 index db8287f97bf..00000000000 --- a/src/TableFunctions/ITableFunctionDataLake.h +++ /dev/null @@ -1,120 +0,0 @@ -#pragma once - -#include "config.h" -#include -#include -#include -#include -#include -#include -#include -#include -#include - - -namespace DB -{ - -template -class ITableFunctionDataLake : public TableFunction -{ -public: - static constexpr auto name = Name::name; - std::string getName() const override { return name; } - -protected: - StoragePtr executeImpl( - const ASTPtr & /* ast_function */, - ContextPtr context, - const std::string & table_name, - ColumnsDescription cached_columns, - bool /*is_insert_query*/) const override - { - ColumnsDescription columns; - auto configuration = TableFunction::getConfiguration(); - if (configuration->structure != "auto") - columns = parseColumnsListFromString(configuration->structure, context); - else if (!cached_columns.empty()) - columns = cached_columns; - - StoragePtr storage = Storage::create( - configuration, context, StorageID(TableFunction::getDatabaseName(), table_name), - columns, ConstraintsDescription{}, String{}, std::nullopt, LoadingStrictnessLevel::CREATE); - - storage->startup(); - return storage; - } - - const char * getStorageTypeName() const override { return name; } - - ColumnsDescription getActualTableStructure(ContextPtr context, bool is_insert_query) const override - { - auto configuration = TableFunction::getConfiguration(); - if (configuration->structure == "auto") - { - context->checkAccess(TableFunction::getSourceAccessType()); - auto object_storage = TableFunction::getObjectStorage(context, !is_insert_query); - return Storage::getTableStructureFromData(object_storage, configuration, std::nullopt, context); - } - else - { - return parseColumnsListFromString(configuration->structure, context); - } - } - - void parseArguments(const ASTPtr & ast_function, ContextPtr context) override - { - auto configuration = TableFunction::getConfiguration(); - configuration->format = "Parquet"; - /// Set default format to Parquet if it's not specified in arguments. - TableFunction::parseArguments(ast_function, context); - } -}; - -struct TableFunctionIcebergName -{ - static constexpr auto name = "iceberg"; -}; - -struct TableFunctionIcebergS3Name -{ - static constexpr auto name = "icebergS3"; -}; - -struct TableFunctionIcebergAzureName -{ - static constexpr auto name = "icebergAzure"; -}; - -struct TableFunctionIcebergLocalName -{ - static constexpr auto name = "icebergLocal"; -}; - -struct TableFunctionDeltaLakeName -{ - static constexpr auto name = "deltaLake"; -}; - -struct TableFunctionHudiName -{ - static constexpr auto name = "hudi"; -}; - -#if USE_AVRO -# if USE_AWS_S3 -using TableFunctionIceberg = ITableFunctionDataLake; -using TableFunctionIcebergS3 = ITableFunctionDataLake; -# endif -# if USE_AZURE_BLOB_STORAGE -using TableFunctionIcebergAzure = ITableFunctionDataLake; -# endif -using TableFunctionIcebergLocal = ITableFunctionDataLake; -#endif -#if USE_AWS_S3 -# if USE_PARQUET -using TableFunctionDeltaLake = ITableFunctionDataLake; -#endif -using TableFunctionHudi = ITableFunctionDataLake; -#endif -} diff --git a/src/TableFunctions/TableFunctionObjectStorage.cpp b/src/TableFunctions/TableFunctionObjectStorage.cpp index 9cebb91044a..60409a732c4 100644 --- a/src/TableFunctions/TableFunctionObjectStorage.cpp +++ b/src/TableFunctions/TableFunctionObjectStorage.cpp @@ -225,4 +225,94 @@ template class TableFunctionObjectStorage; #endif template class TableFunctionObjectStorage; + +#if USE_AVRO +void registerTableFunctionIceberg(TableFunctionFactory & factory) +{ +# if USE_AWS_S3 + factory.registerFunction( + {.documentation + = {.description = R"(The table function can be used to read the Iceberg table stored on S3 object store. Alias to icebergS3)", + .examples{{"iceberg", "SELECT * FROM iceberg(url, access_key_id, secret_access_key)", ""}}, + .categories{"DataLake"}}, + .allow_readonly = false}); + factory.registerFunction( + {.documentation + = {.description = R"(The table function can be used to read the Iceberg table stored on S3 object store.)", + .examples{{"icebergS3", "SELECT * FROM icebergS3(url, access_key_id, secret_access_key)", ""}}, + .categories{"DataLake"}}, + .allow_readonly = false}); + +# endif +# if USE_AZURE_BLOB_STORAGE + factory.registerFunction( + {.documentation + = {.description = R"(The table function can be used to read the Iceberg table stored on Azure object store.)", + .examples{{"icebergAzure", "SELECT * FROM icebergAzure(url, access_key_id, secret_access_key)", ""}}, + .categories{"DataLake"}}, + .allow_readonly = false}); +# endif + factory.registerFunction( + {.documentation + = {.description = R"(The table function can be used to read the Iceberg table stored locally.)", + .examples{{"icebergLocal", "SELECT * FROM icebergLocal(filename)", ""}}, + .categories{"DataLake"}}, + .allow_readonly = false}); +} +#endif + +#if USE_AWS_S3 +# if USE_PARQUET +void registerTableFunctionDeltaLake(TableFunctionFactory & factory) +{ + factory.registerFunction( + {.documentation + = {.description = R"(The table function can be used to read the DeltaLake table stored on object store.)", + .examples{{"deltaLake", "SELECT * FROM deltaLake(url, access_key_id, secret_access_key)", ""}}, + .categories{"DataLake"}}, + .allow_readonly = false}); +} +# endif + +void registerTableFunctionHudi(TableFunctionFactory & factory) +{ + factory.registerFunction( + {.documentation + = {.description = R"(The table function can be used to read the Hudi table stored on object store.)", + .examples{{"hudi", "SELECT * FROM hudi(url, access_key_id, secret_access_key)", ""}}, + .categories{"DataLake"}}, + .allow_readonly = false}); +} +#endif + +void registerDataLakeTableFunctions(TableFunctionFactory & factory) +{ + UNUSED(factory); +#if USE_AVRO + registerTableFunctionIceberg(factory); +#endif +#if USE_AWS_S3 +# if USE_PARQUET + registerTableFunctionDeltaLake(factory); +# endif + registerTableFunctionHudi(factory); +#endif +} + +#if USE_AVRO +# if USE_AWS_S3 +template class TableFunctionObjectStorage; +template class TableFunctionObjectStorage; +# endif +# if USE_AZURE_BLOB_STORAGE +template class TableFunctionObjectStorage; +# endif +template class TableFunctionObjectStorage; +#endif +#if USE_AWS_S3 +# if USE_PARQUET +template class TableFunctionObjectStorage; +# endif +template class TableFunctionObjectStorage; +#endif } diff --git a/src/TableFunctions/TableFunctionObjectStorage.h b/src/TableFunctions/TableFunctionObjectStorage.h index 6b923f93e75..3cf86f982d1 100644 --- a/src/TableFunctions/TableFunctionObjectStorage.h +++ b/src/TableFunctions/TableFunctionObjectStorage.h @@ -2,6 +2,7 @@ #include #include +#include #include #include #include @@ -61,6 +62,42 @@ struct LocalDefinition static constexpr auto storage_type_name = "Local"; }; +struct IcebergDefinition +{ + static constexpr auto name = "iceberg"; + static constexpr auto storage_type_name = "S3"; +}; + +struct IcebergS3Definition +{ + static constexpr auto name = "icebergS3"; + static constexpr auto storage_type_name = "S3"; +}; + +struct IcebergAzureDefinition +{ + static constexpr auto name = "icebergAzure"; + static constexpr auto storage_type_name = "Azure"; +}; + +struct IcebergLocalDefinition +{ + static constexpr auto name = "icebergLocal"; + static constexpr auto storage_type_name = "Local"; +}; + +struct DeltaLakeDefinition +{ + static constexpr auto name = "deltaLake"; + static constexpr auto storage_type_name = "S3"; +}; + +struct HudiDefinition +{ + static constexpr auto name = "hudi"; + static constexpr auto storage_type_name = "S3"; +}; + template class TableFunctionObjectStorage : public ITableFunction { @@ -137,4 +174,22 @@ using TableFunctionHDFS = TableFunctionObjectStorage; + + +#if USE_AVRO +# if USE_AWS_S3 +using TableFunctionIceberg = TableFunctionObjectStorage; +using TableFunctionIcebergS3 = TableFunctionObjectStorage; +# endif +# if USE_AZURE_BLOB_STORAGE +using TableFunctionIcebergAzure = TableFunctionObjectStorage; +# endif +using TableFunctionIcebergLocal = TableFunctionObjectStorage; +#endif +#if USE_AWS_S3 +# if USE_PARQUET +using TableFunctionDeltaLake = TableFunctionObjectStorage; +# endif +using TableFunctionHudi = TableFunctionObjectStorage; +#endif } diff --git a/src/TableFunctions/registerDataLakeTableFunctions.cpp b/src/TableFunctions/registerDataLakeTableFunctions.cpp deleted file mode 100644 index 8361d8a7977..00000000000 --- a/src/TableFunctions/registerDataLakeTableFunctions.cpp +++ /dev/null @@ -1,88 +0,0 @@ -#include -#include - -namespace DB -{ - -#if USE_AVRO -void registerTableFunctionIceberg(TableFunctionFactory & factory) -{ -# if USE_AWS_S3 - factory.registerFunction( - {.documentation - = {.description = R"(The table function can be used to read the Iceberg table stored on S3 object store. Alias to icebergS3)", - .examples{{"iceberg", "SELECT * FROM iceberg(url, access_key_id, secret_access_key)", ""}}, - .categories{"DataLake"}}, - .allow_readonly = false}); - factory.registerFunction( - {.documentation - = {.description = R"(The table function can be used to read the Iceberg table stored on S3 object store.)", - .examples{{"icebergS3", "SELECT * FROM icebergS3(url, access_key_id, secret_access_key)", ""}}, - .categories{"DataLake"}}, - .allow_readonly = false}); - -# endif -# if USE_AZURE_BLOB_STORAGE - factory.registerFunction( - {.documentation - = {.description = R"(The table function can be used to read the Iceberg table stored on Azure object store.)", - .examples{{"icebergAzure", "SELECT * FROM icebergAzure(url, access_key_id, secret_access_key)", ""}}, - .categories{"DataLake"}}, - .allow_readonly = false}); -# endif - factory.registerFunction( - {.documentation - = {.description = R"(The table function can be used to read the Iceberg table stored locally.)", - .examples{{"icebergLocal", "SELECT * FROM icebergLocal(filename)", ""}}, - .categories{"DataLake"}}, - .allow_readonly = false}); -} -#endif - -#if USE_AWS_S3 -# if USE_PARQUET -void registerTableFunctionDeltaLake(TableFunctionFactory & factory) -{ - factory.registerFunction( - { - .documentation = - { - .description=R"(The table function can be used to read the DeltaLake table stored on object store.)", - .examples{{"deltaLake", "SELECT * FROM deltaLake(url, access_key_id, secret_access_key)", ""}}, - .categories{"DataLake"} - }, - .allow_readonly = false - }); -} -#endif - -void registerTableFunctionHudi(TableFunctionFactory & factory) -{ - factory.registerFunction( - { - .documentation = - { - .description=R"(The table function can be used to read the Hudi table stored on object store.)", - .examples{{"hudi", "SELECT * FROM hudi(url, access_key_id, secret_access_key)", ""}}, - .categories{"DataLake"} - }, - .allow_readonly = false - }); -} -#endif - -void registerDataLakeTableFunctions(TableFunctionFactory & factory) -{ - UNUSED(factory); -#if USE_AVRO - registerTableFunctionIceberg(factory); -#endif -#if USE_AWS_S3 -# if USE_PARQUET - registerTableFunctionDeltaLake(factory); -#endif - registerTableFunctionHudi(factory); -#endif -} - -} From 7b01c19d06bf424cfcfaad154a12575a9ad81145 Mon Sep 17 00:00:00 2001 From: vdimir Date: Fri, 27 Sep 2024 15:06:27 +0000 Subject: [PATCH 074/680] fix header... --- src/Interpreters/InterpreterSelectQuery.cpp | 1 + src/Planner/PlannerJoinTree.cpp | 1 + src/Processors/QueryPlan/JoinStep.cpp | 32 ++++++++++++++++--- src/Processors/QueryPlan/JoinStep.h | 4 +++ .../Transforms/ColumnPermuteTransform.cpp | 4 +-- tests/integration/helpers/random_settings.py | 2 +- .../02001_join_on_const_bs_long.sql.j2 | 4 +-- 7 files changed, 39 insertions(+), 9 deletions(-) diff --git a/src/Interpreters/InterpreterSelectQuery.cpp b/src/Interpreters/InterpreterSelectQuery.cpp index bfd9be70bb5..01483b34092 100644 --- a/src/Interpreters/InterpreterSelectQuery.cpp +++ b/src/Interpreters/InterpreterSelectQuery.cpp @@ -1887,6 +1887,7 @@ void InterpreterSelectQuery::executeImpl(QueryPlan & query_plan, std::optional

setStepDescription(fmt::format("JOIN {}", expressions.join->pipelineType())); diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp index 543dc1a88f6..4f4d7e22022 100644 --- a/src/Planner/PlannerJoinTree.cpp +++ b/src/Planner/PlannerJoinTree.cpp @@ -1641,6 +1641,7 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ std::move(join_algorithm), settings[Setting::max_block_size], settings[Setting::max_threads], + outer_scope_columns, false /*optimize_read_in_order*/); join_step->inner_table_selection_mode = settings[Setting::query_plan_join_inner_table_selection]; diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index fefb193827f..9fdfeedb111 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -45,14 +45,13 @@ size_t getPrefixLength(const NameSet & prefix, const Names & names) if (!prefix.contains(names[i])) break; } - LOG_DEBUG(&Poco::Logger::get("XXXX"), "{}:{}: [{}] [{}] -> {}", __FILE__, __LINE__, fmt::join(names, ", "), fmt::join(prefix, ", "), i); return i; } std::vector getPermutationToRotate(size_t prefix_size, size_t total_size) { std::vector permutation(total_size); - size_t i = prefix_size; + size_t i = prefix_size % total_size; for (auto & elem : permutation) { elem = i; @@ -92,8 +91,13 @@ JoinStep::JoinStep( JoinPtr join_, size_t max_block_size_, size_t max_streams_, + NameSet required_output_, bool keep_left_read_in_order_) - : join(std::move(join_)), max_block_size(max_block_size_), max_streams(max_streams_), keep_left_read_in_order(keep_left_read_in_order_) + : join(std::move(join_)) + , max_block_size(max_block_size_) + , max_streams(max_streams_) + , required_output(std::move(required_output_)) + , keep_left_read_in_order(keep_left_read_in_order_) { updateInputStreams(DataStreams{left_stream_, right_stream_}); } @@ -128,9 +132,20 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines const auto & result_names = pipeline->getHeader().getNames(); size_t prefix_size = getPrefixLength(rhs_names, result_names); - if (0 < prefix_size && prefix_size < result_names.size()) + if (!columns_to_remove.empty() || (0 < prefix_size && prefix_size < result_names.size())) { auto column_permutation = getPermutationToRotate(prefix_size, result_names.size()); + size_t n = 0; + auto it = columns_to_remove.begin(); + for (size_t i = 0; i < column_permutation.size(); ++i) + { + if (it != columns_to_remove.end() && *it == i) + ++it; + else + column_permutation[n++] = column_permutation[i]; + } + column_permutation.resize(n); + pipeline->addSimpleTransform([column_perm = std::move(column_permutation)](const Block & header) { return std::make_shared(header, std::move(column_perm)); @@ -174,6 +189,15 @@ void JoinStep::updateOutputStream() if (swap_streams) result_header = rotateBlock(result_header, input_streams[1].header); + columns_to_remove.clear(); + for (size_t i = 0; i < result_header.columns(); ++i) + { + if (required_output.empty()) + break; + if (!required_output.contains(result_header.getByPosition(i).name)) + columns_to_remove.insert(i); + } + result_header.erase(columns_to_remove); output_stream = DataStream { .header = result_header }; } diff --git a/src/Processors/QueryPlan/JoinStep.h b/src/Processors/QueryPlan/JoinStep.h index 96c02f9fd19..30b20a0d3a5 100644 --- a/src/Processors/QueryPlan/JoinStep.h +++ b/src/Processors/QueryPlan/JoinStep.h @@ -20,6 +20,7 @@ public: JoinPtr join_, size_t max_block_size_, size_t max_streams_, + NameSet required_output_, bool keep_left_read_in_order_); String getName() const override { return "Join"; } @@ -48,6 +49,9 @@ private: JoinPtr join; size_t max_block_size; size_t max_streams; + + NameSet required_output; + std::set columns_to_remove; bool keep_left_read_in_order; }; diff --git a/src/Processors/Transforms/ColumnPermuteTransform.cpp b/src/Processors/Transforms/ColumnPermuteTransform.cpp index ac7793bd136..2921bcac177 100644 --- a/src/Processors/Transforms/ColumnPermuteTransform.cpp +++ b/src/Processors/Transforms/ColumnPermuteTransform.cpp @@ -10,8 +10,8 @@ template void applyPermutation(std::vector & data, const std::vector & permutation) { std::vector res; - res.reserve(data.size()); - for (size_t i = 0; i < data.size(); ++i) + res.reserve(permutation.size()); + for (size_t i = 0; i < permutation.size(); ++i) res.emplace_back(std::move(data[permutation[i]])); data = std::move(res); } diff --git a/tests/integration/helpers/random_settings.py b/tests/integration/helpers/random_settings.py index a34d8e93c47..3a51d8cf52f 100644 --- a/tests/integration/helpers/random_settings.py +++ b/tests/integration/helpers/random_settings.py @@ -7,7 +7,7 @@ def randomize_settings(): yield "max_block_size", random.randint(8000, 100000) if random.random() < 0.5: yield "query_plan_join_inner_table_selection", random.choice( - ["auto", "left", "right"] + ["auto", "left"] ) diff --git a/tests/queries/0_stateless/02001_join_on_const_bs_long.sql.j2 b/tests/queries/0_stateless/02001_join_on_const_bs_long.sql.j2 index 1726bcb7062..83548e087bd 100644 --- a/tests/queries/0_stateless/02001_join_on_const_bs_long.sql.j2 +++ b/tests/queries/0_stateless/02001_join_on_const_bs_long.sql.j2 @@ -1,8 +1,8 @@ DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS t2; -CREATE TABLE t1 (id Int) ENGINE = MergeTree ORDER BY id; -CREATE TABLE t2 (id Int) ENGINE = MergeTree ORDER BY id; +CREATE TABLE t1 (id Int) ENGINE = TinyLog; +CREATE TABLE t2 (id Int) ENGINE = TinyLog; INSERT INTO t1 VALUES (1), (2); INSERT INTO t2 SELECT number + 5 AS x FROM (SELECT * FROM system.numbers LIMIT 1111); From de6517367677773b97ddcb0820859493a8295ac0 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Fri, 27 Sep 2024 16:01:52 +0000 Subject: [PATCH 075/680] Automatic style fix --- tests/integration/helpers/random_settings.py | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/tests/integration/helpers/random_settings.py b/tests/integration/helpers/random_settings.py index 3a51d8cf52f..32cde54d0e7 100644 --- a/tests/integration/helpers/random_settings.py +++ b/tests/integration/helpers/random_settings.py @@ -6,9 +6,7 @@ def randomize_settings(): if random.random() < 0.5: yield "max_block_size", random.randint(8000, 100000) if random.random() < 0.5: - yield "query_plan_join_inner_table_selection", random.choice( - ["auto", "left"] - ) + yield "query_plan_join_inner_table_selection", random.choice(["auto", "left"]) def write_random_settings_config(destination): From decfe0b676ab4a334fd2fcc61dd5a211f5fe7d44 Mon Sep 17 00:00:00 2001 From: vdimir Date: Fri, 27 Sep 2024 16:52:44 +0000 Subject: [PATCH 076/680] fix build --- src/Core/Settings.cpp | 2 +- src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp | 8 +++----- 2 files changed, 4 insertions(+), 6 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 4e63c3ae957..dcd1d33ff27 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -345,7 +345,7 @@ namespace ErrorCodes M(Bool, any_join_distinct_right_table_keys, false, "Enable old ANY JOIN logic with many-to-one left-to-right table keys mapping for all ANY JOINs. It leads to confusing not equal results for 't1 ANY LEFT JOIN t2' and 't2 ANY RIGHT JOIN t1'. ANY RIGHT JOIN needs one-to-many keys mapping to be consistent with LEFT one.", IMPORTANT) \ M(Bool, single_join_prefer_left_table, true, "For single JOIN in case of identifier ambiguity prefer left table", IMPORTANT) \ \ - M(JoinInnerTableSelectionMode, query_plan_join_inner_table_selection, JoinInnerTableSelectionMode::Auto, "Select the side of the join to be the inner table in the query plan. Possible values: 'auto', 'left', 'right'.", 0) \ + M(JoinInnerTableSelectionMode, query_plan_join_inner_table_selection, JoinInnerTableSelectionMode::Auto, "Select the side of the join to be the inner table in the query plan. Supported only for `ALL` join strictness with `JOIN ON` clause. Possible values: 'auto', 'left', 'right'.", 0) \ M(UInt64, preferred_block_size_bytes, 1000000, "This setting adjusts the data block size for query processing and represents additional fine-tuning to the more rough 'max_block_size' setting. If the columns are large and with 'max_block_size' rows the block size is likely to be larger than the specified amount of bytes, its size will be lowered for better CPU cache locality.", 0) \ \ M(UInt64, max_replica_delay_for_distributed_queries, 300, "If set, distributed queries of Replicated tables will choose servers with replication delay in seconds less than the specified value (not inclusive). Zero means do not take delay into account.", 0) \ diff --git a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp index 8074304de52..cd66a230038 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp @@ -56,11 +56,9 @@ void optimizeJoin(QueryPlan::Node & node, QueryPlan::Nodes &) return; const auto & table_join = join->getTableJoin(); - auto kind = table_join.kind(); - if (table_join.hasUsing() - || table_join.strictness() != JoinStrictness::All - || (kind != JoinKind::Inner && kind != JoinKind::Left - && kind != JoinKind::Right && kind != JoinKind::Full)) + /// fixme: USING clause handled specially in join algorithm, so swap breaks it + /// fixme: Swapping for SEMI and ANTI joins should be alright, need to try to enable it and test + if (table_join.hasUsing() || table_join.strictness() != JoinStrictness::All) return; bool need_swap = false; From aaabaadf5650f37507ba07b7eca3327b4a41db95 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Sat, 28 Sep 2024 14:15:43 +0000 Subject: [PATCH 077/680] cleanup --- .../Merges/Algorithms/MergeTreeReadInfo.h | 10 +- .../Algorithms/MergingSortedAlgorithm.cpp | 17 --- .../QueryPlan/ReadFromMergeTree.cpp | 25 ---- .../02521_aggregation_by_partitions.reference | 112 +++++------------- 4 files changed, 33 insertions(+), 131 deletions(-) diff --git a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h index a4baaca215b..253d008c21d 100644 --- a/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h +++ b/src/Processors/Merges/Algorithms/MergeTreeReadInfo.h @@ -71,21 +71,13 @@ inline void setVirtualRow(Chunk & chunk, const Block & header, bool apply_virtua { if (!col.type->equals(*pk_col->type)) throw Exception(ErrorCodes::LOGICAL_ERROR, - "Virtual row has different tupe for {}. Expected {}, got {}", + "Virtual row has different type for {}. Expected {}, got {}", col.name, col.dumpStructure(), pk_col->dumpStructure()); ordered_columns.push_back(pk_col->column); } else ordered_columns.push_back(col.type->createColumnConstWithDefaultValue(1)); - - // ColumnPtr current_column = type_and_name.type->createColumn(); - - // size_t pos = type_and_name.name.find_last_of('.'); - // String column_name = (pos == String::npos) ? type_and_name.name : type_and_name.name.substr(pos + 1); - - // const ColumnWithTypeAndName * column = pk_block.findByName(column_name, true); - // ordered_columns.push_back(column ? column->column : current_column->cloneResized(1)); } chunk.setColumns(ordered_columns, 1); diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp index 331b67066be..f2ebf9053ea 100644 --- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp @@ -8,11 +8,6 @@ namespace DB { -namespace ErrorCodes -{ - extern const int NOT_IMPLEMENTED; -} - MergingSortedAlgorithm::MergingSortedAlgorithm( Block header_, size_t num_inputs, @@ -150,9 +145,6 @@ IMergingAlgorithm::Status MergingSortedAlgorithm::mergeImpl(TSortingHeap & queue auto current = queue.current(); - // if (isVirtualRow(current_inputs[current.impl->order].chunk)) - // throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Virtual row is not implemented for Non-batch mode."); - if (current.impl->isLast() && current_inputs[current.impl->order].skip_last_row) { /// Get the next block from the corresponding source, if there is one. @@ -249,15 +241,6 @@ IMergingAlgorithm::Status MergingSortedAlgorithm::mergeBatchImpl(TSortingQueue & auto [current_ptr, initial_batch_size] = queue.current(); auto current = *current_ptr; - // if (isVirtualRow(current_inputs[current.impl->order].chunk)) - // { - // /// If virtual row is detected, there should be only one row as a single chunk, - // /// and always skip this chunk to pull the next one. - // chassert(initial_batch_size == 1); - // queue.removeTop(); - // return Status(current.impl->order); - // } - bool batch_skip_last_row = false; if (current.impl->isLast(initial_batch_size) && current_inputs[current.impl->order].skip_last_row) { diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index e90d6165aa3..6622662d0a0 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -1133,33 +1133,8 @@ Pipe ReadFromMergeTree::spreadMarkRangesAmongStreamsWithOrder( splitted_parts_and_ranges.emplace_back(std::move(new_parts)); } - // bool primary_key_type_supports_virtual_row = true; - // const auto & actions = storage_snapshot->metadata->getPrimaryKey().expression->getActions(); - // for (const auto & action : actions) - // { - // if (action.node->type != ActionsDAG::ActionType::INPUT) - // { - // primary_key_type_supports_virtual_row = false; - // break; - // } - // } - - // /// If possible in the optimization stage, check whether there are more than one branch. - // if (virtual_row_status == VirtualRowStatus::Possible) - // virtual_row_status = splitted_parts_and_ranges.size() > 1 - // || (splitted_parts_and_ranges.size() == 1 && splitted_parts_and_ranges[0].size() > 1) - // ? VirtualRowStatus::Yes : VirtualRowStatus::NoConsiderInLogicalPlan; - for (auto && item : splitted_parts_and_ranges) - { - // bool enable_current_virtual_row = false; - // if (virtual_row_status == VirtualRowStatus::Yes) - // enable_current_virtual_row = true; - // else if (virtual_row_status == VirtualRowStatus::NoConsiderInLogicalPlan) - // enable_current_virtual_row = (need_preliminary_merge || output_each_partition_through_separate_port) && item.size() > 1; - pipes.emplace_back(readInOrder(std::move(item), column_names, pool_settings, read_type, input_order_info->limit)); - } } Block pipe_header; diff --git a/tests/queries/0_stateless/02521_aggregation_by_partitions.reference b/tests/queries/0_stateless/02521_aggregation_by_partitions.reference index addc36421c3..87b2d5c3430 100644 --- a/tests/queries/0_stateless/02521_aggregation_by_partitions.reference +++ b/tests/queries/0_stateless/02521_aggregation_by_partitions.reference @@ -160,100 +160,52 @@ ExpressionTransform × 16 (ReadFromMergeTree) MergingSortedTransform 2 → 1 ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 MergingSortedTransform 2 → 1 ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - MergingSortedTransform 2 → 1 - ExpressionTransform × 2 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 - VirtualRowTransform - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 + MergingSortedTransform 2 → 1 + ExpressionTransform × 2 + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) × 2 0 → 1 1000000 Skip merging: 1 Skip merging: 1 From c7f662dc989833d707d15ef086edd69c1d5b64cd Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Mon, 30 Sep 2024 02:43:53 +0000 Subject: [PATCH 078/680] fix build, add initial fuzzing processing --- .../data_type_deserialization_fuzzer.cpp | 1 + src/Parsers/fuzzers/CMakeLists.txt | 4 +- .../fuzzers/codegen_fuzzer/CMakeLists.txt | 2 +- tests/fuzz/runner.py | 76 +++++++++++++++++-- 4 files changed, 75 insertions(+), 8 deletions(-) diff --git a/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp b/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp index f9a733647e1..216b252ad0f 100644 --- a/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp +++ b/src/DataTypes/fuzzers/data_type_deserialization_fuzzer.cpp @@ -3,6 +3,7 @@ #include #include +#include #include #include diff --git a/src/Parsers/fuzzers/CMakeLists.txt b/src/Parsers/fuzzers/CMakeLists.txt index 903319d733c..c829c26a805 100644 --- a/src/Parsers/fuzzers/CMakeLists.txt +++ b/src/Parsers/fuzzers/CMakeLists.txt @@ -2,10 +2,10 @@ clickhouse_add_executable(lexer_fuzzer lexer_fuzzer.cpp ${SRCS}) target_link_libraries(lexer_fuzzer PRIVATE clickhouse_parsers) clickhouse_add_executable(select_parser_fuzzer select_parser_fuzzer.cpp ${SRCS}) -target_link_libraries(select_parser_fuzzer PRIVATE clickhouse_parsers dbms) +target_link_libraries(select_parser_fuzzer PRIVATE clickhouse_parsers clickhouse_functions dbms) clickhouse_add_executable(create_parser_fuzzer create_parser_fuzzer.cpp ${SRCS}) -target_link_libraries(create_parser_fuzzer PRIVATE clickhouse_parsers dbms) +target_link_libraries(create_parser_fuzzer PRIVATE clickhouse_parsers clickhouse_functions dbms) add_subdirectory(codegen_fuzzer) diff --git a/src/Parsers/fuzzers/codegen_fuzzer/CMakeLists.txt b/src/Parsers/fuzzers/codegen_fuzzer/CMakeLists.txt index 74fdcff79f7..ee17e03fce2 100644 --- a/src/Parsers/fuzzers/codegen_fuzzer/CMakeLists.txt +++ b/src/Parsers/fuzzers/codegen_fuzzer/CMakeLists.txt @@ -47,4 +47,4 @@ target_compile_options (codegen_select_fuzzer PRIVATE -Wno-newline-eof) target_link_libraries(protoc ch_contrib::fuzzer) target_include_directories(codegen_select_fuzzer SYSTEM BEFORE PRIVATE "${CMAKE_CURRENT_BINARY_DIR}") -target_link_libraries(codegen_select_fuzzer PRIVATE ch_contrib::protobuf_mutator ch_contrib::protoc dbms) +target_link_libraries(codegen_select_fuzzer PRIVATE ch_contrib::protobuf_mutator ch_contrib::protoc clickhouse_functions dbms) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 44259228f60..5abab282afd 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -4,19 +4,70 @@ import configparser import logging import os from pathlib import Path +import re import subprocess DEBUGGER = os.getenv("DEBUGGER", "") FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") +def report(source: str, reason: str, call_stack: list, test_unit: str): + print(f"########### REPORT: {source} {reason} {test_unit}") + for line in call_stack: + print(f" {line}") + print("########### END OF REPORT ###########") + +def process_fuzzer_output(output: str): + pass + +def process_error(error: str): + ERROR = r'^==\d+== ERROR: (\S+): (.*)' + error_source = '' + error_reason = '' + SUMMARY = r'^SUMMARY: ' + TEST_UNIT_LINE = r"artifact_prefix='.*/'; Test unit written to (.*)" + test_unit = '' + CALL_STACK_LINE = r'^\s+(#\d+.*)' + call_stack = [] + is_call_stack = False + + for line_num, line in enumerate(error.splitlines(), 1): + + if is_call_stack: + match = re.search(CALL_STACK_LINE, line) + if match: + call_stack.append(match.group(1)) + continue + else: + if re.search(SUMMARY, line): + is_call_stack = False + continue + + if not call_stack and not is_call_stack: + match = re.search(ERROR, line) + if match: + error_source = match.group(1) + error_reason = match.group(2) + is_call_stack = True + continue + + match = re.search(TEST_UNIT_LINE, line) + if match: + test_unit = match.group(1) + + report(error_source, error_reason, call_stack, test_unit) def run_fuzzer(fuzzer: str): logging.info("Running fuzzer %s...", fuzzer) - corpus_dir = f"{fuzzer}.in" - with Path(corpus_dir) as path: + seed_corpus_dir = f"{fuzzer}.in" + with Path(seed_corpus_dir) as path: if not path.exists() or not path.is_dir(): - corpus_dir = "" + seed_corpus_dir = "" + + active_corpus_dir = f"{fuzzer}.corpus" + if not os.path.exists(active_corpus_dir): + os.makedirs(active_corpus_dir) + options_file = f"{fuzzer}.options" custom_libfuzzer_options = "" @@ -53,7 +104,7 @@ def run_fuzzer(fuzzer: str): for key, value in parser["fuzzer_arguments"].items() ) - cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {corpus_dir}" + cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {active_corpus_dir} {seed_corpus_dir}" if custom_libfuzzer_options: cmd_line += f" {custom_libfuzzer_options}" if fuzzer_arguments: @@ -65,8 +116,23 @@ def run_fuzzer(fuzzer: str): cmd_line += " < /dev/null" logging.info("...will execute: %s", cmd_line) - subprocess.check_call(cmd_line, shell=True) + #subprocess.check_call(cmd_line, shell=True) + try: + result = subprocess.run( + cmd_line, + stderr=subprocess.PIPE, + stdout=subprocess.DEVNULL, + text=True, + check=True, + shell=True + ) + except subprocess.CalledProcessError as e: +# print("Command failed with error:", e) + print("Stderr output:", e.stderr) + process_error(e.stderr) + else: + process_fuzzer_output(result.stderr) def main(): logging.basicConfig(level=logging.INFO) From abd3747806dd8f3fb75eac4f0a5cea3c6eacffc2 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Mon, 30 Sep 2024 03:43:34 +0000 Subject: [PATCH 079/680] fix style --- tests/fuzz/runner.py | 30 ++++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 5abab282afd..6825a072e2d 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -3,30 +3,33 @@ import configparser import logging import os -from pathlib import Path import re import subprocess +from pathlib import Path DEBUGGER = os.getenv("DEBUGGER", "") FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") + def report(source: str, reason: str, call_stack: list, test_unit: str): print(f"########### REPORT: {source} {reason} {test_unit}") for line in call_stack: print(f" {line}") print("########### END OF REPORT ###########") + def process_fuzzer_output(output: str): pass + def process_error(error: str): - ERROR = r'^==\d+== ERROR: (\S+): (.*)' - error_source = '' - error_reason = '' - SUMMARY = r'^SUMMARY: ' + ERROR = r"^==\d+== ERROR: (\S+): (.*)" + error_source = "" + error_reason = "" + SUMMARY = r"^SUMMARY: " TEST_UNIT_LINE = r"artifact_prefix='.*/'; Test unit written to (.*)" - test_unit = '' - CALL_STACK_LINE = r'^\s+(#\d+.*)' + test_unit = "" + CALL_STACK_LINE = r"^\s+(#\d+.*)" call_stack = [] is_call_stack = False @@ -56,6 +59,7 @@ def process_error(error: str): report(error_source, error_reason, call_stack, test_unit) + def run_fuzzer(fuzzer: str): logging.info("Running fuzzer %s...", fuzzer) @@ -68,7 +72,6 @@ def run_fuzzer(fuzzer: str): if not os.path.exists(active_corpus_dir): os.makedirs(active_corpus_dir) - options_file = f"{fuzzer}.options" custom_libfuzzer_options = "" fuzzer_arguments = "" @@ -104,7 +107,9 @@ def run_fuzzer(fuzzer: str): for key, value in parser["fuzzer_arguments"].items() ) - cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {active_corpus_dir} {seed_corpus_dir}" + cmd_line = ( + f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {active_corpus_dir} {seed_corpus_dir}" + ) if custom_libfuzzer_options: cmd_line += f" {custom_libfuzzer_options}" if fuzzer_arguments: @@ -116,7 +121,7 @@ def run_fuzzer(fuzzer: str): cmd_line += " < /dev/null" logging.info("...will execute: %s", cmd_line) - #subprocess.check_call(cmd_line, shell=True) + # subprocess.check_call(cmd_line, shell=True) try: result = subprocess.run( @@ -125,15 +130,16 @@ def run_fuzzer(fuzzer: str): stdout=subprocess.DEVNULL, text=True, check=True, - shell=True + shell=True, ) except subprocess.CalledProcessError as e: -# print("Command failed with error:", e) + # print("Command failed with error:", e) print("Stderr output:", e.stderr) process_error(e.stderr) else: process_fuzzer_output(result.stderr) + def main(): logging.basicConfig(level=logging.INFO) From 55ae792706177ce96940f23d7147914db06dcf39 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Mon, 30 Sep 2024 04:02:25 +0000 Subject: [PATCH 080/680] fix style --- tests/fuzz/runner.py | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 6825a072e2d..deb219baff9 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -18,6 +18,7 @@ def report(source: str, reason: str, call_stack: list, test_unit: str): print("########### END OF REPORT ###########") +# pylint: disable=unused-argument def process_fuzzer_output(output: str): pass @@ -33,6 +34,7 @@ def process_error(error: str): call_stack = [] is_call_stack = False + # pylint: disable=unused-variable for line_num, line in enumerate(error.splitlines(), 1): if is_call_stack: @@ -40,10 +42,10 @@ def process_error(error: str): if match: call_stack.append(match.group(1)) continue - else: - if re.search(SUMMARY, line): - is_call_stack = False - continue + + if re.search(SUMMARY, line): + is_call_stack = False + continue if not call_stack and not is_call_stack: match = re.search(ERROR, line) From ba5a0e98e3acc83531542ed6b35b57a1a0c10fee Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Mon, 30 Sep 2024 13:03:17 +0000 Subject: [PATCH 081/680] fix build --- src/AggregateFunctions/fuzzers/CMakeLists.txt | 2 +- src/Core/fuzzers/CMakeLists.txt | 2 +- src/DataTypes/fuzzers/CMakeLists.txt | 2 +- src/Formats/fuzzers/CMakeLists.txt | 2 +- src/Interpreters/fuzzers/CMakeLists.txt | 1 + src/Storages/fuzzers/CMakeLists.txt | 2 +- 6 files changed, 6 insertions(+), 5 deletions(-) diff --git a/src/AggregateFunctions/fuzzers/CMakeLists.txt b/src/AggregateFunctions/fuzzers/CMakeLists.txt index 6a7be0d4377..f01bcb0b631 100644 --- a/src/AggregateFunctions/fuzzers/CMakeLists.txt +++ b/src/AggregateFunctions/fuzzers/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable(aggregate_function_state_deserialization_fuzzer aggregate_function_state_deserialization_fuzzer.cpp ${SRCS}) -target_link_libraries(aggregate_function_state_deserialization_fuzzer PRIVATE clickhouse_aggregate_functions) +target_link_libraries(aggregate_function_state_deserialization_fuzzer PRIVATE clickhouse_aggregate_functions dbms) diff --git a/src/Core/fuzzers/CMakeLists.txt b/src/Core/fuzzers/CMakeLists.txt index c60ce0e097f..51db6fa0b53 100644 --- a/src/Core/fuzzers/CMakeLists.txt +++ b/src/Core/fuzzers/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable (names_and_types_fuzzer names_and_types_fuzzer.cpp) -target_link_libraries (names_and_types_fuzzer PRIVATE) +target_link_libraries (names_and_types_fuzzer PRIVATE dbms) diff --git a/src/DataTypes/fuzzers/CMakeLists.txt b/src/DataTypes/fuzzers/CMakeLists.txt index 9e5b1b3f673..8dedd3470e2 100644 --- a/src/DataTypes/fuzzers/CMakeLists.txt +++ b/src/DataTypes/fuzzers/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable(data_type_deserialization_fuzzer data_type_deserialization_fuzzer.cpp ${SRCS}) -target_link_libraries(data_type_deserialization_fuzzer PRIVATE clickhouse_aggregate_functions) +target_link_libraries(data_type_deserialization_fuzzer PRIVATE clickhouse_aggregate_functions dbms) diff --git a/src/Formats/fuzzers/CMakeLists.txt b/src/Formats/fuzzers/CMakeLists.txt index ee1a4fd4358..83aa5eb781a 100644 --- a/src/Formats/fuzzers/CMakeLists.txt +++ b/src/Formats/fuzzers/CMakeLists.txt @@ -1,2 +1,2 @@ clickhouse_add_executable(format_fuzzer format_fuzzer.cpp ${SRCS}) -target_link_libraries(format_fuzzer PRIVATE clickhouse_aggregate_functions) +target_link_libraries(format_fuzzer PRIVATE clickhouse_aggregate_functions dbms) diff --git a/src/Interpreters/fuzzers/CMakeLists.txt b/src/Interpreters/fuzzers/CMakeLists.txt index 3317bba7e30..174fae299b7 100644 --- a/src/Interpreters/fuzzers/CMakeLists.txt +++ b/src/Interpreters/fuzzers/CMakeLists.txt @@ -3,5 +3,6 @@ target_link_libraries(execute_query_fuzzer PRIVATE dbms clickhouse_table_functions clickhouse_aggregate_functions + clickhouse_functions clickhouse_dictionaries clickhouse_dictionaries_embedded) diff --git a/src/Storages/fuzzers/CMakeLists.txt b/src/Storages/fuzzers/CMakeLists.txt index 2c7c0c16fc2..719b9b77cd9 100644 --- a/src/Storages/fuzzers/CMakeLists.txt +++ b/src/Storages/fuzzers/CMakeLists.txt @@ -4,4 +4,4 @@ clickhouse_add_executable (mergetree_checksum_fuzzer mergetree_checksum_fuzzer.c target_link_libraries (mergetree_checksum_fuzzer PRIVATE dbms) clickhouse_add_executable (columns_description_fuzzer columns_description_fuzzer.cpp) -target_link_libraries (columns_description_fuzzer PRIVATE) +target_link_libraries (columns_description_fuzzer PRIVATE dbms) From 4e6180b50aaf3e39616750f8e4c6b114e0362e97 Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 30 Sep 2024 13:18:44 +0000 Subject: [PATCH 082/680] Resolve conflicts, better exception message --- src/Analyzer/Resolve/QueryAnalyzer.cpp | 8 ++++++-- src/Core/Settings.h | 2 +- src/Interpreters/ExpressionAnalyzer.cpp | 8 ++++++-- 3 files changed, 13 insertions(+), 5 deletions(-) diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index f3d77b0f091..56c96d41c6c 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -103,6 +103,8 @@ namespace Setting extern const SettingsBool single_join_prefer_left_table; extern const SettingsBool transform_null_in; extern const SettingsUInt64 use_structure_from_insertion_table_in_table_functions; + extern const SettingsBool allow_suspicious_types_in_group_by; + extern const SettingsBool allow_suspicious_types_in_order_by; } @@ -4100,7 +4102,7 @@ ProjectionNames QueryAnalyzer::resolveSortNodeList(QueryTreeNodePtr & sort_node_ void QueryAnalyzer::validateSortingKeyType(const DataTypePtr & sorting_key_type, const IdentifierResolveScope & scope) const { - if (scope.context->getSettingsRef().allow_suspicious_types_in_order_by) + if (scope.context->getSettingsRef()[Setting::allow_suspicious_types_in_order_by]) return; auto check = [](const IDataType & type) @@ -4109,6 +4111,7 @@ void QueryAnalyzer::validateSortingKeyType(const DataTypePtr & sorting_key_type, throw Exception( ErrorCodes::ILLEGAL_COLUMN, "Data types Variant/Dynamic are not allowed in ORDER BY keys, because it can lead to unexpected results. " + "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if its a JSON path subcolumn). " "Set setting allow_suspicious_types_in_order_by = 1 in order to allow it"); }; @@ -4189,7 +4192,7 @@ void QueryAnalyzer::resolveGroupByNode(QueryNode & query_node_typed, IdentifierR */ void QueryAnalyzer::validateGroupByKeyType(const DataTypePtr & group_by_key_type, const IdentifierResolveScope & scope) const { - if (scope.context->getSettingsRef().allow_suspicious_types_in_group_by) + if (scope.context->getSettingsRef()[Setting::allow_suspicious_types_in_group_by]) return; auto check = [](const IDataType & type) @@ -4198,6 +4201,7 @@ void QueryAnalyzer::validateGroupByKeyType(const DataTypePtr & group_by_key_type throw Exception( ErrorCodes::ILLEGAL_COLUMN, "Data types Variant/Dynamic are not allowed in GROUP BY keys, because it can lead to unexpected results. " + "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if its a JSON path subcolumn). " "Set setting allow_suspicious_types_in_group_by = 1 in order to allow it"); }; diff --git a/src/Core/Settings.h b/src/Core/Settings.h index bc2d0b423c1..5909ab6314c 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -156,4 +156,4 @@ struct Settings private: std::unique_ptr impl; }; -} \ No newline at end of file +} diff --git a/src/Interpreters/ExpressionAnalyzer.cpp b/src/Interpreters/ExpressionAnalyzer.cpp index dc7dca712a0..9a09bf8e16f 100644 --- a/src/Interpreters/ExpressionAnalyzer.cpp +++ b/src/Interpreters/ExpressionAnalyzer.cpp @@ -106,6 +106,8 @@ namespace Setting extern const SettingsBool query_plan_aggregation_in_order; extern const SettingsBool query_plan_read_in_order; extern const SettingsUInt64 use_index_for_in_with_subqueries_max_values; + extern const SettingsBool allow_suspicious_types_in_group_by; + extern const SettingsBool allow_suspicious_types_in_order_by; } @@ -1409,7 +1411,7 @@ bool SelectQueryExpressionAnalyzer::appendGroupBy(ExpressionActionsChain & chain void SelectQueryExpressionAnalyzer::validateGroupByKeyType(const DB::DataTypePtr & key_type) const { - if (getContext()->getSettingsRef().allow_suspicious_types_in_group_by) + if (getContext()->getSettingsRef()[Setting::allow_suspicious_types_in_group_by]) return; auto check = [](const IDataType & type) @@ -1418,6 +1420,7 @@ void SelectQueryExpressionAnalyzer::validateGroupByKeyType(const DB::DataTypePtr throw Exception( ErrorCodes::ILLEGAL_COLUMN, "Data types Variant/Dynamic are not allowed in GROUP BY keys, because it can lead to unexpected results. " + "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if its a JSON path subcolumn). " "Set setting allow_suspicious_types_in_group_by = 1 in order to allow it"); }; @@ -1692,7 +1695,7 @@ ActionsAndProjectInputsFlagPtr SelectQueryExpressionAnalyzer::appendOrderBy( void SelectQueryExpressionAnalyzer::validateOrderByKeyType(const DataTypePtr & key_type) const { - if (getContext()->getSettingsRef().allow_suspicious_types_in_order_by) + if (getContext()->getSettingsRef()[Setting::allow_suspicious_types_in_order_by]) return; auto check = [](const IDataType & type) @@ -1701,6 +1704,7 @@ void SelectQueryExpressionAnalyzer::validateOrderByKeyType(const DataTypePtr & k throw Exception( ErrorCodes::ILLEGAL_COLUMN, "Data types Variant/Dynamic are not allowed in ORDER BY keys, because it can lead to unexpected results. " + "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if its a JSON path subcolumn). " "Set setting allow_suspicious_types_in_order_by = 1 in order to allow it"); }; From 11c3c0de2447e5fcab999b13d0539cd074f3831d Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 30 Sep 2024 13:22:34 +0000 Subject: [PATCH 083/680] Even better exception message --- src/Analyzer/Resolve/QueryAnalyzer.cpp | 6 ++++-- src/Interpreters/ExpressionAnalyzer.cpp | 6 ++++-- src/Storages/KeyDescription.cpp | 5 ++++- 3 files changed, 12 insertions(+), 5 deletions(-) diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index 56c96d41c6c..7dc1d99efd0 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -4111,7 +4111,8 @@ void QueryAnalyzer::validateSortingKeyType(const DataTypePtr & sorting_key_type, throw Exception( ErrorCodes::ILLEGAL_COLUMN, "Data types Variant/Dynamic are not allowed in ORDER BY keys, because it can lead to unexpected results. " - "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if its a JSON path subcolumn). " + "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if " + "its a JSON path subcolumn) or casting this column to a specific data type. " "Set setting allow_suspicious_types_in_order_by = 1 in order to allow it"); }; @@ -4201,7 +4202,8 @@ void QueryAnalyzer::validateGroupByKeyType(const DataTypePtr & group_by_key_type throw Exception( ErrorCodes::ILLEGAL_COLUMN, "Data types Variant/Dynamic are not allowed in GROUP BY keys, because it can lead to unexpected results. " - "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if its a JSON path subcolumn). " + "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if " + "its a JSON path subcolumn) or casting this column to a specific data type. " "Set setting allow_suspicious_types_in_group_by = 1 in order to allow it"); }; diff --git a/src/Interpreters/ExpressionAnalyzer.cpp b/src/Interpreters/ExpressionAnalyzer.cpp index 9a09bf8e16f..12e769f249a 100644 --- a/src/Interpreters/ExpressionAnalyzer.cpp +++ b/src/Interpreters/ExpressionAnalyzer.cpp @@ -1420,7 +1420,8 @@ void SelectQueryExpressionAnalyzer::validateGroupByKeyType(const DB::DataTypePtr throw Exception( ErrorCodes::ILLEGAL_COLUMN, "Data types Variant/Dynamic are not allowed in GROUP BY keys, because it can lead to unexpected results. " - "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if its a JSON path subcolumn). " + "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if " + "its a JSON path subcolumn) or casting this column to a specific data type. " "Set setting allow_suspicious_types_in_group_by = 1 in order to allow it"); }; @@ -1704,7 +1705,8 @@ void SelectQueryExpressionAnalyzer::validateOrderByKeyType(const DataTypePtr & k throw Exception( ErrorCodes::ILLEGAL_COLUMN, "Data types Variant/Dynamic are not allowed in ORDER BY keys, because it can lead to unexpected results. " - "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if its a JSON path subcolumn). " + "Consider using a subcolumn with a specific data type instead (for example 'column.Int64' or 'json.some.path.:Int64' if " + "its a JSON path subcolumn) or casting this column to a specific data type. " "Set setting allow_suspicious_types_in_order_by = 1 in order to allow it"); }; diff --git a/src/Storages/KeyDescription.cpp b/src/Storages/KeyDescription.cpp index bb0b6d3542d..5c0449612e7 100644 --- a/src/Storages/KeyDescription.cpp +++ b/src/Storages/KeyDescription.cpp @@ -155,7 +155,10 @@ KeyDescription KeyDescription::getSortingKeyFromAST( auto check = [&](const IDataType & type) { if (isDynamic(type) || isVariant(type)) - throw Exception(ErrorCodes::DATA_TYPE_CANNOT_BE_USED_IN_KEY, "Column with type Variant/Dynamic is not allowed in key expression"); + throw Exception( + ErrorCodes::DATA_TYPE_CANNOT_BE_USED_IN_KEY, + "Column with type Variant/Dynamic is not allowed in key expression. Consider using a subcolumn with a specific data " + "type instead (for example 'column.Int64' or 'json.some.path.:Int64' if its a JSON path subcolumn) or casting this column to a specific data type"); }; check(*result.data_types.back()); From dda32963fdd399c2c614b2cb630fb714549e2804 Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 30 Sep 2024 13:57:19 +0000 Subject: [PATCH 084/680] Fix tests --- src/Core/SettingsChangesHistory.cpp | 6 ++---- .../03096_variant_in_primary_key.reference | 4 ---- .../0_stateless/03096_variant_in_primary_key.sql | 8 -------- .../03231_dynamic_incomplete_type_insert_bug.sql | 1 + .../03231_dynamic_not_safe_primary_key.reference | 0 .../03231_dynamic_not_safe_primary_key.sql | 11 ----------- .../0_stateless/03231_dynamic_uniq_group_by.sql | 2 ++ 7 files changed, 5 insertions(+), 27 deletions(-) delete mode 100644 tests/queries/0_stateless/03096_variant_in_primary_key.reference delete mode 100644 tests/queries/0_stateless/03096_variant_in_primary_key.sql delete mode 100644 tests/queries/0_stateless/03231_dynamic_not_safe_primary_key.reference delete mode 100644 tests/queries/0_stateless/03231_dynamic_not_safe_primary_key.sql diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 21a42b970f2..7bc9517a6a6 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -72,6 +72,8 @@ static std::initializer_list Date: Mon, 30 Sep 2024 14:56:34 +0000 Subject: [PATCH 085/680] ignore encoding errors in fuzzers output --- tests/fuzz/runner.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index deb219baff9..6f229725d4e 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -133,6 +133,7 @@ def run_fuzzer(fuzzer: str): text=True, check=True, shell=True, + errors='replace', ) except subprocess.CalledProcessError as e: # print("Command failed with error:", e) From 07fd719c8b2be80d08f088c2849a5fc150b98bc5 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Mon, 30 Sep 2024 15:03:00 +0000 Subject: [PATCH 086/680] Automatic style fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 6f229725d4e..e6eff430d1b 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -133,7 +133,7 @@ def run_fuzzer(fuzzer: str): text=True, check=True, shell=True, - errors='replace', + errors="replace", ) except subprocess.CalledProcessError as e: # print("Command failed with error:", e) From 46ada08197d1dab15116485cf85739a710685b5e Mon Sep 17 00:00:00 2001 From: vdimir Date: Mon, 30 Sep 2024 16:27:27 +0000 Subject: [PATCH 087/680] fix tests --- src/Interpreters/InterpreterSelectQuery.cpp | 3 +- src/Planner/PlannerJoinTree.cpp | 3 +- src/Processors/QueryPlan/JoinStep.cpp | 24 +++++++++-- src/Processors/QueryPlan/JoinStep.h | 4 +- .../0_stateless/00826_cross_to_inner_join.sql | 3 +- .../01107_join_right_table_totals.reference | 7 ++++ .../01107_join_right_table_totals.sql | 10 ++++- .../01881_join_on_conditions_hash.sql.j2 | 10 ++--- ...oin_with_nullable_lowcardinality_crash.sql | 5 ++- .../0_stateless/02282_array_distance.sql | 12 ++++-- .../02381_join_dup_columns_in_plan.reference | 1 - .../0_stateless/02461_join_lc_issue_42380.sql | 3 +- .../02514_analyzer_drop_join_on.reference | 1 - .../02835_join_step_explain.reference | 10 ++--- ...filter_push_down_equivalent_sets.reference | 40 ++++++++++++++----- ..._join_filter_push_down_equivalent_sets.sql | 40 ++++++++++++++----- .../03038_recursive_cte_postgres_4.reference | 4 +- .../03038_recursive_cte_postgres_4.sql | 4 +- ...03130_convert_outer_join_to_inner_join.sql | 13 ++++-- ...ter_push_down_equivalent_columns.reference | 3 +- .../03236_squashing_high_memory.sql | 1 + 21 files changed, 145 insertions(+), 56 deletions(-) diff --git a/src/Interpreters/InterpreterSelectQuery.cpp b/src/Interpreters/InterpreterSelectQuery.cpp index 547f8d63c7f..c830d95eada 100644 --- a/src/Interpreters/InterpreterSelectQuery.cpp +++ b/src/Interpreters/InterpreterSelectQuery.cpp @@ -1886,7 +1886,8 @@ void InterpreterSelectQuery::executeImpl(QueryPlan & query_plan, std::optional

setStepDescription(fmt::format("JOIN {}", expressions.join->pipelineType())); std::vector plans; diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp index 3c540c3ef81..720f0a380ab 100644 --- a/src/Planner/PlannerJoinTree.cpp +++ b/src/Planner/PlannerJoinTree.cpp @@ -1641,7 +1641,8 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ settings[Setting::max_block_size], settings[Setting::max_threads], outer_scope_columns, - false /*optimize_read_in_order*/); + false /*optimize_read_in_order*/, + true /*optimize_skip_unused_shards*/); join_step->inner_table_selection_mode = settings[Setting::query_plan_join_inner_table_selection]; join_step->setStepDescription(fmt::format("JOIN {}", join_pipeline_type)); diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index 9fdfeedb111..8365af4e589 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -92,12 +92,14 @@ JoinStep::JoinStep( size_t max_block_size_, size_t max_streams_, NameSet required_output_, - bool keep_left_read_in_order_) + bool keep_left_read_in_order_, + bool use_new_analyzer_) : join(std::move(join_)) , max_block_size(max_block_size_) , max_streams(max_streams_) , required_output(std::move(required_output_)) , keep_left_read_in_order(keep_left_read_in_order_) + , use_new_analyzer(use_new_analyzer_) { updateInputStreams(DataStreams{left_stream_, right_stream_}); } @@ -130,6 +132,9 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines keep_left_read_in_order, &processors); + if (!use_new_analyzer) + return pipeline; + const auto & result_names = pipeline->getHeader().getNames(); size_t prefix_size = getPrefixLength(rhs_names, result_names); if (!columns_to_remove.empty() || (0 < prefix_size && prefix_size < result_names.size())) @@ -184,19 +189,30 @@ void JoinStep::updateOutputStream() const auto & header = swap_streams ? input_streams[1].header : input_streams[0].header; Block result_header = JoiningTransform::transformHeader(header, join); - join_algorithm_header = result_header; + + if (!use_new_analyzer) + { + if (swap_streams) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot swap streams without new analyzer"); + output_stream = DataStream { .header = result_header }; + return; + } + + if (swap_streams) result_header = rotateBlock(result_header, input_streams[1].header); columns_to_remove.clear(); for (size_t i = 0; i < result_header.columns(); ++i) { - if (required_output.empty()) - break; if (!required_output.contains(result_header.getByPosition(i).name)) columns_to_remove.insert(i); } + /// Do not remove all columns, keep at least one + if (!columns_to_remove.empty() && columns_to_remove.size() == result_header.columns()) + columns_to_remove.erase(columns_to_remove.begin()); + result_header.erase(columns_to_remove); output_stream = DataStream { .header = result_header }; } diff --git a/src/Processors/QueryPlan/JoinStep.h b/src/Processors/QueryPlan/JoinStep.h index 30b20a0d3a5..b0947cb6be7 100644 --- a/src/Processors/QueryPlan/JoinStep.h +++ b/src/Processors/QueryPlan/JoinStep.h @@ -21,7 +21,8 @@ public: size_t max_block_size_, size_t max_streams_, NameSet required_output_, - bool keep_left_read_in_order_); + bool keep_left_read_in_order_, + bool use_new_analyzer_); String getName() const override { return "Join"; } @@ -53,6 +54,7 @@ private: NameSet required_output; std::set columns_to_remove; bool keep_left_read_in_order; + bool use_new_analyzer = false; }; /// Special step for the case when Join is already filled. diff --git a/tests/queries/0_stateless/00826_cross_to_inner_join.sql b/tests/queries/0_stateless/00826_cross_to_inner_join.sql index e9f9e13e2d3..f81832a4109 100644 --- a/tests/queries/0_stateless/00826_cross_to_inner_join.sql +++ b/tests/queries/0_stateless/00826_cross_to_inner_join.sql @@ -48,7 +48,8 @@ SELECT * FROM t1_00826, t2_00826 where t1_00826.a = t2_00826.a; SELECT '--- comma nullable ---'; SELECT * FROM t1_00826, t2_00826 where t1_00826.b = t2_00826.b; SELECT '--- comma and or ---'; -SELECT * FROM t1_00826, t2_00826 where t1_00826.a = t2_00826.a AND (t2_00826.b IS NULL OR t2_00826.b < 2); +SELECT * FROM t1_00826, t2_00826 where t1_00826.a = t2_00826.a AND (t2_00826.b IS NULL OR t2_00826.b < 2) +ORDER BY ALL; SELECT '--- cross ---'; diff --git a/tests/queries/0_stateless/01107_join_right_table_totals.reference b/tests/queries/0_stateless/01107_join_right_table_totals.reference index daf503b776d..aa569ff9331 100644 --- a/tests/queries/0_stateless/01107_join_right_table_totals.reference +++ b/tests/queries/0_stateless/01107_join_right_table_totals.reference @@ -18,28 +18,35 @@ 0 0 0 0 +- 1 1 1 1 0 0 +- 1 1 1 1 0 0 +- 1 1 1 1 0 0 +- 1 1 1 1 0 0 +- 1 1 0 0 +- 1 foo 1 1 300 0 foo 1 0 300 +- 1 100 1970-01-01 1 100 1970-01-01 1 100 1970-01-01 1 200 1970-01-02 1 200 1970-01-02 1 100 1970-01-01 diff --git a/tests/queries/0_stateless/01107_join_right_table_totals.sql b/tests/queries/0_stateless/01107_join_right_table_totals.sql index ad8954d5d70..7e549282489 100644 --- a/tests/queries/0_stateless/01107_join_right_table_totals.sql +++ b/tests/queries/0_stateless/01107_join_right_table_totals.sql @@ -64,39 +64,47 @@ USING (id); INSERT INTO t VALUES (1, 100, '1970-01-01'), (1, 200, '1970-01-02'); +SELECT '-'; SELECT * FROM (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id) l LEFT JOIN (SELECT item_id FROM t ) r ON l.item_id = r.item_id; +SELECT '-'; SELECT * FROM (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id) l RIGHT JOIN (SELECT item_id FROM t ) r ON l.item_id = r.item_id; +SELECT '-'; SELECT * FROM (SELECT item_id FROM t) l LEFT JOIN (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id ) r ON l.item_id = r.item_id; +SELECT '-'; SELECT * FROM (SELECT item_id FROM t) l RIGHT JOIN (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id ) r ON l.item_id = r.item_id; +SELECT '-'; SELECT * FROM (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id) l LEFT JOIN (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id ) r ON l.item_id = r.item_id; +SELECT '-'; SELECT * FROM (SELECT item_id, 'foo' AS key, 1 AS val FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id) l LEFT JOIN (SELECT item_id, sum(price_sold) AS val FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id ) r ON l.item_id = r.item_id; +SELECT '-'; SELECT * FROM (SELECT * FROM t GROUP BY item_id, price_sold, date WITH TOTALS ORDER BY item_id, price_sold, date) l LEFT JOIN (SELECT * FROM t GROUP BY item_id, price_sold, date WITH TOTALS ORDER BY item_id, price_sold, date ) r -ON l.item_id = r.item_id; +ON l.item_id = r.item_id +ORDER BY ALL; DROP TABLE t; diff --git a/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 b/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 index c2d85cefb18..c13722f431a 100644 --- a/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 +++ b/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 @@ -75,7 +75,7 @@ SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t2.key; -- { serverErro SELECT * FROM t1 JOIN t2_nullable as t2 ON t2.key == t2.key2 AND (t1.id == t2.id OR isNull(t2.key2)); -- { serverError 403 } SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 OR t1.id == t2.id; -- { serverError 403 } SELECT * FROM t1 JOIN t2 ON (t2.key == t2.key2 AND (t1.key == t1.key2 AND t1.key != 'XXX' OR t1.id == t2.id)) AND t1.id == t2.id; -- { serverError 403 } -SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 AND t1.key == t1.key2 AND t1.key != 'XXX' AND t1.id == t2.id OR t2.key == t2.key2 AND t1.id == t2.id AND t1.id == t2.id; +SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 AND t1.key == t1.key2 AND t1.key != 'XXX' AND t1.id == t2.id OR t2.key == t2.key2 AND t1.id == t2.id AND t1.id == t2.id ORDER BY ALL; -- non-equi condition containing columns from different tables doesn't supported yet SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t1.id >= t2.id; -- { serverError 403 } SELECT * FROM t1 INNER ANY JOIN t2 ON t1.id == t2.id AND t2.key == t2.key2 AND t1.key == t1.key2 AND t1.id >= length(t2.key); -- { serverError 403 } @@ -89,10 +89,10 @@ SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.idd and (t1.id == t22.id OR t22 SELECT 't22', * FROM t1 JOIN t22 ON (t22.key == t22.key2 OR t1.id == t22.id) and t1.id == t22.idd; -- { serverError 403 } SELECT 't22', * FROM t1 JOIN t22 ON (t1.id == t22.id OR t22.key == t22.key2) and t1.id == t22.idd; -- { serverError 403 } SELECT 't22', * FROM t1 JOIN t22 ON (t1.id == t22.id OR t22.key == t22.key2) and (t1.id == t22.idd AND (t1.key2 = 'a1' OR t1.key2 = 'a2' OR t1.key2 = 'a3' OR t1.key2 = 'a4' OR t1.key2 = 'a5' OR t1.key2 = 'a6' OR t1.key2 = 'a7' OR t1.key2 = 'a8' OR t1.key2 = 'a9' OR t1.key2 = 'a10' OR t1.key2 = 'a11' OR t1.key2 = 'a12' OR t1.key2 = 'a13' OR t1.key2 = 'a14' OR t1.key2 = 'a15' OR t1.key2 = 'a16' OR t1.key2 = 'a17' OR t1.key2 = 'a18' OR t1.key2 = 'a19' OR t1.key2 = '111')); -- { serverError 403 } -SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.idd and t22.key == t22.key2 OR t1.id == t22.idd and t1.id == t22.id; -SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.idd and t1.id == t22.id OR t1.id == t22.idd and t22.key == t22.key2; -SELECT 't22', * FROM t1 JOIN t22 ON t22.key == t22.key2 and t1.id == t22.idd OR t1.id == t22.id and t1.id == t22.idd; -SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.id and t1.id == t22.idd OR t22.key == t22.key2 and t1.id == t22.idd; +SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.idd and t22.key == t22.key2 OR t1.id == t22.idd and t1.id == t22.id ORDER BY ALL; +SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.idd and t1.id == t22.id OR t1.id == t22.idd and t22.key == t22.key2 ORDER BY ALL; +SELECT 't22', * FROM t1 JOIN t22 ON t22.key == t22.key2 and t1.id == t22.idd OR t1.id == t22.id and t1.id == t22.idd ORDER BY ALL; +SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.id and t1.id == t22.idd OR t22.key == t22.key2 and t1.id == t22.idd ORDER BY ALL; {% endfor -%} diff --git a/tests/queries/0_stateless/02245_join_with_nullable_lowcardinality_crash.sql b/tests/queries/0_stateless/02245_join_with_nullable_lowcardinality_crash.sql index abc2ee41402..c3c84ebaded 100644 --- a/tests/queries/0_stateless/02245_join_with_nullable_lowcardinality_crash.sql +++ b/tests/queries/0_stateless/02245_join_with_nullable_lowcardinality_crash.sql @@ -12,8 +12,9 @@ CREATE TABLE without_nullable insert into with_nullable values(0,'f'),(0,'usa'); insert into without_nullable values(0,'usa'),(0,'us2a'); -select if(t0.country is null ,t2.country,t0.country) "country" -from without_nullable t0 right outer join with_nullable t2 on t0.country=t2.country; +select if(t0.country is null ,t2.country,t0.country) "country" +from without_nullable t0 right outer join with_nullable t2 on t0.country=t2.country +ORDER BY 1 DESC; drop table with_nullable; drop table without_nullable; diff --git a/tests/queries/0_stateless/02282_array_distance.sql b/tests/queries/0_stateless/02282_array_distance.sql index 2cca853fd67..85abc8fa381 100644 --- a/tests/queries/0_stateless/02282_array_distance.sql +++ b/tests/queries/0_stateless/02282_array_distance.sql @@ -48,7 +48,8 @@ SELECT L2SquaredDistance(v1.v, v2.v), cosineDistance(v1.v, v2.v) FROM vec2 v1, vec2 v2 -WHERE length(v1.v) == length(v2.v); +WHERE length(v1.v) == length(v2.v) +ORDER BY ALL; INSERT INTO vec2f VALUES (1, [100, 200, 0]), (2, [888, 777, 666]), (3, range(1, 35, 1)), (4, range(3, 37, 1)), (5, range(1, 135, 1)), (6, range(3, 137, 1)); SELECT @@ -61,7 +62,8 @@ SELECT L2SquaredDistance(v1.v, v2.v), cosineDistance(v1.v, v2.v) FROM vec2f v1, vec2f v2 -WHERE length(v1.v) == length(v2.v); +WHERE length(v1.v) == length(v2.v) +ORDER BY ALL; INSERT INTO vec2d VALUES (1, [100, 200, 0]), (2, [888, 777, 666]), (3, range(1, 35, 1)), (4, range(3, 37, 1)), (5, range(1, 135, 1)), (6, range(3, 137, 1)); SELECT @@ -74,7 +76,8 @@ SELECT L2SquaredDistance(v1.v, v2.v), cosineDistance(v1.v, v2.v) FROM vec2d v1, vec2d v2 -WHERE length(v1.v) == length(v2.v); +WHERE length(v1.v) == length(v2.v) +ORDER BY ALL; SELECT v1.id, @@ -86,7 +89,8 @@ SELECT L2SquaredDistance(v1.v, v2.v), cosineDistance(v1.v, v2.v) FROM vec2f v1, vec2d v2 -WHERE length(v1.v) == length(v2.v); +WHERE length(v1.v) == length(v2.v) +ORDER BY ALL; SELECT L1Distance([0, 0], [1]); -- { serverError SIZES_OF_ARRAYS_DONT_MATCH } SELECT L2Distance([1, 2], (3,4)); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } diff --git a/tests/queries/0_stateless/02381_join_dup_columns_in_plan.reference b/tests/queries/0_stateless/02381_join_dup_columns_in_plan.reference index 365725f8ffe..90aab0a0eb2 100644 --- a/tests/queries/0_stateless/02381_join_dup_columns_in_plan.reference +++ b/tests/queries/0_stateless/02381_join_dup_columns_in_plan.reference @@ -148,7 +148,6 @@ Header: key String value String Join Header: __table1.key String - __table3.key String __table3.value String Sorting Header: __table1.key String diff --git a/tests/queries/0_stateless/02461_join_lc_issue_42380.sql b/tests/queries/0_stateless/02461_join_lc_issue_42380.sql index f0ecbf64e58..8b5c6846bd0 100644 --- a/tests/queries/0_stateless/02461_join_lc_issue_42380.sql +++ b/tests/queries/0_stateless/02461_join_lc_issue_42380.sql @@ -9,4 +9,5 @@ CREATE TABLE t2__fuzz_47 (id LowCardinality(Int16)) ENGINE = MergeTree() ORDER B INSERT INTO t1__fuzz_13 VALUES (1); INSERT INTO t2__fuzz_47 VALUES (1); -SELECT * FROM t1__fuzz_13 FULL OUTER JOIN t2__fuzz_47 ON 1 = 2; +SELECT * FROM t1__fuzz_13 FULL OUTER JOIN t2__fuzz_47 ON 1 = 2 +ORDER BY ALL; diff --git a/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference b/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference index 59983fff778..d407a4c7985 100644 --- a/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference +++ b/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference @@ -50,7 +50,6 @@ Header: a2 String d2 String Join (JOIN FillRightFirst) Header: __table1.a2 String - __table1.k UInt64 __table4.d2 String Expression (DROP unused columns after JOIN) Header: __table1.a2 String diff --git a/tests/queries/0_stateless/02835_join_step_explain.reference b/tests/queries/0_stateless/02835_join_step_explain.reference index 31205956662..2f641d4aa44 100644 --- a/tests/queries/0_stateless/02835_join_step_explain.reference +++ b/tests/queries/0_stateless/02835_join_step_explain.reference @@ -58,18 +58,16 @@ Header: id UInt64 Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value_1 String : 1 INPUT : 2 -> __table2.value_1 String : 2 - INPUT :: 3 -> __table2.value_2 UInt64 : 3 - INPUT : 4 -> __table2.id UInt64 : 4 - ALIAS __table1.id :: 0 -> id UInt64 : 5 + INPUT : 3 -> __table2.id UInt64 : 3 + ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value_1 :: 1 -> value_1 String : 0 ALIAS __table2.value_1 :: 2 -> rhs.value_1 String : 1 - ALIAS __table2.id :: 4 -> rhs.id UInt64 : 2 -Positions: 5 0 2 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 0 2 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value_1 String __table2.value_1 String - __table2.value_2 UInt64 __table2.id UInt64 Type: INNER Strictness: ASOF diff --git a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference index 80f4e309505..c98a98b236c 100644 --- a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference +++ b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference @@ -2,7 +2,9 @@ EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5; +WHERE lhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; Expression ((Project names + (Projection + ))) Header: id UInt64 rhs.id UInt64 @@ -69,7 +71,9 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5; +WHERE rhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right'; +; Expression ((Project names + (Projection + ))) Header: id UInt64 rhs.id UInt64 @@ -136,7 +140,9 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 AND rhs.id = 6; +WHERE lhs.id = 5 AND rhs.id = 6 +SETTINGS query_plan_join_inner_table_selection = 'right' +; Expression ((Project names + (Projection + ))) Header: id UInt64 rhs.id UInt64 @@ -206,7 +212,9 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5; +WHERE lhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; Expression ((Project names + (Projection + ))) Header: id UInt64 rhs.id UInt64 @@ -273,7 +281,9 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5; +WHERE rhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; Expression ((Project names + Projection)) Header: id UInt64 rhs.id UInt64 @@ -347,7 +357,9 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5; +WHERE lhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; Expression ((Project names + Projection)) Header: id UInt64 rhs.id UInt64 @@ -421,7 +433,9 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5; +WHERE rhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; Expression ((Project names + (Projection + ))) Header: id UInt64 rhs.id UInt64 @@ -488,7 +502,9 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5; +WHERE lhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; Expression ((Project names + Projection)) Header: id UInt64 rhs.id UInt64 @@ -562,7 +578,9 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5; +WHERE rhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; Expression ((Project names + Projection)) Header: id UInt64 rhs.id UInt64 @@ -636,7 +654,9 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 AND rhs.id = 6; +WHERE lhs.id = 5 AND rhs.id = 6 +SETTINGS query_plan_join_inner_table_selection = 'right' +; Expression ((Project names + Projection)) Header: id UInt64 rhs.id UInt64 diff --git a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql index e1a13d1ce71..d6dcc34c796 100644 --- a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql +++ b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql @@ -22,7 +22,9 @@ INSERT INTO test_table_2 SELECT number, number FROM numbers(10); EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5; +WHERE lhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT '--'; @@ -33,7 +35,9 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5; +WHERE rhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right'; +; SELECT '--'; @@ -44,7 +48,9 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 AND rhs.id = 6; +WHERE lhs.id = 5 AND rhs.id = 6 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE lhs.id = 5 AND rhs.id = 6; @@ -53,7 +59,9 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5; +WHERE lhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT '--'; @@ -64,7 +72,9 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5; +WHERE rhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT '--'; @@ -75,7 +85,9 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5; +WHERE lhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT '--'; @@ -86,7 +98,9 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5; +WHERE rhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT '--'; @@ -97,7 +111,9 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5; +WHERE lhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT '--'; @@ -108,7 +124,9 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5; +WHERE rhs.id = 5 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT '--'; @@ -119,7 +137,9 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 AND rhs.id = 6; +WHERE lhs.id = 5 AND rhs.id = 6 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT '--'; diff --git a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference index cf070eebc38..7df38e855f6 100644 --- a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference +++ b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference @@ -52,7 +52,9 @@ WITH RECURSIVE search_graph AS ( FROM graph g, search_graph sg WHERE g.f = sg.t AND NOT is_cycle ) -SELECT * FROM search_graph; +SELECT * FROM search_graph +SETTINGS query_plan_join_inner_table_selection = 'right' +; 1 2 arc 1 -> 2 false [(1,2)] 1 3 arc 1 -> 3 false [(1,3)] 2 3 arc 2 -> 3 false [(2,3)] diff --git a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql index 7dad74893b9..d33ca7b078e 100644 --- a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql +++ b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql @@ -55,7 +55,9 @@ WITH RECURSIVE search_graph AS ( FROM graph g, search_graph sg WHERE g.f = sg.t AND NOT is_cycle ) -SELECT * FROM search_graph; +SELECT * FROM search_graph +SETTINGS query_plan_join_inner_table_selection = 'right' +; -- ordering by the path column has same effect as SEARCH DEPTH FIRST WITH RECURSIVE search_graph AS ( diff --git a/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql b/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql index b3d1827d98f..ddefc322b4f 100644 --- a/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql +++ b/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql @@ -22,7 +22,10 @@ SETTINGS index_granularity = 16 INSERT INTO test_table_1 VALUES (1, 'Value_1'), (2, 'Value_2'); INSERT INTO test_table_2 VALUES (2, 'Value_2'), (3, 'Value_3'); -EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE rhs.id != 0; + +EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE rhs.id != 0 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT '--'; @@ -30,7 +33,9 @@ SELECT * FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs. SELECT '--'; -EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE lhs.id != 0; +EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE lhs.id != 0 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT '--'; @@ -38,7 +43,9 @@ SELECT * FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs SELECT '--'; -EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE lhs.id != 0 AND rhs.id != 0; +EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE lhs.id != 0 AND rhs.id != 0 +SETTINGS query_plan_join_inner_table_selection = 'right' +; SELECT '--'; diff --git a/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.reference b/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.reference index 7058d36aaf9..1c82e76cc65 100644 --- a/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.reference +++ b/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.reference @@ -65,8 +65,7 @@ SELECT name FROM users RIGHT JOIN users2 USING name WHERE users2.name ='Alice'; Expression ((Project names + (Projection + ))) Header: name String Join (JOIN FillRightFirst) - Header: __table1.name String - __table2.name String + Header: __table2.name String Filter (( + Change column names to column identifiers)) Header: __table1.name String ReadFromMergeTree (default.users) diff --git a/tests/queries/0_stateless/03236_squashing_high_memory.sql b/tests/queries/0_stateless/03236_squashing_high_memory.sql index f6e5dbdef03..eeb3ae85e84 100644 --- a/tests/queries/0_stateless/03236_squashing_high_memory.sql +++ b/tests/queries/0_stateless/03236_squashing_high_memory.sql @@ -11,6 +11,7 @@ CREATE TABLE id_values ENGINE MergeTree ORDER BY id1 AS SELECT arrayJoin(range(500000)) AS id1, arrayJoin(range(1000)) AS id2; SET max_memory_usage = '1G'; +SET query_plan_join_inner_table_selection = 'right'; CREATE TABLE test_table ENGINE MergeTree ORDER BY id AS SELECT id_values.id1 AS id, From e28171d2b6ea1ffb8783f6141f59763684b4dfd4 Mon Sep 17 00:00:00 2001 From: vdimir Date: Tue, 1 Oct 2024 11:38:38 +0000 Subject: [PATCH 088/680] fix clang tidy --- src/Processors/Transforms/ColumnPermuteTransform.cpp | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/Processors/Transforms/ColumnPermuteTransform.cpp b/src/Processors/Transforms/ColumnPermuteTransform.cpp index 2921bcac177..169dd2dc67e 100644 --- a/src/Processors/Transforms/ColumnPermuteTransform.cpp +++ b/src/Processors/Transforms/ColumnPermuteTransform.cpp @@ -11,8 +11,8 @@ void applyPermutation(std::vector & data, const std::vector & permuta { std::vector res; res.reserve(permutation.size()); - for (size_t i = 0; i < permutation.size(); ++i) - res.emplace_back(std::move(data[permutation[i]])); + for (size_t i : permutation) + res.emplace_back(std::move(data[i])); data = std::move(res); } From 335e1847fee258ce75639dbdce34fd0bdf5b040a Mon Sep 17 00:00:00 2001 From: vdimir Date: Tue, 1 Oct 2024 11:39:51 +0000 Subject: [PATCH 089/680] up src/Core/SettingsChangesHistory.cpp --- src/Core/SettingsChangesHistory.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index d1f90f378e6..54c9f53f41b 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -71,10 +71,10 @@ static std::initializer_list Date: Tue, 1 Oct 2024 12:00:11 +0000 Subject: [PATCH 090/680] randomize only latest version settings --- tests/integration/helpers/cluster.py | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index f5f87947c0f..1687f049b25 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -4592,7 +4592,12 @@ class ClickHouseInstance: if len(self.custom_dictionaries_paths): write_embedded_config("0_common_enable_dictionaries.xml", self.config_d_dir) - if self.randomize_settings and self.base_config_dir == DEFAULT_BASE_CONFIG_DIR: + if ( + self.randomize_settings + and self.image == "clickhouse/integration-test" + and self.tag == "latest" + and self.base_config_dir == DEFAULT_BASE_CONFIG_DIR + ): # If custom main config is used, do not apply random settings to it write_random_settings_config(Path(users_d_dir) / "0_random_settings.xml") From a1a571c45e43b767d4c2f2a7c4114020513882b9 Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 1 Oct 2024 12:59:46 +0000 Subject: [PATCH 091/680] Fix tests --- tests/queries/0_stateless/01825_new_type_json_10.sql | 1 + tests/queries/0_stateless/01825_new_type_json_11.sh | 6 +++--- tests/queries/0_stateless/01825_new_type_json_12.sh | 2 +- tests/queries/0_stateless/01825_new_type_json_13.sh | 2 +- tests/queries/0_stateless/01825_new_type_json_6.sh | 2 +- tests/queries/0_stateless/01825_new_type_json_7.sh | 2 +- tests/queries/0_stateless/01825_new_type_json_ghdata.sh | 2 +- tests/queries/0_stateless/01825_new_type_json_in_array.sql | 3 +++ .../0_stateless/01825_new_type_json_insert_select.sql | 2 ++ .../queries/0_stateless/02421_new_type_json_async_insert.sh | 2 +- .../0_stateless/03151_dynamic_type_scale_max_types.sql | 5 +++-- 11 files changed, 18 insertions(+), 11 deletions(-) diff --git a/tests/queries/0_stateless/01825_new_type_json_10.sql b/tests/queries/0_stateless/01825_new_type_json_10.sql index f586cc4477b..9aac35e2c88 100644 --- a/tests/queries/0_stateless/01825_new_type_json_10.sql +++ b/tests/queries/0_stateless/01825_new_type_json_10.sql @@ -1,6 +1,7 @@ -- Tags: no-fasttest SET allow_experimental_json_type = 1; +SET allow_suspicious_types_in_order_by = 1; DROP TABLE IF EXISTS t_json_10; CREATE TABLE t_json_10 (o JSON) ENGINE = Memory; diff --git a/tests/queries/0_stateless/01825_new_type_json_11.sh b/tests/queries/0_stateless/01825_new_type_json_11.sh index f448b7433ab..e9b90af4499 100755 --- a/tests/queries/0_stateless/01825_new_type_json_11.sh +++ b/tests/queries/0_stateless/01825_new_type_json_11.sh @@ -57,8 +57,8 @@ $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(obj)) as $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(arrayJoin(obj.key_1[]))) as path FROM t_json_11 order by path;" $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(arrayJoin(arrayJoin(obj.key_1[].key_3[])))) as path FROM t_json_11 order by path;" $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(arrayJoin(arrayJoin(arrayJoin(obj.key_1[].key_3[].key_4[]))))) as path FROM t_json_11 order by path;" -$CLICKHOUSE_CLIENT -q "SELECT obj FROM t_json_11 ORDER BY obj.id FORMAT JSONEachRow" -$CLICKHOUSE_CLIENT -q "SELECT obj.key_1[].key_3 FROM t_json_11 ORDER BY obj.id FORMAT JSONEachRow" -$CLICKHOUSE_CLIENT -q "SELECT obj.key_1[].key_3[].key_4[].key_5, obj.key_1[].key_3[].key_7 FROM t_json_11 ORDER BY obj.id" +$CLICKHOUSE_CLIENT -q "SELECT obj FROM t_json_11 ORDER BY obj.id FORMAT JSONEachRow" --allow_suspicious_types_in_order_by 1 +$CLICKHOUSE_CLIENT -q "SELECT obj.key_1[].key_3 FROM t_json_11 ORDER BY obj.id FORMAT JSONEachRow" --allow_suspicious_types_in_order_by 1 +$CLICKHOUSE_CLIENT -q "SELECT obj.key_1[].key_3[].key_4[].key_5, obj.key_1[].key_3[].key_7 FROM t_json_11 ORDER BY obj.id" --allow_suspicious_types_in_order_by 1 $CLICKHOUSE_CLIENT -q "DROP TABLE t_json_11;" diff --git a/tests/queries/0_stateless/01825_new_type_json_12.sh b/tests/queries/0_stateless/01825_new_type_json_12.sh index d7c938d7cd1..e3909787690 100755 --- a/tests/queries/0_stateless/01825_new_type_json_12.sh +++ b/tests/queries/0_stateless/01825_new_type_json_12.sh @@ -49,6 +49,6 @@ $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(arrayJoin $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(arrayJoin(arrayJoin(arrayJoin(obj.key_0[].key_1[].key_3[]))))) as path FROM t_json_12 order by path;" $CLICKHOUSE_CLIENT -q "SELECT obj FROM t_json_12 ORDER BY obj.id FORMAT JSONEachRow" --output_format_json_named_tuples_as_objects 1 $CLICKHOUSE_CLIENT -q "SELECT obj.key_0[].key_1[].key_3[].key_4, obj.key_0[].key_1[].key_3[].key_5, \ - obj.key_0[].key_1[].key_3[].key_6, obj.key_0[].key_1[].key_3[].key_7 FROM t_json_12 ORDER BY obj.id" + obj.key_0[].key_1[].key_3[].key_6, obj.key_0[].key_1[].key_3[].key_7 FROM t_json_12 ORDER BY obj.id" --allow_suspicious_types_in_order_by 1 $CLICKHOUSE_CLIENT -q "DROP TABLE t_json_12;" diff --git a/tests/queries/0_stateless/01825_new_type_json_13.sh b/tests/queries/0_stateless/01825_new_type_json_13.sh index 316e6890d5e..e7d9f556be7 100755 --- a/tests/queries/0_stateless/01825_new_type_json_13.sh +++ b/tests/queries/0_stateless/01825_new_type_json_13.sh @@ -45,6 +45,6 @@ $CLICKHOUSE_CLIENT -q "SELECT \ obj.key_1.key_2.key_3.key_4.key_5, \ obj.key_1.key_2.key_3.key_4.key_6, \ obj.key_1.key_2.key_3.key_4.key_7 \ -FROM t_json_13 ORDER BY obj.id" +FROM t_json_13 ORDER BY obj.id" --allow_suspicious_types_in_order_by 1 $CLICKHOUSE_CLIENT -q "DROP TABLE t_json_13;" diff --git a/tests/queries/0_stateless/01825_new_type_json_6.sh b/tests/queries/0_stateless/01825_new_type_json_6.sh index 6b9a7e71f50..a2102636c42 100755 --- a/tests/queries/0_stateless/01825_new_type_json_6.sh +++ b/tests/queries/0_stateless/01825_new_type_json_6.sh @@ -54,6 +54,6 @@ EOF $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(data)) as path FROM t_json_6 order by path;" $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(arrayJoin(data.out[]))) as path FROM t_json_6 order by path;" $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(arrayJoin(arrayJoin(data.out[].outputs[])))) as path FROM t_json_6 order by path;" -$CLICKHOUSE_CLIENT -q "SELECT data.key, data.out[].type, data.out[].value, data.out[].outputs[].index, data.out[].outputs[].n FROM t_json_6 ORDER BY data.key" +$CLICKHOUSE_CLIENT -q "SELECT data.key, data.out[].type, data.out[].value, data.out[].outputs[].index, data.out[].outputs[].n FROM t_json_6 ORDER BY data.key" --allow_suspicious_types_in_order_by 1 $CLICKHOUSE_CLIENT -q "DROP TABLE t_json_6;" diff --git a/tests/queries/0_stateless/01825_new_type_json_7.sh b/tests/queries/0_stateless/01825_new_type_json_7.sh index 36483175df6..b6ea46f5ff8 100755 --- a/tests/queries/0_stateless/01825_new_type_json_7.sh +++ b/tests/queries/0_stateless/01825_new_type_json_7.sh @@ -25,6 +25,6 @@ cat < Date: Tue, 1 Oct 2024 14:02:17 +0000 Subject: [PATCH 092/680] add timeout for every fuzzer --- tests/fuzz/runner.py | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index e6eff430d1b..cfd60d8f259 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -62,7 +62,7 @@ def process_error(error: str): report(error_source, error_reason, call_stack, test_unit) -def run_fuzzer(fuzzer: str): +def run_fuzzer(fuzzer: str, timeout: int): logging.info("Running fuzzer %s...", fuzzer) seed_corpus_dir = f"{fuzzer}.in" @@ -134,6 +134,7 @@ def run_fuzzer(fuzzer: str): check=True, shell=True, errors="replace", + timeout=timeout, ) except subprocess.CalledProcessError as e: # print("Command failed with error:", e) @@ -148,10 +149,16 @@ def main(): subprocess.check_call("ls -al", shell=True) + timeout = 30 + + match = re.search(r"(^|\s+)-max_total_time=(\d+)($|\s)", FUZZER_ARGS) + if match: + timeout += match.group(2) + with Path() as current: for fuzzer in current.iterdir(): if (current / fuzzer).is_file() and os.access(current / fuzzer, os.X_OK): - run_fuzzer(fuzzer) + run_fuzzer(fuzzer, timeout) if __name__ == "__main__": From 77e13544d6d5641a68a765c7e15f7af4b9bfec00 Mon Sep 17 00:00:00 2001 From: Igor Nikonov Date: Tue, 1 Oct 2024 14:03:05 +0000 Subject: [PATCH 093/680] Parallel relicas: use local plan for local replica by default --- src/Core/Settings.cpp | 2 +- src/Core/SettingsChangesHistory.cpp | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index d0ce90e6fdd..dfba3b128bb 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -965,7 +965,7 @@ namespace ErrorCodes 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, 0, "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. Value should be in range [128; 16384]", 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, parallel_replicas_local_plan, false, "Build local plan for local replica", 0) \ + M(Bool, parallel_replicas_local_plan, true, "If true, use local plan for local replica in a query with parallel replicas, otherwise all replicas in a used cluster considered as remote", 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 560f144866b..92cf586b9c6 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -67,6 +67,7 @@ static std::initializer_list Date: Tue, 1 Oct 2024 14:26:44 +0000 Subject: [PATCH 094/680] w --- src/Parsers/IAST.cpp | 1 - src/Processors/QueryPlan/JoinStep.cpp | 2 +- 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/src/Parsers/IAST.cpp b/src/Parsers/IAST.cpp index d6daf9bd78b..ad95f69b220 100644 --- a/src/Parsers/IAST.cpp +++ b/src/Parsers/IAST.cpp @@ -174,7 +174,6 @@ String IAST::formatWithPossiblyHidingSensitiveData( IdentifierQuotingRule identifier_quoting_rule, IdentifierQuotingStyle identifier_quoting_style) const { - WriteBufferFromOwnString buf; FormatSettings settings(buf, one_line); settings.show_secrets = show_secrets; diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index 8365af4e589..2d7dd689149 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -151,7 +151,7 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines } column_permutation.resize(n); - pipeline->addSimpleTransform([column_perm = std::move(column_permutation)](const Block & header) + pipeline->addSimpleTransform([column_perm = std::move(column_permutation)](const Block & header) mutable { return std::make_shared(header, std::move(column_perm)); }); From a7da67069ab92c06e069d0f91132b8b12e0c2eda Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Tue, 1 Oct 2024 15:49:26 +0000 Subject: [PATCH 095/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index cfd60d8f259..ccc5a4b7465 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -153,7 +153,7 @@ def main(): match = re.search(r"(^|\s+)-max_total_time=(\d+)($|\s)", FUZZER_ARGS) if match: - timeout += match.group(2) + timeout += int(match.group(2)) with Path() as current: for fuzzer in current.iterdir(): From da525b6ab5b752c5029433e3513007e6b5e8759b Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Tue, 1 Oct 2024 18:25:22 +0000 Subject: [PATCH 096/680] process timeout --- tests/fuzz/runner.py | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index ccc5a4b7465..f4a6a67e1f8 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -138,8 +138,11 @@ def run_fuzzer(fuzzer: str, timeout: int): ) except subprocess.CalledProcessError as e: # print("Command failed with error:", e) - print("Stderr output:", e.stderr) + print("Stderr output: ", e.stderr) process_error(e.stderr) + except subprocess.TimeoutExpired as e: + print("Timeout: ", e.stderr) + process_fuzzer_output(e.stderr) else: process_fuzzer_output(result.stderr) From fec1b32a79987767618e44dc06a04ac8f6762a09 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 2 Oct 2024 14:01:02 +0000 Subject: [PATCH 097/680] fix parser --- tests/fuzz/runner.py | 34 +++++++++++++--------------------- 1 file changed, 13 insertions(+), 21 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index f4a6a67e1f8..4099ff940e8 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -24,42 +24,34 @@ def process_fuzzer_output(output: str): def process_error(error: str): - ERROR = r"^==\d+== ERROR: (\S+): (.*)" + ERROR = r"^==\d+==\s?ERROR: (\S+): (.*)" error_source = "" error_reason = "" - SUMMARY = r"^SUMMARY: " TEST_UNIT_LINE = r"artifact_prefix='.*/'; Test unit written to (.*)" - test_unit = "" - CALL_STACK_LINE = r"^\s+(#\d+.*)" call_stack = [] is_call_stack = False # pylint: disable=unused-variable for line_num, line in enumerate(error.splitlines(), 1): - if is_call_stack: - match = re.search(CALL_STACK_LINE, line) - if match: - call_stack.append(match.group(1)) - continue - - if re.search(SUMMARY, line): + if re.search(r"^==\d+==", line): is_call_stack = False + continue + call_stack.append(line) continue - if not call_stack and not is_call_stack: - match = re.search(ERROR, line) + if call_stack: + match = re.search(TEST_UNIT_LINE, line) if match: - error_source = match.group(1) - error_reason = match.group(2) - is_call_stack = True - continue + report(error_source, error_reason, call_stack, match.group(1)) + call_stack.clear() + continue - match = re.search(TEST_UNIT_LINE, line) + match = re.search(ERROR, line) if match: - test_unit = match.group(1) - - report(error_source, error_reason, call_stack, test_unit) + error_source = match.group(1) + error_reason = match.group(2) + is_call_stack = True def run_fuzzer(fuzzer: str, timeout: int): From 28b4c8cba32fe57840529f3e2d3298c27564cafe Mon Sep 17 00:00:00 2001 From: avogar Date: Wed, 2 Oct 2024 15:16:38 +0000 Subject: [PATCH 098/680] Fix tests --- tests/queries/0_stateless/01825_new_type_json_12.sh | 2 +- tests/queries/0_stateless/01825_new_type_json_13.sh | 2 +- tests/queries/0_stateless/01825_new_type_json_in_array.sql | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/queries/0_stateless/01825_new_type_json_12.sh b/tests/queries/0_stateless/01825_new_type_json_12.sh index e3909787690..fd5b9fddd75 100755 --- a/tests/queries/0_stateless/01825_new_type_json_12.sh +++ b/tests/queries/0_stateless/01825_new_type_json_12.sh @@ -47,7 +47,7 @@ $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(obj)) as $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(arrayJoin(obj.key_0[]))) as path FROM t_json_12 order by path;" $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(arrayJoin(arrayJoin(obj.key_0[].key_1[])))) as path FROM t_json_12 order by path;" $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(arrayJoin(arrayJoin(arrayJoin(obj.key_0[].key_1[].key_3[]))))) as path FROM t_json_12 order by path;" -$CLICKHOUSE_CLIENT -q "SELECT obj FROM t_json_12 ORDER BY obj.id FORMAT JSONEachRow" --output_format_json_named_tuples_as_objects 1 +$CLICKHOUSE_CLIENT -q "SELECT obj FROM t_json_12 ORDER BY obj.id FORMAT JSONEachRow" --output_format_json_named_tuples_as_objects 1 --allow_suspicious_types_in_order_by 1 $CLICKHOUSE_CLIENT -q "SELECT obj.key_0[].key_1[].key_3[].key_4, obj.key_0[].key_1[].key_3[].key_5, \ obj.key_0[].key_1[].key_3[].key_6, obj.key_0[].key_1[].key_3[].key_7 FROM t_json_12 ORDER BY obj.id" --allow_suspicious_types_in_order_by 1 diff --git a/tests/queries/0_stateless/01825_new_type_json_13.sh b/tests/queries/0_stateless/01825_new_type_json_13.sh index e7d9f556be7..116665e58e3 100755 --- a/tests/queries/0_stateless/01825_new_type_json_13.sh +++ b/tests/queries/0_stateless/01825_new_type_json_13.sh @@ -39,7 +39,7 @@ EOF $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(obj)) as path FROM t_json_13 order by path;" $CLICKHOUSE_CLIENT -q "SELECT DISTINCT arrayJoin(JSONAllPathsWithTypes(arrayJoin(obj.key1[]))) as path FROM t_json_13 order by path;" -$CLICKHOUSE_CLIENT -q "SELECT obj FROM t_json_13 ORDER BY obj.id FORMAT JSONEachRow" --output_format_json_named_tuples_as_objects 1 +$CLICKHOUSE_CLIENT -q "SELECT obj FROM t_json_13 ORDER BY obj.id FORMAT JSONEachRow" --output_format_json_named_tuples_as_objects 1 --allow_suspicious_types_in_order_by 1 $CLICKHOUSE_CLIENT -q "SELECT \ obj.key_1.key_2.key_3.key_8, \ obj.key_1.key_2.key_3.key_4.key_5, \ diff --git a/tests/queries/0_stateless/01825_new_type_json_in_array.sql b/tests/queries/0_stateless/01825_new_type_json_in_array.sql index 3d2e04a1bfd..ef15061e6c8 100644 --- a/tests/queries/0_stateless/01825_new_type_json_in_array.sql +++ b/tests/queries/0_stateless/01825_new_type_json_in_array.sql @@ -3,7 +3,7 @@ SET allow_experimental_json_type = 1; SET allow_experimental_analyzer = 1; SET allow_suspicious_types_in_order_by = 1; -SET allow_suspicious_types_in_order_by = 1; +SET allow_suspicious_types_in_group_by = 1; DROP TABLE IF EXISTS t_json_array; From c367d63c5089cb1f1810bd4f3f767f551b6fed7f Mon Sep 17 00:00:00 2001 From: vdimir Date: Wed, 2 Oct 2024 15:46:40 +0000 Subject: [PATCH 099/680] fix --- src/Processors/QueryPlan/JoinStep.cpp | 5 +++-- src/Processors/Transforms/ColumnPermuteTransform.cpp | 4 ++-- src/Processors/Transforms/ColumnPermuteTransform.h | 2 +- 3 files changed, 6 insertions(+), 5 deletions(-) diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index 2d7dd689149..d6f9590d240 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -151,12 +151,13 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines } column_permutation.resize(n); - pipeline->addSimpleTransform([column_perm = std::move(column_permutation)](const Block & header) mutable + pipeline->addSimpleTransform([&column_permutation](const Block & header) { - return std::make_shared(header, std::move(column_perm)); + return std::make_shared(header, column_permutation); }); } + return pipeline; } diff --git a/src/Processors/Transforms/ColumnPermuteTransform.cpp b/src/Processors/Transforms/ColumnPermuteTransform.cpp index 169dd2dc67e..eb2a691d6d1 100644 --- a/src/Processors/Transforms/ColumnPermuteTransform.cpp +++ b/src/Processors/Transforms/ColumnPermuteTransform.cpp @@ -33,9 +33,9 @@ void permuteChunk(Chunk & chunk, const std::vector & permutation) } -ColumnPermuteTransform::ColumnPermuteTransform(const Block & header_, std::vector permutation_) +ColumnPermuteTransform::ColumnPermuteTransform(const Block & header_, const std::vector & permutation_) : ISimpleTransform(header_, permuteBlock(header_, permutation_), false) - , permutation(std::move(permutation_)) + , permutation(permutation_) { } diff --git a/src/Processors/Transforms/ColumnPermuteTransform.h b/src/Processors/Transforms/ColumnPermuteTransform.h index b2e3c469833..f4d68850193 100644 --- a/src/Processors/Transforms/ColumnPermuteTransform.h +++ b/src/Processors/Transforms/ColumnPermuteTransform.h @@ -13,7 +13,7 @@ namespace DB class ColumnPermuteTransform : public ISimpleTransform { public: - ColumnPermuteTransform(const Block & header_, std::vector permutation_); + ColumnPermuteTransform(const Block & header_, const std::vector & permutation_); String getName() const override { return "ColumnPermuteTransform"; } From ab89e4daa0fe9cf6035c030b1863d64c4c2d8ce0 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 2 Oct 2024 15:51:41 +0000 Subject: [PATCH 100/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 4099ff940e8..d752fce1bd0 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -133,7 +133,7 @@ def run_fuzzer(fuzzer: str, timeout: int): print("Stderr output: ", e.stderr) process_error(e.stderr) except subprocess.TimeoutExpired as e: - print("Timeout: ", e.stderr) + print("Timeout") process_fuzzer_output(e.stderr) else: process_fuzzer_output(result.stderr) From 228b01331d1099f68bc086945a3924e981634cfa Mon Sep 17 00:00:00 2001 From: vdimir Date: Wed, 2 Oct 2024 15:56:26 +0000 Subject: [PATCH 101/680] fix conflict in src/Core/SettingsChangesHistory.cpp --- src/Core/SettingsChangesHistory.cpp | 1 - 1 file changed, 1 deletion(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 1769eebbe8b..a488e6dd203 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -96,7 +96,6 @@ static std::initializer_list Date: Wed, 2 Oct 2024 17:14:11 +0000 Subject: [PATCH 102/680] debugging timeouts --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index d752fce1bd0..05b8faa96a2 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -133,7 +133,7 @@ def run_fuzzer(fuzzer: str, timeout: int): print("Stderr output: ", e.stderr) process_error(e.stderr) except subprocess.TimeoutExpired as e: - print("Timeout") + print("Timeout for %s", cmd_line) process_fuzzer_output(e.stderr) else: process_fuzzer_output(result.stderr) From 0f8fed3d83bac3f9a91225c5c190fa1d6624ebe3 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 2 Oct 2024 20:07:02 +0000 Subject: [PATCH 103/680] add s3 corpus --- tests/fuzz/runner.py | 24 ++++++++++++++++++++---- 1 file changed, 20 insertions(+), 4 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 05b8faa96a2..3b916145e0c 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -6,6 +6,8 @@ import os import re import subprocess from pathlib import Path +from tests.ci.env_helper import S3_BUILDS_BUCKET +from tests.ci.s3_helper import S3Helper DEBUGGER = os.getenv("DEBUGGER", "") FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") @@ -55,6 +57,8 @@ def process_error(error: str): def run_fuzzer(fuzzer: str, timeout: int): + s3 = S3Helper() + logging.info("Running fuzzer %s...", fuzzer) seed_corpus_dir = f"{fuzzer}.in" @@ -63,8 +67,14 @@ def run_fuzzer(fuzzer: str, timeout: int): seed_corpus_dir = "" active_corpus_dir = f"{fuzzer}.corpus" - if not os.path.exists(active_corpus_dir): - os.makedirs(active_corpus_dir) + s3.download_files(bucket=S3_BUILDS_BUCKET, + s3_path=f"fuzzer/corpus/{fuzzer}/", + file_suffix="", + local_directory=active_corpus_dir,) + + new_corpus_dir = f"{fuzzer}.corpus_new" + if not os.path.exists(new_corpus_dir): + os.makedirs(new_corpus_dir) options_file = f"{fuzzer}.options" custom_libfuzzer_options = "" @@ -102,7 +112,7 @@ def run_fuzzer(fuzzer: str, timeout: int): ) cmd_line = ( - f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {active_corpus_dir} {seed_corpus_dir}" + f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {new_corpus_dir} {active_corpus_dir} {seed_corpus_dir}" ) if custom_libfuzzer_options: cmd_line += f" {custom_libfuzzer_options}" @@ -133,11 +143,17 @@ def run_fuzzer(fuzzer: str, timeout: int): print("Stderr output: ", e.stderr) process_error(e.stderr) except subprocess.TimeoutExpired as e: - print("Timeout for %s", cmd_line) + print("Timeout for ", cmd_line) process_fuzzer_output(e.stderr) else: process_fuzzer_output(result.stderr) + f = open(f"{new_corpus_dir}/testfile", "a") + f.write("Now the file has more content!") + f.close() + + s3.upload_build_directory_to_s3(new_corpus_dir, "fuzzer/corpus/") + def main(): logging.basicConfig(level=logging.INFO) From f43ebf004f334ec782fdccd2aa38c1846288fe4a Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 2 Oct 2024 20:24:13 +0000 Subject: [PATCH 104/680] fix style --- tests/fuzz/runner.py | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 3b916145e0c..8e1de7ca38d 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -6,6 +6,7 @@ import os import re import subprocess from pathlib import Path + from tests.ci.env_helper import S3_BUILDS_BUCKET from tests.ci.s3_helper import S3Helper @@ -68,9 +69,10 @@ def run_fuzzer(fuzzer: str, timeout: int): active_corpus_dir = f"{fuzzer}.corpus" s3.download_files(bucket=S3_BUILDS_BUCKET, - s3_path=f"fuzzer/corpus/{fuzzer}/", - file_suffix="", - local_directory=active_corpus_dir,) + s3_path=f"fuzzer/corpus/{fuzzer}/", + file_suffix="", + local_directory=active_corpus_dir, + ) new_corpus_dir = f"{fuzzer}.corpus_new" if not os.path.exists(new_corpus_dir): @@ -111,9 +113,8 @@ def run_fuzzer(fuzzer: str, timeout: int): for key, value in parser["fuzzer_arguments"].items() ) - cmd_line = ( - f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {new_corpus_dir} {active_corpus_dir} {seed_corpus_dir}" - ) + cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {new_corpus_dir} {active_corpus_dir} {seed_corpus_dir}" + if custom_libfuzzer_options: cmd_line += f" {custom_libfuzzer_options}" if fuzzer_arguments: From 245e76a5d3be2dd78cf072ef9c4810da4a497d29 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 2 Oct 2024 20:36:31 +0000 Subject: [PATCH 105/680] fix style --- tests/fuzz/runner.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 8e1de7ca38d..7f398d2124a 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -68,7 +68,8 @@ def run_fuzzer(fuzzer: str, timeout: int): seed_corpus_dir = "" active_corpus_dir = f"{fuzzer}.corpus" - s3.download_files(bucket=S3_BUILDS_BUCKET, + s3.download_files( + bucket=S3_BUILDS_BUCKET, s3_path=f"fuzzer/corpus/{fuzzer}/", file_suffix="", local_directory=active_corpus_dir, From 55fd44935d70195fa969941ee3d98b636bdcfe42 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 2 Oct 2024 20:57:16 +0000 Subject: [PATCH 106/680] fix style --- tests/fuzz/runner.py | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 7f398d2124a..dbe9511b85c 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -7,8 +7,8 @@ import re import subprocess from pathlib import Path -from tests.ci.env_helper import S3_BUILDS_BUCKET -from tests.ci.s3_helper import S3Helper +from ci.env_helper import S3_BUILDS_BUCKET +from ci.s3_helper import S3Helper DEBUGGER = os.getenv("DEBUGGER", "") FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") @@ -150,9 +150,8 @@ def run_fuzzer(fuzzer: str, timeout: int): else: process_fuzzer_output(result.stderr) - f = open(f"{new_corpus_dir}/testfile", "a") - f.write("Now the file has more content!") - f.close() + with open(f"{new_corpus_dir}/testfile", "a", encoding='ascii') as f: + f.write("Now the file has more content!") s3.upload_build_directory_to_s3(new_corpus_dir, "fuzzer/corpus/") From f490d835136e0e28557ffc654e6cb87e13bde65e Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 2 Oct 2024 21:09:31 +0000 Subject: [PATCH 107/680] fix style --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index dbe9511b85c..ac6cbc56a7e 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -150,7 +150,7 @@ def run_fuzzer(fuzzer: str, timeout: int): else: process_fuzzer_output(result.stderr) - with open(f"{new_corpus_dir}/testfile", "a", encoding='ascii') as f: + with open(f"{new_corpus_dir}/testfile", "a", encoding="ascii") as f: f.write("Now the file has more content!") s3.upload_build_directory_to_s3(new_corpus_dir, "fuzzer/corpus/") From 4f23f16417c62057f721273492a0d60441588477 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 2 Oct 2024 22:39:20 +0000 Subject: [PATCH 108/680] fix --- tests/fuzz/runner.py | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index ac6cbc56a7e..d85bc018739 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -7,9 +7,6 @@ import re import subprocess from pathlib import Path -from ci.env_helper import S3_BUILDS_BUCKET -from ci.s3_helper import S3Helper - DEBUGGER = os.getenv("DEBUGGER", "") FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") @@ -174,4 +171,9 @@ def main(): if __name__ == "__main__": + from os import sys, path + ACTIVE_DIR = path.dirname(path.abspath(__file__)) + sys.path.append(path.dirname(ACTIVE_DIR)) + from ci.env_helper import S3_BUILDS_BUCKET + from ci.s3_helper import S3Helper main() From 5e95ce8a485f1497af06b144c3754941fb1fba93 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 2 Oct 2024 23:03:08 +0000 Subject: [PATCH 109/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index d85bc018739..fc93c7437ca 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -171,7 +171,7 @@ def main(): if __name__ == "__main__": - from os import sys, path + from os import path, sys ACTIVE_DIR = path.dirname(path.abspath(__file__)) sys.path.append(path.dirname(ACTIVE_DIR)) from ci.env_helper import S3_BUILDS_BUCKET From dff243a132c5014c1485133d92812bfb3750e67d Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 2 Oct 2024 23:19:06 +0000 Subject: [PATCH 110/680] fix --- tests/fuzz/runner.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index fc93c7437ca..d03bc6f5bed 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -172,8 +172,10 @@ def main(): if __name__ == "__main__": from os import path, sys + ACTIVE_DIR = path.dirname(path.abspath(__file__)) sys.path.append(path.dirname(ACTIVE_DIR)) from ci.env_helper import S3_BUILDS_BUCKET from ci.s3_helper import S3Helper + main() From d022c4615b851b58aaa0f5dbdb1ab3b05b22ab83 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 00:10:59 +0000 Subject: [PATCH 111/680] fix --- tests/fuzz/runner.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index d03bc6f5bed..ffd319cf16c 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -175,7 +175,7 @@ if __name__ == "__main__": ACTIVE_DIR = path.dirname(path.abspath(__file__)) sys.path.append(path.dirname(ACTIVE_DIR)) - from ci.env_helper import S3_BUILDS_BUCKET - from ci.s3_helper import S3Helper + from ci.env_helper import S3_BUILDS_BUCKET # pylint: disable=import-error + from ci.s3_helper import S3Helper # pylint: disable=import-error main() From f009d1e7d5c7c605874c637977e1639455086b67 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 00:28:15 +0000 Subject: [PATCH 112/680] fix --- tests/fuzz/runner.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index ffd319cf16c..171c99698a7 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -175,7 +175,7 @@ if __name__ == "__main__": ACTIVE_DIR = path.dirname(path.abspath(__file__)) sys.path.append(path.dirname(ACTIVE_DIR)) - from ci.env_helper import S3_BUILDS_BUCKET # pylint: disable=import-error - from ci.s3_helper import S3Helper # pylint: disable=import-error + from ci.env_helper import S3_BUILDS_BUCKET # pylint: disable=import-error + from ci.s3_helper import S3Helper # pylint: disable=import-error main() From 4a7de86089ac2bdcad31791d1db717f25c656b5d Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 00:42:53 +0000 Subject: [PATCH 113/680] fix --- tests/fuzz/runner.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 171c99698a7..af3f2ff6040 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -175,7 +175,7 @@ if __name__ == "__main__": ACTIVE_DIR = path.dirname(path.abspath(__file__)) sys.path.append(path.dirname(ACTIVE_DIR)) - from ci.env_helper import S3_BUILDS_BUCKET # pylint: disable=import-error - from ci.s3_helper import S3Helper # pylint: disable=import-error + from ci.env_helper import S3_BUILDS_BUCKET # pylint: disable=import-error,no-name-in-module + from ci.s3_helper import S3Helper # pylint: disable=import-error,no-name-in-module main() From bf292bcc45a131a589bbb0ba113bcc80db380b07 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 3 Oct 2024 00:52:51 +0000 Subject: [PATCH 114/680] Automatic style fix --- tests/fuzz/runner.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index af3f2ff6040..718799a7f63 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -175,7 +175,9 @@ if __name__ == "__main__": ACTIVE_DIR = path.dirname(path.abspath(__file__)) sys.path.append(path.dirname(ACTIVE_DIR)) - from ci.env_helper import S3_BUILDS_BUCKET # pylint: disable=import-error,no-name-in-module + from ci.env_helper import ( # pylint: disable=import-error,no-name-in-module + S3_BUILDS_BUCKET, + ) from ci.s3_helper import S3Helper # pylint: disable=import-error,no-name-in-module main() From d279be6ac2683cfebe56399b8d1e60cca085eb1e Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 02:10:07 +0000 Subject: [PATCH 115/680] add boto3 to requirements --- docker/test/fuzzer/requirements.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/docker/test/fuzzer/requirements.txt b/docker/test/fuzzer/requirements.txt index 3dce93e023b..74147513e76 100644 --- a/docker/test/fuzzer/requirements.txt +++ b/docker/test/fuzzer/requirements.txt @@ -25,3 +25,4 @@ six==1.16.0 wadllib==1.3.6 wheel==0.37.1 zipp==1.0.0 +boto3 From ce3983d757e032cdcbd3af81f0a79a959bf036bc Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 02:20:14 +0000 Subject: [PATCH 116/680] fix --- docker/test/fuzzer/requirements.txt | 1 - docker/test/libfuzzer/requirements.txt | 1 + 2 files changed, 1 insertion(+), 1 deletion(-) diff --git a/docker/test/fuzzer/requirements.txt b/docker/test/fuzzer/requirements.txt index 74147513e76..3dce93e023b 100644 --- a/docker/test/fuzzer/requirements.txt +++ b/docker/test/fuzzer/requirements.txt @@ -25,4 +25,3 @@ six==1.16.0 wadllib==1.3.6 wheel==0.37.1 zipp==1.0.0 -boto3 diff --git a/docker/test/libfuzzer/requirements.txt b/docker/test/libfuzzer/requirements.txt index 3dce93e023b..74147513e76 100644 --- a/docker/test/libfuzzer/requirements.txt +++ b/docker/test/libfuzzer/requirements.txt @@ -25,3 +25,4 @@ six==1.16.0 wadllib==1.3.6 wheel==0.37.1 zipp==1.0.0 +boto3 From c7b8a98fa6a2d0c914112562834c52f4acd04b9a Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 03:12:58 +0000 Subject: [PATCH 117/680] fix --- tests/fuzz/runner.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 718799a7f63..6c4c2930a90 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -174,10 +174,10 @@ if __name__ == "__main__": from os import path, sys ACTIVE_DIR = path.dirname(path.abspath(__file__)) - sys.path.append(path.dirname(ACTIVE_DIR)) - from ci.env_helper import ( # pylint: disable=import-error,no-name-in-module + sys.path.append(path.dirname(ACTIVE_DIR) / "ci") + from env_helper import ( # pylint: disable=import-error,no-name-in-module S3_BUILDS_BUCKET, ) - from ci.s3_helper import S3Helper # pylint: disable=import-error,no-name-in-module + from s3_helper import S3Helper # pylint: disable=import-error,no-name-in-module main() From 2bb3dd7cbc6c860849add0adcd32c296a00d349c Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 04:09:00 +0000 Subject: [PATCH 118/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 6c4c2930a90..a64af5bab66 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -174,7 +174,7 @@ if __name__ == "__main__": from os import path, sys ACTIVE_DIR = path.dirname(path.abspath(__file__)) - sys.path.append(path.dirname(ACTIVE_DIR) / "ci") + sys.path.append(Path(path.dirname(ACTIVE_DIR)) / "ci") from env_helper import ( # pylint: disable=import-error,no-name-in-module S3_BUILDS_BUCKET, ) From 582e01ba57218480a2ef485ccc5f8c4ff440bfc3 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 05:39:42 +0000 Subject: [PATCH 119/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index a64af5bab66..51201e85224 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -174,7 +174,7 @@ if __name__ == "__main__": from os import path, sys ACTIVE_DIR = path.dirname(path.abspath(__file__)) - sys.path.append(Path(path.dirname(ACTIVE_DIR)) / "ci") + sys.path.append((Path(path.dirname(ACTIVE_DIR)) / "ci").as_posix()) from env_helper import ( # pylint: disable=import-error,no-name-in-module S3_BUILDS_BUCKET, ) From 1dc67425bdc084346bafa1264828e979b7909071 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 06:45:03 +0000 Subject: [PATCH 120/680] fix --- tests/fuzz/runner.py | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 51201e85224..e11a5415227 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -1,5 +1,6 @@ #!/usr/bin/env python3 +import botocore import configparser import logging import os @@ -65,12 +66,15 @@ def run_fuzzer(fuzzer: str, timeout: int): seed_corpus_dir = "" active_corpus_dir = f"{fuzzer}.corpus" - s3.download_files( - bucket=S3_BUILDS_BUCKET, - s3_path=f"fuzzer/corpus/{fuzzer}/", - file_suffix="", - local_directory=active_corpus_dir, - ) + try: + s3.download_files( + bucket=S3_BUILDS_BUCKET, + s3_path=f"fuzzer/corpus/{fuzzer}/", + file_suffix="", + local_directory=active_corpus_dir, + ) + except botocore.errorfactory.NoSuchKey as e: + logging.debug("No active corpus exists for %s", fuzzer) new_corpus_dir = f"{fuzzer}.corpus_new" if not os.path.exists(new_corpus_dir): From 0a08ec018a1626a823d4496f57843e24816bf12c Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 07:02:11 +0000 Subject: [PATCH 121/680] fix --- tests/fuzz/runner.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index e11a5415227..06a232a0e5a 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -1,6 +1,5 @@ #!/usr/bin/env python3 -import botocore import configparser import logging import os @@ -8,6 +7,8 @@ import re import subprocess from pathlib import Path +import botocore + DEBUGGER = os.getenv("DEBUGGER", "") FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") From 55ff81518f9a35dc3797b1c80acd6d4ef990c5d3 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 07:24:50 +0000 Subject: [PATCH 122/680] fix --- tests/fuzz/runner.py | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 06a232a0e5a..ccd7cbc475a 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -7,7 +7,7 @@ import re import subprocess from pathlib import Path -import botocore +from botocore.exceptions import ClientError DEBUGGER = os.getenv("DEBUGGER", "") FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") @@ -74,8 +74,11 @@ def run_fuzzer(fuzzer: str, timeout: int): file_suffix="", local_directory=active_corpus_dir, ) - except botocore.errorfactory.NoSuchKey as e: - logging.debug("No active corpus exists for %s", fuzzer) + except ClientError as e: + if e.response['Error']['Code'] == 'NoSuchKey': + logging.debug("No active corpus exists for %s", fuzzer) + else: + raise new_corpus_dir = f"{fuzzer}.corpus_new" if not os.path.exists(new_corpus_dir): From 3008330afec6c45fd3badf335cca57cb173ecadc Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 3 Oct 2024 07:33:39 +0000 Subject: [PATCH 123/680] Automatic style fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index ccd7cbc475a..e1860d60081 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -75,7 +75,7 @@ def run_fuzzer(fuzzer: str, timeout: int): local_directory=active_corpus_dir, ) except ClientError as e: - if e.response['Error']['Code'] == 'NoSuchKey': + if e.response["Error"]["Code"] == "NoSuchKey": logging.debug("No active corpus exists for %s", fuzzer) else: raise From ad05a454352c882e1e81250d99c8d73669d9c2c9 Mon Sep 17 00:00:00 2001 From: vdimir Date: Thu, 3 Oct 2024 11:44:35 +0000 Subject: [PATCH 124/680] upd tests --- .../0_stateless/00826_cross_to_inner_join.sql | 10 +++++----- .../00847_multiple_join_same_column.sql | 14 +++++++------- .../01015_empty_in_inner_right_join.sql.j2 | 2 ++ .../0_stateless/02000_join_on_const.reference | 18 +++++++++--------- .../0_stateless/02000_join_on_const.sql | 16 ++++++++-------- .../0_stateless/03094_one_thousand_joins.sql | 1 + 6 files changed, 32 insertions(+), 29 deletions(-) diff --git a/tests/queries/0_stateless/00826_cross_to_inner_join.sql b/tests/queries/0_stateless/00826_cross_to_inner_join.sql index f81832a4109..5ab7a2d0626 100644 --- a/tests/queries/0_stateless/00826_cross_to_inner_join.sql +++ b/tests/queries/0_stateless/00826_cross_to_inner_join.sql @@ -15,9 +15,9 @@ INSERT INTO t2_00826 values (1,1), (1,2); INSERT INTO t2_00826 (a) values (2), (3); SELECT '--- cross ---'; -SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.a = t2_00826.a; +SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.a = t2_00826.a ORDER BY ALL; SELECT '--- cross nullable ---'; -SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.b = t2_00826.b; +SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.b = t2_00826.b ORDER BY ALL; SELECT '--- cross nullable vs not nullable ---'; SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.a = t2_00826.b ORDER BY t1_00826.a; SELECT '--- cross self ---'; @@ -41,12 +41,12 @@ SELECT '--- is null or ---'; SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.b = t2_00826.a AND (t2_00826.b IS NULL OR t2_00826.b > t2_00826.a) ORDER BY t1_00826.a; SELECT '--- do not rewrite alias ---'; -SELECT a as b FROM t1_00826 cross join t2_00826 where t1_00826.b = t2_00826.a AND b > 0; +SELECT a as b FROM t1_00826 cross join t2_00826 where t1_00826.b = t2_00826.a AND b > 0 ORDER BY ALL; SELECT '--- comma ---'; -SELECT * FROM t1_00826, t2_00826 where t1_00826.a = t2_00826.a; +SELECT * FROM t1_00826, t2_00826 where t1_00826.a = t2_00826.a ORDER BY ALL; SELECT '--- comma nullable ---'; -SELECT * FROM t1_00826, t2_00826 where t1_00826.b = t2_00826.b; +SELECT * FROM t1_00826, t2_00826 where t1_00826.b = t2_00826.b ORDER BY ALL; SELECT '--- comma and or ---'; SELECT * FROM t1_00826, t2_00826 where t1_00826.a = t2_00826.a AND (t2_00826.b IS NULL OR t2_00826.b < 2) ORDER BY ALL; diff --git a/tests/queries/0_stateless/00847_multiple_join_same_column.sql b/tests/queries/0_stateless/00847_multiple_join_same_column.sql index c7f0c6383c2..bbb4eb12466 100644 --- a/tests/queries/0_stateless/00847_multiple_join_same_column.sql +++ b/tests/queries/0_stateless/00847_multiple_join_same_column.sql @@ -20,42 +20,42 @@ select t.a, s.b, s.a, s.b, y.a, y.b from t left join s on (t.a = s.a and s.b = t.b) left join y on (y.a = s.a and y.b = s.b) order by t.a -format PrettyCompactNoEscapes; +format PrettyCompactMonoBlock; select t.a as t_a from t left join s on s.a = t_a order by t.a -format PrettyCompactNoEscapes; +format PrettyCompactMonoBlock; select t.a, s.a as s_a from t left join s on s.a = t.a left join y on y.b = s.b order by t.a -format PrettyCompactNoEscapes; +format PrettyCompactMonoBlock; select t.a, t.a, t.b as t_b from t left join s on t.a = s.a left join y on y.b = s.b order by t.a -format PrettyCompactNoEscapes; +format PrettyCompactMonoBlock; select s.a, s.a, s.b as s_b, s.b from t left join s on s.a = t.a left join y on s.b = y.b order by t.a -format PrettyCompactNoEscapes; +format PrettyCompactMonoBlock; select y.a, y.a, y.b as y_b, y.b from t left join s on s.a = t.a left join y on y.b = s.b order by t.a -format PrettyCompactNoEscapes; +format PrettyCompactMonoBlock; select t.a, t.a as t_a, s.a, s.a as s_a, y.a, y.a as y_a from t left join s on t.a = s.a left join y on y.b = s.b order by t.a -format PrettyCompactNoEscapes; +format PrettyCompactMonoBlock; drop table t; drop table s; diff --git a/tests/queries/0_stateless/01015_empty_in_inner_right_join.sql.j2 b/tests/queries/0_stateless/01015_empty_in_inner_right_join.sql.j2 index cdb9d253b9b..cdbb0542ffb 100644 --- a/tests/queries/0_stateless/01015_empty_in_inner_right_join.sql.j2 +++ b/tests/queries/0_stateless/01015_empty_in_inner_right_join.sql.j2 @@ -1,5 +1,7 @@ SET joined_subquery_requires_alias = 0; +SET query_plan_join_inner_table_selection = 'auto'; + {% for join_algorithm in ['partial_merge', 'hash'] -%} SET join_algorithm = '{{ join_algorithm }}'; diff --git a/tests/queries/0_stateless/02000_join_on_const.reference b/tests/queries/0_stateless/02000_join_on_const.reference index 3bd1633ce32..f8e46a2b976 100644 --- a/tests/queries/0_stateless/02000_join_on_const.reference +++ b/tests/queries/0_stateless/02000_join_on_const.reference @@ -33,23 +33,23 @@ 2 2 2 2 -- { echoOn } -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 1 SETTINGS enable_analyzer = 1; 1 0 2 2 -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; -2 2 +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 1 SETTINGS enable_analyzer = 1; 0 3 -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; +2 2 +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 2, 1 SETTINGS enable_analyzer = 1; 1 0 2 2 0 3 -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 1 SETTINGS enable_analyzer = 1; 1 0 2 0 -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 2 SETTINGS enable_analyzer = 1; 0 2 0 3 -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 2, 1 SETTINGS enable_analyzer = 1; 1 0 2 0 0 2 @@ -59,11 +59,11 @@ SELECT * FROM (SELECT 1 as a) as t1 LEFT JOIN ( SELECT ('b', 256) as b ) AS t2 1 ('',0) SELECT * FROM (SELECT 1 as a) as t1 RIGHT JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; 0 ('b',256) -SELECT * FROM (SELECT 1 as a) as t1 FULL JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; +SELECT * FROM (SELECT 1 as a) as t1 FULL JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL ORDER BY 2; 1 ('',0) 0 ('b',256) SELECT * FROM (SELECT 1 as a) as t1 SEMI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; -SELECT * FROM (SELECT 1 as a) as t1 ANTI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; +SELECT * FROM (SELECT 1 as a) as t1 ANTI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL ORDER BY 2; 1 ('',0) 2 4 2 Nullable(UInt64) UInt8 diff --git a/tests/queries/0_stateless/02000_join_on_const.sql b/tests/queries/0_stateless/02000_join_on_const.sql index da70973ed87..33638edafa5 100644 --- a/tests/queries/0_stateless/02000_join_on_const.sql +++ b/tests/queries/0_stateless/02000_join_on_const.sql @@ -73,20 +73,20 @@ SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 SETTINGS enable_analyzer = 0; -- SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 SETTINGS enable_analyzer = 1; -- { echoOn } -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 2, 1 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 2 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 2, 1 SETTINGS enable_analyzer = 1; SELECT * FROM (SELECT 1 as a) as t1 INNER JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; SELECT * FROM (SELECT 1 as a) as t1 LEFT JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; SELECT * FROM (SELECT 1 as a) as t1 RIGHT JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; -SELECT * FROM (SELECT 1 as a) as t1 FULL JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; +SELECT * FROM (SELECT 1 as a) as t1 FULL JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL ORDER BY 2; SELECT * FROM (SELECT 1 as a) as t1 SEMI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; -SELECT * FROM (SELECT 1 as a) as t1 ANTI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; +SELECT * FROM (SELECT 1 as a) as t1 ANTI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL ORDER BY 2; -- { echoOff } diff --git a/tests/queries/0_stateless/03094_one_thousand_joins.sql b/tests/queries/0_stateless/03094_one_thousand_joins.sql index 6ae4e4d4d3c..69c4fb42a6b 100644 --- a/tests/queries/0_stateless/03094_one_thousand_joins.sql +++ b/tests/queries/0_stateless/03094_one_thousand_joins.sql @@ -3,6 +3,7 @@ SET join_algorithm = 'default'; -- for 'full_sorting_merge' the query is 10x slower SET enable_analyzer = 1; -- old analyzer returns TOO_DEEP_SUBQUERIES +SET query_plan_join_inner_table_selection = 'auto'; -- 'left' is slower -- Bug 33446, marked as 'long' because it still runs around 10 sec SELECT * FROM (SELECT 1 AS x) t1 JOIN (SELECT 1 AS x) t2 ON t1.x = t2.x JOIN (SELECT 1 AS x) t3 ON t1.x = t3.x JOIN (SELECT 1 AS x) t4 ON t1.x = t4.x JOIN (SELECT 1 AS x) t5 ON t1.x = t5.x JOIN (SELECT 1 AS x) t6 ON t1.x = t6.x JOIN (SELECT 1 AS x) t7 ON t1.x = t7.x JOIN (SELECT 1 AS x) t8 ON t1.x = t8.x JOIN (SELECT 1 AS x) t9 ON t1.x = t9.x JOIN (SELECT 1 AS x) t10 ON t1.x = t10.x JOIN (SELECT 1 AS x) t11 ON t1.x = t11.x JOIN (SELECT 1 AS x) t12 ON t1.x = t12.x JOIN (SELECT 1 AS x) t13 ON t1.x = t13.x JOIN (SELECT 1 AS x) t14 ON t1.x = t14.x JOIN (SELECT 1 AS x) t15 ON t1.x = t15.x JOIN (SELECT 1 AS x) t16 ON t1.x = t16.x JOIN (SELECT 1 AS x) t17 ON t1.x = t17.x JOIN (SELECT 1 AS x) t18 ON t1.x = t18.x JOIN (SELECT 1 AS x) t19 ON t1.x = t19.x JOIN (SELECT 1 AS x) t20 ON t1.x = t20.x JOIN (SELECT 1 AS x) t21 ON t1.x = t21.x JOIN (SELECT 1 AS x) t22 ON t1.x = t22.x JOIN (SELECT 1 AS x) t23 ON t1.x = t23.x JOIN (SELECT 1 AS x) t24 ON t1.x = t24.x JOIN (SELECT 1 AS x) t25 ON t1.x = t25.x JOIN (SELECT 1 AS x) t26 ON t1.x = t26.x JOIN (SELECT 1 AS x) t27 ON t1.x = t27.x JOIN (SELECT 1 AS x) t28 ON t1.x = t28.x JOIN (SELECT 1 AS x) t29 ON t1.x = t29.x JOIN (SELECT 1 AS x) t30 ON t1.x = t30.x JOIN (SELECT 1 AS x) t31 ON t1.x = t31.x JOIN (SELECT 1 AS x) t32 ON t1.x = t32.x JOIN (SELECT 1 AS x) t33 ON t1.x = t33.x JOIN (SELECT 1 AS x) t34 ON t1.x = t34.x JOIN (SELECT 1 AS x) t35 ON t1.x = t35.x JOIN (SELECT 1 AS x) t36 ON t1.x = t36.x JOIN (SELECT 1 AS x) t37 ON t1.x = t37.x JOIN (SELECT 1 AS x) t38 ON t1.x = t38.x JOIN (SELECT 1 AS x) t39 ON t1.x = t39.x JOIN (SELECT 1 AS x) t40 ON t1.x = t40.x JOIN (SELECT 1 AS x) t41 ON t1.x = t41.x JOIN (SELECT 1 AS x) t42 ON t1.x = t42.x JOIN (SELECT 1 AS x) t43 ON t1.x = t43.x JOIN (SELECT 1 AS x) t44 ON t1.x = t44.x JOIN (SELECT 1 AS x) t45 ON t1.x = t45.x JOIN (SELECT 1 AS x) t46 ON t1.x = t46.x JOIN (SELECT 1 AS x) t47 ON t1.x = t47.x JOIN (SELECT 1 AS x) t48 ON t1.x = t48.x JOIN (SELECT 1 AS x) t49 ON t1.x = t49.x JOIN (SELECT 1 AS x) t50 ON t1.x = t50.x JOIN (SELECT 1 AS x) t51 ON t1.x = t51.x JOIN (SELECT 1 AS x) t52 ON t1.x = t52.x JOIN (SELECT 1 AS x) t53 ON t1.x = t53.x JOIN (SELECT 1 AS x) t54 ON t1.x = t54.x JOIN (SELECT 1 AS x) t55 ON t1.x = t55.x JOIN (SELECT 1 AS x) t56 ON t1.x = t56.x JOIN (SELECT 1 AS x) t57 ON t1.x = t57.x JOIN (SELECT 1 AS x) t58 ON t1.x = t58.x JOIN (SELECT 1 AS x) t59 ON t1.x = t59.x JOIN (SELECT 1 AS x) t60 ON t1.x = t60.x JOIN (SELECT 1 AS x) t61 ON t1.x = t61.x JOIN (SELECT 1 AS x) t62 ON t1.x = t62.x JOIN (SELECT 1 AS x) t63 ON t1.x = t63.x JOIN (SELECT 1 AS x) t64 ON t1.x = t64.x JOIN (SELECT 1 AS x) t65 ON t1.x = t65.x JOIN (SELECT 1 AS x) t66 ON t1.x = t66.x JOIN (SELECT 1 AS x) t67 ON t1.x = t67.x JOIN (SELECT 1 AS x) t68 ON t1.x = t68.x JOIN (SELECT 1 AS x) t69 ON t1.x = t69.x JOIN (SELECT 1 AS x) t70 ON t1.x = t70.x JOIN (SELECT 1 AS x) t71 ON t1.x = t71.x JOIN (SELECT 1 AS x) t72 ON t1.x = t72.x JOIN (SELECT 1 AS x) t73 ON t1.x = t73.x JOIN (SELECT 1 AS x) t74 ON t1.x = t74.x JOIN (SELECT 1 AS x) t75 ON t1.x = t75.x JOIN (SELECT 1 AS x) t76 ON t1.x = t76.x JOIN (SELECT 1 AS x) t77 ON t1.x = t77.x JOIN (SELECT 1 AS x) t78 ON t1.x = t78.x JOIN (SELECT 1 AS x) t79 ON t1.x = t79.x JOIN (SELECT 1 AS x) t80 ON t1.x = t80.x JOIN (SELECT 1 AS x) t81 ON t1.x = t81.x JOIN (SELECT 1 AS x) t82 ON t1.x = t82.x JOIN (SELECT 1 AS x) t83 ON t1.x = t83.x JOIN (SELECT 1 AS x) t84 ON t1.x = t84.x JOIN (SELECT 1 AS x) t85 ON t1.x = t85.x JOIN (SELECT 1 AS x) t86 ON t1.x = t86.x JOIN (SELECT 1 AS x) t87 ON t1.x = t87.x JOIN (SELECT 1 AS x) t88 ON t1.x = t88.x JOIN (SELECT 1 AS x) t89 ON t1.x = t89.x JOIN (SELECT 1 AS x) t90 ON t1.x = t90.x JOIN (SELECT 1 AS x) t91 ON t1.x = t91.x JOIN (SELECT 1 AS x) t92 ON t1.x = t92.x JOIN (SELECT 1 AS x) t93 ON t1.x = t93.x JOIN (SELECT 1 AS x) t94 ON t1.x = t94.x JOIN (SELECT 1 AS x) t95 ON t1.x = t95.x JOIN (SELECT 1 AS x) t96 ON t1.x = t96.x JOIN (SELECT 1 AS x) t97 ON t1.x = t97.x JOIN (SELECT 1 AS x) t98 ON t1.x = t98.x JOIN (SELECT 1 AS x) t99 ON t1.x = t99.x JOIN (SELECT 1 AS x) t100 ON t1.x = t100.x JOIN (SELECT 1 AS x) t101 ON t1.x = t101.x JOIN (SELECT 1 AS x) t102 ON t1.x = t102.x JOIN (SELECT 1 AS x) t103 ON t1.x = t103.x JOIN (SELECT 1 AS x) t104 ON t1.x = t104.x JOIN (SELECT 1 AS x) t105 ON t1.x = t105.x JOIN (SELECT 1 AS x) t106 ON t1.x = t106.x JOIN (SELECT 1 AS x) t107 ON t1.x = t107.x JOIN (SELECT 1 AS x) t108 ON t1.x = t108.x JOIN (SELECT 1 AS x) t109 ON t1.x = t109.x JOIN (SELECT 1 AS x) t110 ON t1.x = t110.x JOIN (SELECT 1 AS x) t111 ON t1.x = t111.x JOIN (SELECT 1 AS x) t112 ON t1.x = t112.x JOIN (SELECT 1 AS x) t113 ON t1.x = t113.x JOIN (SELECT 1 AS x) t114 ON t1.x = t114.x JOIN (SELECT 1 AS x) t115 ON t1.x = t115.x JOIN (SELECT 1 AS x) t116 ON t1.x = t116.x JOIN (SELECT 1 AS x) t117 ON t1.x = t117.x JOIN (SELECT 1 AS x) t118 ON t1.x = t118.x JOIN (SELECT 1 AS x) t119 ON t1.x = t119.x JOIN (SELECT 1 AS x) t120 ON t1.x = t120.x JOIN (SELECT 1 AS x) t121 ON t1.x = t121.x JOIN (SELECT 1 AS x) t122 ON t1.x = t122.x JOIN (SELECT 1 AS x) t123 ON t1.x = t123.x JOIN (SELECT 1 AS x) t124 ON t1.x = t124.x JOIN (SELECT 1 AS x) t125 ON t1.x = t125.x JOIN (SELECT 1 AS x) t126 ON t1.x = t126.x JOIN (SELECT 1 AS x) t127 ON t1.x = t127.x JOIN (SELECT 1 AS x) t128 ON t1.x = t128.x JOIN (SELECT 1 AS x) t129 ON t1.x = t129.x JOIN (SELECT 1 AS x) t130 ON t1.x = t130.x JOIN (SELECT 1 AS x) t131 ON t1.x = t131.x JOIN (SELECT 1 AS x) t132 ON t1.x = t132.x JOIN (SELECT 1 AS x) t133 ON t1.x = t133.x JOIN (SELECT 1 AS x) t134 ON t1.x = t134.x JOIN (SELECT 1 AS x) t135 ON t1.x = t135.x JOIN (SELECT 1 AS x) t136 ON t1.x = t136.x JOIN (SELECT 1 AS x) t137 ON t1.x = t137.x JOIN (SELECT 1 AS x) t138 ON t1.x = t138.x JOIN (SELECT 1 AS x) t139 ON t1.x = t139.x JOIN (SELECT 1 AS x) t140 ON t1.x = t140.x JOIN (SELECT 1 AS x) t141 ON t1.x = t141.x JOIN (SELECT 1 AS x) t142 ON t1.x = t142.x JOIN (SELECT 1 AS x) t143 ON t1.x = t143.x JOIN (SELECT 1 AS x) t144 ON t1.x = t144.x JOIN (SELECT 1 AS x) t145 ON t1.x = t145.x JOIN (SELECT 1 AS x) t146 ON t1.x = t146.x JOIN (SELECT 1 AS x) t147 ON t1.x = t147.x JOIN (SELECT 1 AS x) t148 ON t1.x = t148.x JOIN (SELECT 1 AS x) t149 ON t1.x = t149.x JOIN (SELECT 1 AS x) t150 ON t1.x = t150.x JOIN (SELECT 1 AS x) t151 ON t1.x = t151.x JOIN (SELECT 1 AS x) t152 ON t1.x = t152.x JOIN (SELECT 1 AS x) t153 ON t1.x = t153.x JOIN (SELECT 1 AS x) t154 ON t1.x = t154.x JOIN (SELECT 1 AS x) t155 ON t1.x = t155.x JOIN (SELECT 1 AS x) t156 ON t1.x = t156.x JOIN (SELECT 1 AS x) t157 ON t1.x = t157.x JOIN (SELECT 1 AS x) t158 ON t1.x = t158.x JOIN (SELECT 1 AS x) t159 ON t1.x = t159.x JOIN (SELECT 1 AS x) t160 ON t1.x = t160.x JOIN (SELECT 1 AS x) t161 ON t1.x = t161.x JOIN (SELECT 1 AS x) t162 ON t1.x = t162.x JOIN (SELECT 1 AS x) t163 ON t1.x = t163.x JOIN (SELECT 1 AS x) t164 ON t1.x = t164.x JOIN (SELECT 1 AS x) t165 ON t1.x = t165.x JOIN (SELECT 1 AS x) t166 ON t1.x = t166.x JOIN (SELECT 1 AS x) t167 ON t1.x = t167.x JOIN (SELECT 1 AS x) t168 ON t1.x = t168.x JOIN (SELECT 1 AS x) t169 ON t1.x = t169.x JOIN (SELECT 1 AS x) t170 ON t1.x = t170.x JOIN (SELECT 1 AS x) t171 ON t1.x = t171.x JOIN (SELECT 1 AS x) t172 ON t1.x = t172.x JOIN (SELECT 1 AS x) t173 ON t1.x = t173.x JOIN (SELECT 1 AS x) t174 ON t1.x = t174.x JOIN (SELECT 1 AS x) t175 ON t1.x = t175.x JOIN (SELECT 1 AS x) t176 ON t1.x = t176.x JOIN (SELECT 1 AS x) t177 ON t1.x = t177.x JOIN (SELECT 1 AS x) t178 ON t1.x = t178.x JOIN (SELECT 1 AS x) t179 ON t1.x = t179.x JOIN (SELECT 1 AS x) t180 ON t1.x = t180.x JOIN (SELECT 1 AS x) t181 ON t1.x = t181.x JOIN (SELECT 1 AS x) t182 ON t1.x = t182.x JOIN (SELECT 1 AS x) t183 ON t1.x = t183.x JOIN (SELECT 1 AS x) t184 ON t1.x = t184.x JOIN (SELECT 1 AS x) t185 ON t1.x = t185.x JOIN (SELECT 1 AS x) t186 ON t1.x = t186.x JOIN (SELECT 1 AS x) t187 ON t1.x = t187.x JOIN (SELECT 1 AS x) t188 ON t1.x = t188.x JOIN (SELECT 1 AS x) t189 ON t1.x = t189.x JOIN (SELECT 1 AS x) t190 ON t1.x = t190.x JOIN (SELECT 1 AS x) t191 ON t1.x = t191.x JOIN (SELECT 1 AS x) t192 ON t1.x = t192.x JOIN (SELECT 1 AS x) t193 ON t1.x = t193.x JOIN (SELECT 1 AS x) t194 ON t1.x = t194.x JOIN (SELECT 1 AS x) t195 ON t1.x = t195.x JOIN (SELECT 1 AS x) t196 ON t1.x = t196.x JOIN (SELECT 1 AS x) t197 ON t1.x = t197.x JOIN (SELECT 1 AS x) t198 ON t1.x = t198.x JOIN (SELECT 1 AS x) t199 ON t1.x = t199.x JOIN (SELECT 1 AS x) t200 ON t1.x = t200.x JOIN (SELECT 1 AS x) t201 ON t1.x = t201.x JOIN (SELECT 1 AS x) t202 ON t1.x = t202.x JOIN (SELECT 1 AS x) t203 ON t1.x = t203.x JOIN (SELECT 1 AS x) t204 ON t1.x = t204.x JOIN (SELECT 1 AS x) t205 ON t1.x = t205.x JOIN (SELECT 1 AS x) t206 ON t1.x = t206.x JOIN (SELECT 1 AS x) t207 ON t1.x = t207.x JOIN (SELECT 1 AS x) t208 ON t1.x = t208.x JOIN (SELECT 1 AS x) t209 ON t1.x = t209.x JOIN (SELECT 1 AS x) t210 ON t1.x = t210.x JOIN (SELECT 1 AS x) t211 ON t1.x = t211.x JOIN (SELECT 1 AS x) t212 ON t1.x = t212.x JOIN (SELECT 1 AS x) t213 ON t1.x = t213.x JOIN (SELECT 1 AS x) t214 ON t1.x = t214.x JOIN (SELECT 1 AS x) t215 ON t1.x = t215.x JOIN (SELECT 1 AS x) t216 ON t1.x = t216.x JOIN (SELECT 1 AS x) t217 ON t1.x = t217.x JOIN (SELECT 1 AS x) t218 ON t1.x = t218.x JOIN (SELECT 1 AS x) t219 ON t1.x = t219.x JOIN (SELECT 1 AS x) t220 ON t1.x = t220.x JOIN (SELECT 1 AS x) t221 ON t1.x = t221.x JOIN (SELECT 1 AS x) t222 ON t1.x = t222.x JOIN (SELECT 1 AS x) t223 ON t1.x = t223.x JOIN (SELECT 1 AS x) t224 ON t1.x = t224.x JOIN (SELECT 1 AS x) t225 ON t1.x = t225.x JOIN (SELECT 1 AS x) t226 ON t1.x = t226.x JOIN (SELECT 1 AS x) t227 ON t1.x = t227.x JOIN (SELECT 1 AS x) t228 ON t1.x = t228.x JOIN (SELECT 1 AS x) t229 ON t1.x = t229.x JOIN (SELECT 1 AS x) t230 ON t1.x = t230.x JOIN (SELECT 1 AS x) t231 ON t1.x = t231.x JOIN (SELECT 1 AS x) t232 ON t1.x = t232.x JOIN (SELECT 1 AS x) t233 ON t1.x = t233.x JOIN (SELECT 1 AS x) t234 ON t1.x = t234.x JOIN (SELECT 1 AS x) t235 ON t1.x = t235.x JOIN (SELECT 1 AS x) t236 ON t1.x = t236.x JOIN (SELECT 1 AS x) t237 ON t1.x = t237.x JOIN (SELECT 1 AS x) t238 ON t1.x = t238.x JOIN (SELECT 1 AS x) t239 ON t1.x = t239.x JOIN (SELECT 1 AS x) t240 ON t1.x = t240.x JOIN (SELECT 1 AS x) t241 ON t1.x = t241.x JOIN (SELECT 1 AS x) t242 ON t1.x = t242.x JOIN (SELECT 1 AS x) t243 ON t1.x = t243.x JOIN (SELECT 1 AS x) t244 ON t1.x = t244.x JOIN (SELECT 1 AS x) t245 ON t1.x = t245.x JOIN (SELECT 1 AS x) t246 ON t1.x = t246.x JOIN (SELECT 1 AS x) t247 ON t1.x = t247.x JOIN (SELECT 1 AS x) t248 ON t1.x = t248.x JOIN (SELECT 1 AS x) t249 ON t1.x = t249.x JOIN (SELECT 1 AS x) t250 ON t1.x = t250.x JOIN (SELECT 1 AS x) t251 ON t1.x = t251.x JOIN (SELECT 1 AS x) t252 ON t1.x = t252.x JOIN (SELECT 1 AS x) t253 ON t1.x = t253.x JOIN (SELECT 1 AS x) t254 ON t1.x = t254.x JOIN (SELECT 1 AS x) t255 ON t1.x = t255.x JOIN (SELECT 1 AS x) t256 ON t1.x = t256.x JOIN (SELECT 1 AS x) t257 ON t1.x = t257.x JOIN (SELECT 1 AS x) t258 ON t1.x = t258.x JOIN (SELECT 1 AS x) t259 ON t1.x = t259.x JOIN (SELECT 1 AS x) t260 ON t1.x = t260.x JOIN (SELECT 1 AS x) t261 ON t1.x = t261.x JOIN (SELECT 1 AS x) t262 ON t1.x = t262.x JOIN (SELECT 1 AS x) t263 ON t1.x = t263.x JOIN (SELECT 1 AS x) t264 ON t1.x = t264.x JOIN (SELECT 1 AS x) t265 ON t1.x = t265.x JOIN (SELECT 1 AS x) t266 ON t1.x = t266.x JOIN (SELECT 1 AS x) t267 ON t1.x = t267.x JOIN (SELECT 1 AS x) t268 ON t1.x = t268.x JOIN (SELECT 1 AS x) t269 ON t1.x = t269.x JOIN (SELECT 1 AS x) t270 ON t1.x = t270.x JOIN (SELECT 1 AS x) t271 ON t1.x = t271.x JOIN (SELECT 1 AS x) t272 ON t1.x = t272.x JOIN (SELECT 1 AS x) t273 ON t1.x = t273.x JOIN (SELECT 1 AS x) t274 ON t1.x = t274.x JOIN (SELECT 1 AS x) t275 ON t1.x = t275.x JOIN (SELECT 1 AS x) t276 ON t1.x = t276.x JOIN (SELECT 1 AS x) t277 ON t1.x = t277.x JOIN (SELECT 1 AS x) t278 ON t1.x = t278.x JOIN (SELECT 1 AS x) t279 ON t1.x = t279.x JOIN (SELECT 1 AS x) t280 ON t1.x = t280.x JOIN (SELECT 1 AS x) t281 ON t1.x = t281.x JOIN (SELECT 1 AS x) t282 ON t1.x = t282.x JOIN (SELECT 1 AS x) t283 ON t1.x = t283.x JOIN (SELECT 1 AS x) t284 ON t1.x = t284.x JOIN (SELECT 1 AS x) t285 ON t1.x = t285.x JOIN (SELECT 1 AS x) t286 ON t1.x = t286.x JOIN (SELECT 1 AS x) t287 ON t1.x = t287.x JOIN (SELECT 1 AS x) t288 ON t1.x = t288.x JOIN (SELECT 1 AS x) t289 ON t1.x = t289.x JOIN (SELECT 1 AS x) t290 ON t1.x = t290.x JOIN (SELECT 1 AS x) t291 ON t1.x = t291.x JOIN (SELECT 1 AS x) t292 ON t1.x = t292.x JOIN (SELECT 1 AS x) t293 ON t1.x = t293.x JOIN (SELECT 1 AS x) t294 ON t1.x = t294.x JOIN (SELECT 1 AS x) t295 ON t1.x = t295.x JOIN (SELECT 1 AS x) t296 ON t1.x = t296.x JOIN (SELECT 1 AS x) t297 ON t1.x = t297.x JOIN (SELECT 1 AS x) t298 ON t1.x = t298.x JOIN (SELECT 1 AS x) t299 ON t1.x = t299.x JOIN (SELECT 1 AS x) t300 ON t1.x = t300.x JOIN (SELECT 1 AS x) t301 ON t1.x = t301.x JOIN (SELECT 1 AS x) t302 ON t1.x = t302.x JOIN (SELECT 1 AS x) t303 ON t1.x = t303.x JOIN (SELECT 1 AS x) t304 ON t1.x = t304.x JOIN (SELECT 1 AS x) t305 ON t1.x = t305.x JOIN (SELECT 1 AS x) t306 ON t1.x = t306.x JOIN (SELECT 1 AS x) t307 ON t1.x = t307.x JOIN (SELECT 1 AS x) t308 ON t1.x = t308.x JOIN (SELECT 1 AS x) t309 ON t1.x = t309.x JOIN (SELECT 1 AS x) t310 ON t1.x = t310.x JOIN (SELECT 1 AS x) t311 ON t1.x = t311.x JOIN (SELECT 1 AS x) t312 ON t1.x = t312.x JOIN (SELECT 1 AS x) t313 ON t1.x = t313.x JOIN (SELECT 1 AS x) t314 ON t1.x = t314.x JOIN (SELECT 1 AS x) t315 ON t1.x = t315.x JOIN (SELECT 1 AS x) t316 ON t1.x = t316.x JOIN (SELECT 1 AS x) t317 ON t1.x = t317.x JOIN (SELECT 1 AS x) t318 ON t1.x = t318.x JOIN (SELECT 1 AS x) t319 ON t1.x = t319.x JOIN (SELECT 1 AS x) t320 ON t1.x = t320.x JOIN (SELECT 1 AS x) t321 ON t1.x = t321.x JOIN (SELECT 1 AS x) t322 ON t1.x = t322.x JOIN (SELECT 1 AS x) t323 ON t1.x = t323.x JOIN (SELECT 1 AS x) t324 ON t1.x = t324.x JOIN (SELECT 1 AS x) t325 ON t1.x = t325.x JOIN (SELECT 1 AS x) t326 ON t1.x = t326.x JOIN (SELECT 1 AS x) t327 ON t1.x = t327.x JOIN (SELECT 1 AS x) t328 ON t1.x = t328.x JOIN (SELECT 1 AS x) t329 ON t1.x = t329.x JOIN (SELECT 1 AS x) t330 ON t1.x = t330.x JOIN (SELECT 1 AS x) t331 ON t1.x = t331.x JOIN (SELECT 1 AS x) t332 ON t1.x = t332.x JOIN (SELECT 1 AS x) t333 ON t1.x = t333.x JOIN (SELECT 1 AS x) t334 ON t1.x = t334.x JOIN (SELECT 1 AS x) t335 ON t1.x = t335.x JOIN (SELECT 1 AS x) t336 ON t1.x = t336.x JOIN (SELECT 1 AS x) t337 ON t1.x = t337.x JOIN (SELECT 1 AS x) t338 ON t1.x = t338.x JOIN (SELECT 1 AS x) t339 ON t1.x = t339.x JOIN (SELECT 1 AS x) t340 ON t1.x = t340.x JOIN (SELECT 1 AS x) t341 ON t1.x = t341.x JOIN (SELECT 1 AS x) t342 ON t1.x = t342.x JOIN (SELECT 1 AS x) t343 ON t1.x = t343.x JOIN (SELECT 1 AS x) t344 ON t1.x = t344.x JOIN (SELECT 1 AS x) t345 ON t1.x = t345.x JOIN (SELECT 1 AS x) t346 ON t1.x = t346.x JOIN (SELECT 1 AS x) t347 ON t1.x = t347.x JOIN (SELECT 1 AS x) t348 ON t1.x = t348.x JOIN (SELECT 1 AS x) t349 ON t1.x = t349.x JOIN (SELECT 1 AS x) t350 ON t1.x = t350.x JOIN (SELECT 1 AS x) t351 ON t1.x = t351.x JOIN (SELECT 1 AS x) t352 ON t1.x = t352.x JOIN (SELECT 1 AS x) t353 ON t1.x = t353.x JOIN (SELECT 1 AS x) t354 ON t1.x = t354.x JOIN (SELECT 1 AS x) t355 ON t1.x = t355.x JOIN (SELECT 1 AS x) t356 ON t1.x = t356.x JOIN (SELECT 1 AS x) t357 ON t1.x = t357.x JOIN (SELECT 1 AS x) t358 ON t1.x = t358.x JOIN (SELECT 1 AS x) t359 ON t1.x = t359.x JOIN (SELECT 1 AS x) t360 ON t1.x = t360.x JOIN (SELECT 1 AS x) t361 ON t1.x = t361.x JOIN (SELECT 1 AS x) t362 ON t1.x = t362.x JOIN (SELECT 1 AS x) t363 ON t1.x = t363.x JOIN (SELECT 1 AS x) t364 ON t1.x = t364.x JOIN (SELECT 1 AS x) t365 ON t1.x = t365.x JOIN (SELECT 1 AS x) t366 ON t1.x = t366.x JOIN (SELECT 1 AS x) t367 ON t1.x = t367.x JOIN (SELECT 1 AS x) t368 ON t1.x = t368.x JOIN (SELECT 1 AS x) t369 ON t1.x = t369.x JOIN (SELECT 1 AS x) t370 ON t1.x = t370.x JOIN (SELECT 1 AS x) t371 ON t1.x = t371.x JOIN (SELECT 1 AS x) t372 ON t1.x = t372.x JOIN (SELECT 1 AS x) t373 ON t1.x = t373.x JOIN (SELECT 1 AS x) t374 ON t1.x = t374.x JOIN (SELECT 1 AS x) t375 ON t1.x = t375.x JOIN (SELECT 1 AS x) t376 ON t1.x = t376.x JOIN (SELECT 1 AS x) t377 ON t1.x = t377.x JOIN (SELECT 1 AS x) t378 ON t1.x = t378.x JOIN (SELECT 1 AS x) t379 ON t1.x = t379.x JOIN (SELECT 1 AS x) t380 ON t1.x = t380.x JOIN (SELECT 1 AS x) t381 ON t1.x = t381.x JOIN (SELECT 1 AS x) t382 ON t1.x = t382.x JOIN (SELECT 1 AS x) t383 ON t1.x = t383.x JOIN (SELECT 1 AS x) t384 ON t1.x = t384.x JOIN (SELECT 1 AS x) t385 ON t1.x = t385.x JOIN (SELECT 1 AS x) t386 ON t1.x = t386.x JOIN (SELECT 1 AS x) t387 ON t1.x = t387.x JOIN (SELECT 1 AS x) t388 ON t1.x = t388.x JOIN (SELECT 1 AS x) t389 ON t1.x = t389.x JOIN (SELECT 1 AS x) t390 ON t1.x = t390.x JOIN (SELECT 1 AS x) t391 ON t1.x = t391.x JOIN (SELECT 1 AS x) t392 ON t1.x = t392.x JOIN (SELECT 1 AS x) t393 ON t1.x = t393.x JOIN (SELECT 1 AS x) t394 ON t1.x = t394.x JOIN (SELECT 1 AS x) t395 ON t1.x = t395.x JOIN (SELECT 1 AS x) t396 ON t1.x = t396.x JOIN (SELECT 1 AS x) t397 ON t1.x = t397.x JOIN (SELECT 1 AS x) t398 ON t1.x = t398.x JOIN (SELECT 1 AS x) t399 ON t1.x = t399.x JOIN (SELECT 1 AS x) t400 ON t1.x = t400.x JOIN (SELECT 1 AS x) t401 ON t1.x = t401.x JOIN (SELECT 1 AS x) t402 ON t1.x = t402.x JOIN (SELECT 1 AS x) t403 ON t1.x = t403.x JOIN (SELECT 1 AS x) t404 ON t1.x = t404.x JOIN (SELECT 1 AS x) t405 ON t1.x = t405.x JOIN (SELECT 1 AS x) t406 ON t1.x = t406.x JOIN (SELECT 1 AS x) t407 ON t1.x = t407.x JOIN (SELECT 1 AS x) t408 ON t1.x = t408.x JOIN (SELECT 1 AS x) t409 ON t1.x = t409.x JOIN (SELECT 1 AS x) t410 ON t1.x = t410.x JOIN (SELECT 1 AS x) t411 ON t1.x = t411.x JOIN (SELECT 1 AS x) t412 ON t1.x = t412.x JOIN (SELECT 1 AS x) t413 ON t1.x = t413.x JOIN (SELECT 1 AS x) t414 ON t1.x = t414.x JOIN (SELECT 1 AS x) t415 ON t1.x = t415.x JOIN (SELECT 1 AS x) t416 ON t1.x = t416.x JOIN (SELECT 1 AS x) t417 ON t1.x = t417.x JOIN (SELECT 1 AS x) t418 ON t1.x = t418.x JOIN (SELECT 1 AS x) t419 ON t1.x = t419.x JOIN (SELECT 1 AS x) t420 ON t1.x = t420.x JOIN (SELECT 1 AS x) t421 ON t1.x = t421.x JOIN (SELECT 1 AS x) t422 ON t1.x = t422.x JOIN (SELECT 1 AS x) t423 ON t1.x = t423.x JOIN (SELECT 1 AS x) t424 ON t1.x = t424.x JOIN (SELECT 1 AS x) t425 ON t1.x = t425.x JOIN (SELECT 1 AS x) t426 ON t1.x = t426.x JOIN (SELECT 1 AS x) t427 ON t1.x = t427.x JOIN (SELECT 1 AS x) t428 ON t1.x = t428.x JOIN (SELECT 1 AS x) t429 ON t1.x = t429.x JOIN (SELECT 1 AS x) t430 ON t1.x = t430.x JOIN (SELECT 1 AS x) t431 ON t1.x = t431.x JOIN (SELECT 1 AS x) t432 ON t1.x = t432.x JOIN (SELECT 1 AS x) t433 ON t1.x = t433.x JOIN (SELECT 1 AS x) t434 ON t1.x = t434.x JOIN (SELECT 1 AS x) t435 ON t1.x = t435.x JOIN (SELECT 1 AS x) t436 ON t1.x = t436.x JOIN (SELECT 1 AS x) t437 ON t1.x = t437.x JOIN (SELECT 1 AS x) t438 ON t1.x = t438.x JOIN (SELECT 1 AS x) t439 ON t1.x = t439.x JOIN (SELECT 1 AS x) t440 ON t1.x = t440.x JOIN (SELECT 1 AS x) t441 ON t1.x = t441.x JOIN (SELECT 1 AS x) t442 ON t1.x = t442.x JOIN (SELECT 1 AS x) t443 ON t1.x = t443.x JOIN (SELECT 1 AS x) t444 ON t1.x = t444.x JOIN (SELECT 1 AS x) t445 ON t1.x = t445.x JOIN (SELECT 1 AS x) t446 ON t1.x = t446.x JOIN (SELECT 1 AS x) t447 ON t1.x = t447.x JOIN (SELECT 1 AS x) t448 ON t1.x = t448.x JOIN (SELECT 1 AS x) t449 ON t1.x = t449.x JOIN (SELECT 1 AS x) t450 ON t1.x = t450.x JOIN (SELECT 1 AS x) t451 ON t1.x = t451.x JOIN (SELECT 1 AS x) t452 ON t1.x = t452.x JOIN (SELECT 1 AS x) t453 ON t1.x = t453.x JOIN (SELECT 1 AS x) t454 ON t1.x = t454.x JOIN (SELECT 1 AS x) t455 ON t1.x = t455.x JOIN (SELECT 1 AS x) t456 ON t1.x = t456.x JOIN (SELECT 1 AS x) t457 ON t1.x = t457.x JOIN (SELECT 1 AS x) t458 ON t1.x = t458.x JOIN (SELECT 1 AS x) t459 ON t1.x = t459.x JOIN (SELECT 1 AS x) t460 ON t1.x = t460.x JOIN (SELECT 1 AS x) t461 ON t1.x = t461.x JOIN (SELECT 1 AS x) t462 ON t1.x = t462.x JOIN (SELECT 1 AS x) t463 ON t1.x = t463.x JOIN (SELECT 1 AS x) t464 ON t1.x = t464.x JOIN (SELECT 1 AS x) t465 ON t1.x = t465.x JOIN (SELECT 1 AS x) t466 ON t1.x = t466.x JOIN (SELECT 1 AS x) t467 ON t1.x = t467.x JOIN (SELECT 1 AS x) t468 ON t1.x = t468.x JOIN (SELECT 1 AS x) t469 ON t1.x = t469.x JOIN (SELECT 1 AS x) t470 ON t1.x = t470.x JOIN (SELECT 1 AS x) t471 ON t1.x = t471.x JOIN (SELECT 1 AS x) t472 ON t1.x = t472.x JOIN (SELECT 1 AS x) t473 ON t1.x = t473.x JOIN (SELECT 1 AS x) t474 ON t1.x = t474.x JOIN (SELECT 1 AS x) t475 ON t1.x = t475.x JOIN (SELECT 1 AS x) t476 ON t1.x = t476.x JOIN (SELECT 1 AS x) t477 ON t1.x = t477.x JOIN (SELECT 1 AS x) t478 ON t1.x = t478.x JOIN (SELECT 1 AS x) t479 ON t1.x = t479.x JOIN (SELECT 1 AS x) t480 ON t1.x = t480.x JOIN (SELECT 1 AS x) t481 ON t1.x = t481.x JOIN (SELECT 1 AS x) t482 ON t1.x = t482.x JOIN (SELECT 1 AS x) t483 ON t1.x = t483.x JOIN (SELECT 1 AS x) t484 ON t1.x = t484.x JOIN (SELECT 1 AS x) t485 ON t1.x = t485.x JOIN (SELECT 1 AS x) t486 ON t1.x = t486.x JOIN (SELECT 1 AS x) t487 ON t1.x = t487.x JOIN (SELECT 1 AS x) t488 ON t1.x = t488.x JOIN (SELECT 1 AS x) t489 ON t1.x = t489.x JOIN (SELECT 1 AS x) t490 ON t1.x = t490.x JOIN (SELECT 1 AS x) t491 ON t1.x = t491.x JOIN (SELECT 1 AS x) t492 ON t1.x = t492.x JOIN (SELECT 1 AS x) t493 ON t1.x = t493.x JOIN (SELECT 1 AS x) t494 ON t1.x = t494.x JOIN (SELECT 1 AS x) t495 ON t1.x = t495.x JOIN (SELECT 1 AS x) t496 ON t1.x = t496.x JOIN (SELECT 1 AS x) t497 ON t1.x = t497.x JOIN (SELECT 1 AS x) t498 ON t1.x = t498.x JOIN (SELECT 1 AS x) t499 ON t1.x = t499.x JOIN (SELECT 1 AS x) t500 ON t1.x = t500.x JOIN (SELECT 1 AS x) t501 ON t1.x = t501.x JOIN (SELECT 1 AS x) t502 ON t1.x = t502.x JOIN (SELECT 1 AS x) t503 ON t1.x = t503.x JOIN (SELECT 1 AS x) t504 ON t1.x = t504.x JOIN (SELECT 1 AS x) t505 ON t1.x = t505.x JOIN (SELECT 1 AS x) t506 ON t1.x = t506.x JOIN (SELECT 1 AS x) t507 ON t1.x = t507.x JOIN (SELECT 1 AS x) t508 ON t1.x = t508.x JOIN (SELECT 1 AS x) t509 ON t1.x = t509.x JOIN (SELECT 1 AS x) t510 ON t1.x = t510.x JOIN (SELECT 1 AS x) t511 ON t1.x = t511.x JOIN (SELECT 1 AS x) t512 ON t1.x = t512.x JOIN (SELECT 1 AS x) t513 ON t1.x = t513.x JOIN (SELECT 1 AS x) t514 ON t1.x = t514.x JOIN (SELECT 1 AS x) t515 ON t1.x = t515.x JOIN (SELECT 1 AS x) t516 ON t1.x = t516.x JOIN (SELECT 1 AS x) t517 ON t1.x = t517.x JOIN (SELECT 1 AS x) t518 ON t1.x = t518.x JOIN (SELECT 1 AS x) t519 ON t1.x = t519.x JOIN (SELECT 1 AS x) t520 ON t1.x = t520.x JOIN (SELECT 1 AS x) t521 ON t1.x = t521.x JOIN (SELECT 1 AS x) t522 ON t1.x = t522.x JOIN (SELECT 1 AS x) t523 ON t1.x = t523.x JOIN (SELECT 1 AS x) t524 ON t1.x = t524.x JOIN (SELECT 1 AS x) t525 ON t1.x = t525.x JOIN (SELECT 1 AS x) t526 ON t1.x = t526.x JOIN (SELECT 1 AS x) t527 ON t1.x = t527.x JOIN (SELECT 1 AS x) t528 ON t1.x = t528.x JOIN (SELECT 1 AS x) t529 ON t1.x = t529.x JOIN (SELECT 1 AS x) t530 ON t1.x = t530.x JOIN (SELECT 1 AS x) t531 ON t1.x = t531.x JOIN (SELECT 1 AS x) t532 ON t1.x = t532.x JOIN (SELECT 1 AS x) t533 ON t1.x = t533.x JOIN (SELECT 1 AS x) t534 ON t1.x = t534.x JOIN (SELECT 1 AS x) t535 ON t1.x = t535.x JOIN (SELECT 1 AS x) t536 ON t1.x = t536.x JOIN (SELECT 1 AS x) t537 ON t1.x = t537.x JOIN (SELECT 1 AS x) t538 ON t1.x = t538.x JOIN (SELECT 1 AS x) t539 ON t1.x = t539.x JOIN (SELECT 1 AS x) t540 ON t1.x = t540.x JOIN (SELECT 1 AS x) t541 ON t1.x = t541.x JOIN (SELECT 1 AS x) t542 ON t1.x = t542.x JOIN (SELECT 1 AS x) t543 ON t1.x = t543.x JOIN (SELECT 1 AS x) t544 ON t1.x = t544.x JOIN (SELECT 1 AS x) t545 ON t1.x = t545.x JOIN (SELECT 1 AS x) t546 ON t1.x = t546.x JOIN (SELECT 1 AS x) t547 ON t1.x = t547.x JOIN (SELECT 1 AS x) t548 ON t1.x = t548.x JOIN (SELECT 1 AS x) t549 ON t1.x = t549.x JOIN (SELECT 1 AS x) t550 ON t1.x = t550.x JOIN (SELECT 1 AS x) t551 ON t1.x = t551.x JOIN (SELECT 1 AS x) t552 ON t1.x = t552.x JOIN (SELECT 1 AS x) t553 ON t1.x = t553.x JOIN (SELECT 1 AS x) t554 ON t1.x = t554.x JOIN (SELECT 1 AS x) t555 ON t1.x = t555.x JOIN (SELECT 1 AS x) t556 ON t1.x = t556.x JOIN (SELECT 1 AS x) t557 ON t1.x = t557.x JOIN (SELECT 1 AS x) t558 ON t1.x = t558.x JOIN (SELECT 1 AS x) t559 ON t1.x = t559.x JOIN (SELECT 1 AS x) t560 ON t1.x = t560.x JOIN (SELECT 1 AS x) t561 ON t1.x = t561.x JOIN (SELECT 1 AS x) t562 ON t1.x = t562.x JOIN (SELECT 1 AS x) t563 ON t1.x = t563.x JOIN (SELECT 1 AS x) t564 ON t1.x = t564.x JOIN (SELECT 1 AS x) t565 ON t1.x = t565.x JOIN (SELECT 1 AS x) t566 ON t1.x = t566.x JOIN (SELECT 1 AS x) t567 ON t1.x = t567.x JOIN (SELECT 1 AS x) t568 ON t1.x = t568.x JOIN (SELECT 1 AS x) t569 ON t1.x = t569.x JOIN (SELECT 1 AS x) t570 ON t1.x = t570.x JOIN (SELECT 1 AS x) t571 ON t1.x = t571.x JOIN (SELECT 1 AS x) t572 ON t1.x = t572.x JOIN (SELECT 1 AS x) t573 ON t1.x = t573.x JOIN (SELECT 1 AS x) t574 ON t1.x = t574.x JOIN (SELECT 1 AS x) t575 ON t1.x = t575.x JOIN (SELECT 1 AS x) t576 ON t1.x = t576.x JOIN (SELECT 1 AS x) t577 ON t1.x = t577.x JOIN (SELECT 1 AS x) t578 ON t1.x = t578.x JOIN (SELECT 1 AS x) t579 ON t1.x = t579.x JOIN (SELECT 1 AS x) t580 ON t1.x = t580.x JOIN (SELECT 1 AS x) t581 ON t1.x = t581.x JOIN (SELECT 1 AS x) t582 ON t1.x = t582.x JOIN (SELECT 1 AS x) t583 ON t1.x = t583.x JOIN (SELECT 1 AS x) t584 ON t1.x = t584.x JOIN (SELECT 1 AS x) t585 ON t1.x = t585.x JOIN (SELECT 1 AS x) t586 ON t1.x = t586.x JOIN (SELECT 1 AS x) t587 ON t1.x = t587.x JOIN (SELECT 1 AS x) t588 ON t1.x = t588.x JOIN (SELECT 1 AS x) t589 ON t1.x = t589.x JOIN (SELECT 1 AS x) t590 ON t1.x = t590.x JOIN (SELECT 1 AS x) t591 ON t1.x = t591.x JOIN (SELECT 1 AS x) t592 ON t1.x = t592.x JOIN (SELECT 1 AS x) t593 ON t1.x = t593.x JOIN (SELECT 1 AS x) t594 ON t1.x = t594.x JOIN (SELECT 1 AS x) t595 ON t1.x = t595.x JOIN (SELECT 1 AS x) t596 ON t1.x = t596.x JOIN (SELECT 1 AS x) t597 ON t1.x = t597.x JOIN (SELECT 1 AS x) t598 ON t1.x = t598.x JOIN (SELECT 1 AS x) t599 ON t1.x = t599.x JOIN (SELECT 1 AS x) t600 ON t1.x = t600.x JOIN (SELECT 1 AS x) t601 ON t1.x = t601.x JOIN (SELECT 1 AS x) t602 ON t1.x = t602.x JOIN (SELECT 1 AS x) t603 ON t1.x = t603.x JOIN (SELECT 1 AS x) t604 ON t1.x = t604.x JOIN (SELECT 1 AS x) t605 ON t1.x = t605.x JOIN (SELECT 1 AS x) t606 ON t1.x = t606.x JOIN (SELECT 1 AS x) t607 ON t1.x = t607.x JOIN (SELECT 1 AS x) t608 ON t1.x = t608.x JOIN (SELECT 1 AS x) t609 ON t1.x = t609.x JOIN (SELECT 1 AS x) t610 ON t1.x = t610.x JOIN (SELECT 1 AS x) t611 ON t1.x = t611.x JOIN (SELECT 1 AS x) t612 ON t1.x = t612.x JOIN (SELECT 1 AS x) t613 ON t1.x = t613.x JOIN (SELECT 1 AS x) t614 ON t1.x = t614.x JOIN (SELECT 1 AS x) t615 ON t1.x = t615.x JOIN (SELECT 1 AS x) t616 ON t1.x = t616.x JOIN (SELECT 1 AS x) t617 ON t1.x = t617.x JOIN (SELECT 1 AS x) t618 ON t1.x = t618.x JOIN (SELECT 1 AS x) t619 ON t1.x = t619.x JOIN (SELECT 1 AS x) t620 ON t1.x = t620.x JOIN (SELECT 1 AS x) t621 ON t1.x = t621.x JOIN (SELECT 1 AS x) t622 ON t1.x = t622.x JOIN (SELECT 1 AS x) t623 ON t1.x = t623.x JOIN (SELECT 1 AS x) t624 ON t1.x = t624.x JOIN (SELECT 1 AS x) t625 ON t1.x = t625.x JOIN (SELECT 1 AS x) t626 ON t1.x = t626.x JOIN (SELECT 1 AS x) t627 ON t1.x = t627.x JOIN (SELECT 1 AS x) t628 ON t1.x = t628.x JOIN (SELECT 1 AS x) t629 ON t1.x = t629.x JOIN (SELECT 1 AS x) t630 ON t1.x = t630.x JOIN (SELECT 1 AS x) t631 ON t1.x = t631.x JOIN (SELECT 1 AS x) t632 ON t1.x = t632.x JOIN (SELECT 1 AS x) t633 ON t1.x = t633.x JOIN (SELECT 1 AS x) t634 ON t1.x = t634.x JOIN (SELECT 1 AS x) t635 ON t1.x = t635.x JOIN (SELECT 1 AS x) t636 ON t1.x = t636.x JOIN (SELECT 1 AS x) t637 ON t1.x = t637.x JOIN (SELECT 1 AS x) t638 ON t1.x = t638.x JOIN (SELECT 1 AS x) t639 ON t1.x = t639.x JOIN (SELECT 1 AS x) t640 ON t1.x = t640.x JOIN (SELECT 1 AS x) t641 ON t1.x = t641.x JOIN (SELECT 1 AS x) t642 ON t1.x = t642.x JOIN (SELECT 1 AS x) t643 ON t1.x = t643.x JOIN (SELECT 1 AS x) t644 ON t1.x = t644.x JOIN (SELECT 1 AS x) t645 ON t1.x = t645.x JOIN (SELECT 1 AS x) t646 ON t1.x = t646.x JOIN (SELECT 1 AS x) t647 ON t1.x = t647.x JOIN (SELECT 1 AS x) t648 ON t1.x = t648.x JOIN (SELECT 1 AS x) t649 ON t1.x = t649.x JOIN (SELECT 1 AS x) t650 ON t1.x = t650.x JOIN (SELECT 1 AS x) t651 ON t1.x = t651.x JOIN (SELECT 1 AS x) t652 ON t1.x = t652.x JOIN (SELECT 1 AS x) t653 ON t1.x = t653.x JOIN (SELECT 1 AS x) t654 ON t1.x = t654.x JOIN (SELECT 1 AS x) t655 ON t1.x = t655.x JOIN (SELECT 1 AS x) t656 ON t1.x = t656.x JOIN (SELECT 1 AS x) t657 ON t1.x = t657.x JOIN (SELECT 1 AS x) t658 ON t1.x = t658.x JOIN (SELECT 1 AS x) t659 ON t1.x = t659.x JOIN (SELECT 1 AS x) t660 ON t1.x = t660.x JOIN (SELECT 1 AS x) t661 ON t1.x = t661.x JOIN (SELECT 1 AS x) t662 ON t1.x = t662.x JOIN (SELECT 1 AS x) t663 ON t1.x = t663.x JOIN (SELECT 1 AS x) t664 ON t1.x = t664.x JOIN (SELECT 1 AS x) t665 ON t1.x = t665.x JOIN (SELECT 1 AS x) t666 ON t1.x = t666.x From e60ae9c64a237d0a7c9fba5a1e83ff611e0f8c58 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 12:44:02 +0000 Subject: [PATCH 125/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index e1860d60081..2e779401e0b 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -158,7 +158,7 @@ def run_fuzzer(fuzzer: str, timeout: int): with open(f"{new_corpus_dir}/testfile", "a", encoding="ascii") as f: f.write("Now the file has more content!") - s3.upload_build_directory_to_s3(new_corpus_dir, "fuzzer/corpus/") + s3.upload_build_directory_to_s3(new_corpus_dir, Path("fuzzer/corpus/")) def main(): From de69aa8c946258ebd25fc4e0a131b0244f5cbac1 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 13:42:24 +0000 Subject: [PATCH 126/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 2e779401e0b..c6b1b2a623b 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -158,7 +158,7 @@ def run_fuzzer(fuzzer: str, timeout: int): with open(f"{new_corpus_dir}/testfile", "a", encoding="ascii") as f: f.write("Now the file has more content!") - s3.upload_build_directory_to_s3(new_corpus_dir, Path("fuzzer/corpus/")) + s3.upload_build_directory_to_s3(Path(new_corpus_dir), "fuzzer/corpus/") def main(): From da5ebde4d5db8d2838c4473fe21e69d3b5a9ae4e Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 16:16:39 +0000 Subject: [PATCH 127/680] add CI env --- tests/ci/libfuzzer_test_check.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 8f19dd7d023..46406dc3557 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -133,6 +133,8 @@ def main(): check_name, run_by_hash_num, run_by_hash_total ) + additional_envs.append("CI=1") + ci_logs_credentials = CiLogsCredentials(Path(temp_path) / "export-logs-config.sh") ci_logs_args = ci_logs_credentials.get_docker_arguments( pr_info, stopwatch.start_time_str, check_name From 3fb92a61a0c115fd564913fa918acf1c0e5db987 Mon Sep 17 00:00:00 2001 From: vdimir Date: Thu, 3 Oct 2024 16:49:19 +0000 Subject: [PATCH 128/680] t --- src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp | 5 +++++ tests/integration/helpers/cluster.py | 2 ++ 2 files changed, 7 insertions(+) diff --git a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp index cd66a230038..d0f4371fac6 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp @@ -56,6 +56,11 @@ void optimizeJoin(QueryPlan::Node & node, QueryPlan::Nodes &) return; const auto & table_join = join->getTableJoin(); + + /// Algorithms other than HashJoin may not support OUTER JOINs + if (table_join.kind() != JoinKind::Inner && !typeid_cast(join.get())) + return; + /// fixme: USING clause handled specially in join algorithm, so swap breaks it /// fixme: Swapping for SEMI and ANTI joins should be alright, need to try to enable it and test if (table_join.hasUsing() || table_join.strictness() != JoinStrictness::All) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index 8cf3e318797..8fe2932137c 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -4524,6 +4524,8 @@ class ClickHouseInstance: ): # If custom main config is used, do not apply random settings to it write_random_settings_config(Path(users_d_dir) / "0_random_settings.xml") + else: + print(f"XXXX Skip random settings for {self.name}, {self.randomize_settings} {self.image}:{self.tag} @ {self.base_config_dir} ?= {DEFAULT_BASE_CONFIG_DIR}") version = None version_parts = self.tag.split(".") From 2c8c5629d95941da2cef30ce373751a83a95b8d4 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 3 Oct 2024 16:57:19 +0000 Subject: [PATCH 129/680] Automatic style fix --- tests/integration/helpers/cluster.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index 8fe2932137c..f9d3746b2d9 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -4525,7 +4525,9 @@ class ClickHouseInstance: # If custom main config is used, do not apply random settings to it write_random_settings_config(Path(users_d_dir) / "0_random_settings.xml") else: - print(f"XXXX Skip random settings for {self.name}, {self.randomize_settings} {self.image}:{self.tag} @ {self.base_config_dir} ?= {DEFAULT_BASE_CONFIG_DIR}") + print( + f"XXXX Skip random settings for {self.name}, {self.randomize_settings} {self.image}:{self.tag} @ {self.base_config_dir} ?= {DEFAULT_BASE_CONFIG_DIR}" + ) version = None version_parts = self.tag.split(".") From 4d917d80b42f8dedbd4ddbfecc1c6d9c5fa87c01 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 17:32:05 +0000 Subject: [PATCH 130/680] fix --- tests/ci/libfuzzer_test_check.py | 2 +- tests/fuzz/runner.py | 5 +---- 2 files changed, 2 insertions(+), 5 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 46406dc3557..5de28d5641a 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -59,7 +59,7 @@ def get_run_command( envs = [ # a static link, don't use S3_URL or S3_DOWNLOAD - '-e S3_URL="https://s3.amazonaws.com/clickhouse-datasets"', + '-e S3_URL="https://s3.amazonaws.com"', ] envs += [f"-e {e}" for e in additional_envs] diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index c6b1b2a623b..5d0f2865422 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -155,10 +155,7 @@ def run_fuzzer(fuzzer: str, timeout: int): else: process_fuzzer_output(result.stderr) - with open(f"{new_corpus_dir}/testfile", "a", encoding="ascii") as f: - f.write("Now the file has more content!") - - s3.upload_build_directory_to_s3(Path(new_corpus_dir), "fuzzer/corpus/") + s3.upload_build_directory_to_s3(Path(new_corpus_dir), f"fuzzer/corpus/{fuzzer}", False) def main(): From f66bc05c0188d5873696d01b2d80486c73625bb2 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 3 Oct 2024 17:39:14 +0000 Subject: [PATCH 131/680] Automatic style fix --- tests/fuzz/runner.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 5d0f2865422..2e7c1184bcc 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -155,7 +155,9 @@ def run_fuzzer(fuzzer: str, timeout: int): else: process_fuzzer_output(result.stderr) - s3.upload_build_directory_to_s3(Path(new_corpus_dir), f"fuzzer/corpus/{fuzzer}", False) + s3.upload_build_directory_to_s3( + Path(new_corpus_dir), f"fuzzer/corpus/{fuzzer}", False + ) def main(): From 6fa23c4b72747293d58aebb11d1bb7d2a15b4647 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 23:44:40 +0000 Subject: [PATCH 132/680] kill all fuzzers on timeout --- tests/fuzz/runner.py | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 2e7c1184bcc..42e54acfecc 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -4,6 +4,7 @@ import configparser import logging import os import re +import signal import subprocess from pathlib import Path @@ -56,6 +57,15 @@ def process_error(error: str): is_call_stack = True +def kill_fuzzer(fuzzer: str): + p = subprocess.Popen(['ps', '-A'], stdout=subprocess.PIPE) + out, err = p.communicate() + for line in out.splitlines(): + if fuzzer in line: + pid = int(line.split(None, 1)[0]) + os.kill(pid, signal.SIGKILL) + + def run_fuzzer(fuzzer: str, timeout: int): s3 = S3Helper() @@ -151,6 +161,7 @@ def run_fuzzer(fuzzer: str, timeout: int): process_error(e.stderr) except subprocess.TimeoutExpired as e: print("Timeout for ", cmd_line) + kill_fuzzer(fuzzer) process_fuzzer_output(e.stderr) else: process_fuzzer_output(result.stderr) From a0d2f2085d56252eb689a72909b567db9325fdc1 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 3 Oct 2024 23:57:05 +0000 Subject: [PATCH 133/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 42e54acfecc..512a20e58c5 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -58,7 +58,7 @@ def process_error(error: str): def kill_fuzzer(fuzzer: str): - p = subprocess.Popen(['ps', '-A'], stdout=subprocess.PIPE) + p = subprocess.Popen(["ps", "-A"], stdout=subprocess.PIPE) out, err = p.communicate() for line in out.splitlines(): if fuzzer in line: From 08d098a2f486fab845fa46459b5e842132028ea4 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 4 Oct 2024 00:15:36 +0000 Subject: [PATCH 134/680] fix --- tests/fuzz/runner.py | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 512a20e58c5..8e05625a6d9 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -58,12 +58,12 @@ def process_error(error: str): def kill_fuzzer(fuzzer: str): - p = subprocess.Popen(["ps", "-A"], stdout=subprocess.PIPE) - out, err = p.communicate() - for line in out.splitlines(): - if fuzzer in line: - pid = int(line.split(None, 1)[0]) - os.kill(pid, signal.SIGKILL) + with subprocess.Popen(["ps", "-A"], stdout=subprocess.PIPE) as p + out, _ = p.communicate() + for line in out.splitlines(): + if fuzzer in line: + pid = int(line.split(None, 1)[0]) + os.kill(pid, signal.SIGKILL) def run_fuzzer(fuzzer: str, timeout: int): From bfb2e7c04413f467e310231830f6701b39739e5e Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 4 Oct 2024 00:16:16 +0000 Subject: [PATCH 135/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 8e05625a6d9..81a76fbcdb9 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -58,7 +58,7 @@ def process_error(error: str): def kill_fuzzer(fuzzer: str): - with subprocess.Popen(["ps", "-A"], stdout=subprocess.PIPE) as p + with subprocess.Popen(["ps", "-A"], stdout=subprocess.PIPE) as p: out, _ = p.communicate() for line in out.splitlines(): if fuzzer in line: From 9d81ff0a8906ed5549ca3a75a0540b4fb0e13dfc Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 4 Oct 2024 01:22:26 +0000 Subject: [PATCH 136/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 81a76fbcdb9..702014ce04f 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -185,7 +185,7 @@ def main(): with Path() as current: for fuzzer in current.iterdir(): if (current / fuzzer).is_file() and os.access(current / fuzzer, os.X_OK): - run_fuzzer(fuzzer, timeout) + run_fuzzer(fuzzer.name, timeout) if __name__ == "__main__": From 5cf7a777a2b7bf80c9a3eba1d89a5a3bbfa2c86f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 4 Oct 2024 02:31:34 +0000 Subject: [PATCH 137/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 702014ce04f..b51b0f99abc 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -61,7 +61,7 @@ def kill_fuzzer(fuzzer: str): with subprocess.Popen(["ps", "-A"], stdout=subprocess.PIPE) as p: out, _ = p.communicate() for line in out.splitlines(): - if fuzzer in line: + if fuzzer.encode("utf-8") in line: pid = int(line.split(None, 1)[0]) os.kill(pid, signal.SIGKILL) From db69e018bf31acf0ec0c22e63bebe1448429e4fc Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 4 Oct 2024 03:18:01 +0000 Subject: [PATCH 138/680] fix --- tests/fuzz/runner.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index b51b0f99abc..bcfc7e6146f 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -63,6 +63,7 @@ def kill_fuzzer(fuzzer: str): for line in out.splitlines(): if fuzzer.encode("utf-8") in line: pid = int(line.split(None, 1)[0]) + logging.info("Killing fuzzer %s, pid %d", fuzzer, pid) os.kill(pid, signal.SIGKILL) From 530d034302720ec3c479e38ba18ac432e27f6ab3 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 4 Oct 2024 04:35:35 +0000 Subject: [PATCH 139/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index bcfc7e6146f..948bc9d48ed 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -161,7 +161,7 @@ def run_fuzzer(fuzzer: str, timeout: int): print("Stderr output: ", e.stderr) process_error(e.stderr) except subprocess.TimeoutExpired as e: - print("Timeout for ", cmd_line) + logging.info("Timeout for %s", cmd_line) kill_fuzzer(fuzzer) process_fuzzer_output(e.stderr) else: From e9e35eb118f35ecfa0b6d21fe4a9be7e87443a1f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 4 Oct 2024 05:31:17 +0000 Subject: [PATCH 140/680] fix --- tests/fuzz/runner.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 948bc9d48ed..ac2bb78b7f0 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -7,6 +7,7 @@ import re import signal import subprocess from pathlib import Path +from time import sleep from botocore.exceptions import ClientError @@ -163,6 +164,7 @@ def run_fuzzer(fuzzer: str, timeout: int): except subprocess.TimeoutExpired as e: logging.info("Timeout for %s", cmd_line) kill_fuzzer(fuzzer) + sleep(10) process_fuzzer_output(e.stderr) else: process_fuzzer_output(result.stderr) From 0872cc0dd7a78b26fb16c98c04894bf168bd199a Mon Sep 17 00:00:00 2001 From: vdimir Date: Fri, 4 Oct 2024 10:05:16 +0000 Subject: [PATCH 141/680] fix inegration settings randomization with non default tag --- tests/integration/helpers/cluster.py | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index f9d3746b2d9..5fa4c5dfce3 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -67,6 +67,7 @@ DEFAULT_ENV_NAME = ".env" DEFAULT_BASE_CONFIG_DIR = os.environ.get( "CLICKHOUSE_TESTS_BASE_CONFIG_DIR", "/etc/clickhouse-server/" ) +DOCKER_BASE_TAG = os.environ.get("DOCKER_BASE_TAG", "latest") SANITIZER_SIGN = "==================" @@ -504,7 +505,6 @@ class ClickHouseCluster: "CLICKHOUSE_TESTS_DOCKERD_HOST" ) self.docker_api_version = os.environ.get("DOCKER_API_VERSION") - self.docker_base_tag = os.environ.get("DOCKER_BASE_TAG", "latest") self.base_cmd = ["docker", "compose"] if custom_dockerd_host: @@ -1079,7 +1079,7 @@ class ClickHouseCluster: env_variables["keeper_binary"] = binary_path env_variables["keeper_cmd_prefix"] = keeper_cmd_prefix - env_variables["image"] = "clickhouse/integration-test:" + self.docker_base_tag + env_variables["image"] = "clickhouse/integration-test:" + DOCKER_BASE_TAG env_variables["user"] = str(os.getuid()) env_variables["keeper_fs"] = "bind" for i in range(1, 4): @@ -1672,7 +1672,7 @@ class ClickHouseCluster: ) if tag is None: - tag = self.docker_base_tag + tag = DOCKER_BASE_TAG if not env_variables: env_variables = {} self.use_keeper = use_keeper @@ -4519,15 +4519,11 @@ class ClickHouseInstance: if ( self.randomize_settings and self.image == "clickhouse/integration-test" - and self.tag == "latest" + and self.tag == DOCKER_BASE_TAG and self.base_config_dir == DEFAULT_BASE_CONFIG_DIR ): # If custom main config is used, do not apply random settings to it write_random_settings_config(Path(users_d_dir) / "0_random_settings.xml") - else: - print( - f"XXXX Skip random settings for {self.name}, {self.randomize_settings} {self.image}:{self.tag} @ {self.base_config_dir} ?= {DEFAULT_BASE_CONFIG_DIR}" - ) version = None version_parts = self.tag.split(".") From c91b0563de2ffb81fa5c10655c8711c894792aac Mon Sep 17 00:00:00 2001 From: vdimir Date: Fri, 4 Oct 2024 11:05:21 +0000 Subject: [PATCH 142/680] materialize block in JoiningTransform::transformHeader --- src/Processors/Transforms/JoiningTransform.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/src/Processors/Transforms/JoiningTransform.cpp b/src/Processors/Transforms/JoiningTransform.cpp index f2fb6327129..187f4bf6728 100644 --- a/src/Processors/Transforms/JoiningTransform.cpp +++ b/src/Processors/Transforms/JoiningTransform.cpp @@ -19,6 +19,7 @@ Block JoiningTransform::transformHeader(Block header, const JoinPtr & join) join->initialize(header); ExtraBlockPtr tmp; join->joinBlock(header, tmp); + materializeBlockInplace(header); LOG_TEST(getLogger("JoiningTransform"), "After join block: '{}'", header.dumpStructure()); return header; } From eb8ae504db5b7d04ff1d9f04f6068e91472153eb Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 4 Oct 2024 12:03:21 +0000 Subject: [PATCH 143/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index ac2bb78b7f0..e842f40f8d8 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -59,7 +59,7 @@ def process_error(error: str): def kill_fuzzer(fuzzer: str): - with subprocess.Popen(["ps", "-A"], stdout=subprocess.PIPE) as p: + with subprocess.Popen(["ps", "-A", "u"], stdout=subprocess.PIPE) as p: out, _ = p.communicate() for line in out.splitlines(): if fuzzer.encode("utf-8") in line: From c7902255ba868af3903e075bb69e27381f062351 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 4 Oct 2024 12:54:13 +0000 Subject: [PATCH 144/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index e842f40f8d8..b3c19fbb0a4 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -63,7 +63,7 @@ def kill_fuzzer(fuzzer: str): out, _ = p.communicate() for line in out.splitlines(): if fuzzer.encode("utf-8") in line: - pid = int(line.split(None, 1)[0]) + pid = int(line.split(None, 2)[1]) logging.info("Killing fuzzer %s, pid %d", fuzzer, pid) os.kill(pid, signal.SIGKILL) From c555eb4ba50734f9c3a760af44bd1edb702d26f1 Mon Sep 17 00:00:00 2001 From: vdimir Date: Fri, 4 Oct 2024 13:24:54 +0000 Subject: [PATCH 145/680] optimize join step planning a bit --- src/Planner/PlannerJoinTree.cpp | 92 +++++++++++-------- src/Processors/QueryPlan/JoinStep.cpp | 11 ++- src/Processors/QueryPlan/JoinStep.h | 6 +- .../QueryPlan/Optimizations/optimizeJoin.cpp | 3 +- 4 files changed, 70 insertions(+), 42 deletions(-) diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp index 720f0a380ab..19fd896f9a8 100644 --- a/src/Planner/PlannerJoinTree.cpp +++ b/src/Planner/PlannerJoinTree.cpp @@ -1241,6 +1241,55 @@ void joinCastPlanColumnsToNullable(QueryPlan & plan_to_add_cast, PlannerContextP plan_to_add_cast.addStep(std::move(cast_join_columns_step)); } +std::optional createStepToDropColumns( + const Block & header, + const ColumnIdentifierSet & outer_scope_columns, + const PlannerContextPtr & planner_context) +{ + ActionsDAG drop_unused_columns_after_join_actions_dag(header.getColumnsWithTypeAndName()); + ActionsDAG::NodeRawConstPtrs drop_unused_columns_after_join_actions_dag_updated_outputs; + std::unordered_set drop_unused_columns_after_join_actions_dag_updated_outputs_names; + std::optional first_skipped_column_node_index; + + auto & drop_unused_columns_after_join_actions_dag_outputs = drop_unused_columns_after_join_actions_dag.getOutputs(); + size_t drop_unused_columns_after_join_actions_dag_outputs_size = drop_unused_columns_after_join_actions_dag_outputs.size(); + + const auto & global_planner_context = planner_context->getGlobalPlannerContext(); + + for (size_t i = 0; i < drop_unused_columns_after_join_actions_dag_outputs_size; ++i) + { + const auto & output = drop_unused_columns_after_join_actions_dag_outputs[i]; + + if (drop_unused_columns_after_join_actions_dag_updated_outputs_names.contains(output->result_name) + || !global_planner_context->hasColumnIdentifier(output->result_name)) + continue; + + if (!outer_scope_columns.contains(output->result_name)) + { + if (!first_skipped_column_node_index) + first_skipped_column_node_index = i; + continue; + } + + drop_unused_columns_after_join_actions_dag_updated_outputs.push_back(output); + drop_unused_columns_after_join_actions_dag_updated_outputs_names.insert(output->result_name); + } + + if (!first_skipped_column_node_index) + return {}; + + /** It is expected that JOIN TREE query plan will contain at least 1 column, even if there are no columns in outer scope. + * + * Example: SELECT count() FROM test_table_1 AS t1, test_table_2 AS t2; + */ + if (drop_unused_columns_after_join_actions_dag_updated_outputs.empty() && first_skipped_column_node_index) + drop_unused_columns_after_join_actions_dag_updated_outputs.push_back(drop_unused_columns_after_join_actions_dag_outputs[*first_skipped_column_node_index]); + + drop_unused_columns_after_join_actions_dag_outputs = std::move(drop_unused_columns_after_join_actions_dag_updated_outputs); + + return drop_unused_columns_after_join_actions_dag; +} + JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_expression, JoinTreeQueryPlan left_join_tree_query_plan, JoinTreeQueryPlan right_join_tree_query_plan, @@ -1654,47 +1703,18 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ result_plan.unitePlans(std::move(join_step), {std::move(plans)}); } - ActionsDAG drop_unused_columns_after_join_actions_dag(result_plan.getCurrentDataStream().header.getColumnsWithTypeAndName()); - ActionsDAG::NodeRawConstPtrs drop_unused_columns_after_join_actions_dag_updated_outputs; - std::unordered_set drop_unused_columns_after_join_actions_dag_updated_outputs_names; - std::optional first_skipped_column_node_index; - - auto & drop_unused_columns_after_join_actions_dag_outputs = drop_unused_columns_after_join_actions_dag.getOutputs(); - size_t drop_unused_columns_after_join_actions_dag_outputs_size = drop_unused_columns_after_join_actions_dag_outputs.size(); - - for (size_t i = 0; i < drop_unused_columns_after_join_actions_dag_outputs_size; ++i) + const auto & header_after_join = result_plan.getCurrentDataStream().header; + if (header_after_join.columns() > outer_scope_columns.size()) { - const auto & output = drop_unused_columns_after_join_actions_dag_outputs[i]; - - const auto & global_planner_context = planner_context->getGlobalPlannerContext(); - if (drop_unused_columns_after_join_actions_dag_updated_outputs_names.contains(output->result_name) - || !global_planner_context->hasColumnIdentifier(output->result_name)) - continue; - - if (!outer_scope_columns.contains(output->result_name)) + auto drop_unused_columns_after_join_actions_dag = createStepToDropColumns(header_after_join, outer_scope_columns, planner_context); + if (drop_unused_columns_after_join_actions_dag) { - if (!first_skipped_column_node_index) - first_skipped_column_node_index = i; - continue; + auto drop_unused_columns_after_join_transform_step = std::make_unique(result_plan.getCurrentDataStream(), std::move(*drop_unused_columns_after_join_actions_dag)); + drop_unused_columns_after_join_transform_step->setStepDescription("Drop unused columns after JOIN"); + result_plan.addStep(std::move(drop_unused_columns_after_join_transform_step)); } - - drop_unused_columns_after_join_actions_dag_updated_outputs.push_back(output); - drop_unused_columns_after_join_actions_dag_updated_outputs_names.insert(output->result_name); } - /** It is expected that JOIN TREE query plan will contain at least 1 column, even if there are no columns in outer scope. - * - * Example: SELECT count() FROM test_table_1 AS t1, test_table_2 AS t2; - */ - if (drop_unused_columns_after_join_actions_dag_updated_outputs.empty() && first_skipped_column_node_index) - drop_unused_columns_after_join_actions_dag_updated_outputs.push_back(drop_unused_columns_after_join_actions_dag_outputs[*first_skipped_column_node_index]); - - drop_unused_columns_after_join_actions_dag_outputs = std::move(drop_unused_columns_after_join_actions_dag_updated_outputs); - - auto drop_unused_columns_after_join_transform_step = std::make_unique(result_plan.getCurrentDataStream(), std::move(drop_unused_columns_after_join_actions_dag)); - drop_unused_columns_after_join_transform_step->setStepDescription("DROP unused columns after JOIN"); - result_plan.addStep(std::move(drop_unused_columns_after_join_transform_step)); - for (const auto & right_join_tree_query_plan_row_policy : right_join_tree_query_plan.used_row_policies) left_join_tree_query_plan.used_row_policies.insert(right_join_tree_query_plan_row_policy); diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index d6f9590d240..3edc64ef967 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -185,8 +185,18 @@ void JoinStep::describeActions(JSONBuilder::JSONMap & map) const map.add(name, value); } +void JoinStep::setJoin(JoinPtr join_, bool swap_streams_) +{ + join_algorithm_header.clear(); + swap_streams = swap_streams_; + join = std::move(join_); +} + void JoinStep::updateOutputStream() { + if (join_algorithm_header) + return; + const auto & header = swap_streams ? input_streams[1].header : input_streams[0].header; Block result_header = JoiningTransform::transformHeader(header, join); @@ -200,7 +210,6 @@ void JoinStep::updateOutputStream() return; } - if (swap_streams) result_header = rotateBlock(result_header, input_streams[1].header); diff --git a/src/Processors/QueryPlan/JoinStep.h b/src/Processors/QueryPlan/JoinStep.h index b0947cb6be7..bf6560d5a07 100644 --- a/src/Processors/QueryPlan/JoinStep.h +++ b/src/Processors/QueryPlan/JoinStep.h @@ -34,13 +34,12 @@ public: void describeActions(FormatSettings & settings) const override; const JoinPtr & getJoin() const { return join; } - void setJoin(JoinPtr join_) { join = std::move(join_); } + void setJoin(JoinPtr join_, bool swap_streams_ = false); bool allowPushDownToRight() const; bool canUpdateInputStream() const override { return true; } JoinInnerTableSelectionMode inner_table_selection_mode = JoinInnerTableSelectionMode::Right; - bool swap_streams = false; private: void updateOutputStream() override; @@ -51,10 +50,11 @@ private: size_t max_block_size; size_t max_streams; - NameSet required_output; + const NameSet required_output; std::set columns_to_remove; bool keep_left_read_in_order; bool use_new_analyzer = false; + bool swap_streams = false; }; /// Special step for the case when Join is already filled. diff --git a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp index d0f4371fac6..ced3b987b64 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp @@ -92,12 +92,11 @@ void optimizeJoin(QueryPlan::Node & node, QueryPlan::Nodes &) const auto & left_stream_input_header = streams.front().header; const auto & right_stream_input_header = streams.back().header; - join_step->swap_streams = true; auto updated_table_join = std::make_shared(table_join); updated_table_join->swapSides(); auto updated_join = join->clone(updated_table_join, right_stream_input_header, left_stream_input_header); - join_step->setJoin(std::move(updated_join)); + join_step->setJoin(std::move(updated_join), /* swap_streams= */ true); } } From 2f923ee24278a22e2c78d957f76077dff21176a5 Mon Sep 17 00:00:00 2001 From: avogar Date: Fri, 4 Oct 2024 14:36:28 +0000 Subject: [PATCH 146/680] Fix old analyzer --- src/Interpreters/ExpressionAnalyzer.cpp | 17 +++++++++++++---- 1 file changed, 13 insertions(+), 4 deletions(-) diff --git a/src/Interpreters/ExpressionAnalyzer.cpp b/src/Interpreters/ExpressionAnalyzer.cpp index 12e769f249a..5913cf644d8 100644 --- a/src/Interpreters/ExpressionAnalyzer.cpp +++ b/src/Interpreters/ExpressionAnalyzer.cpp @@ -1372,6 +1372,7 @@ bool SelectQueryExpressionAnalyzer::appendGroupBy(ExpressionActionsChain & chain ExpressionActionsChain::Step & step = chain.lastStep(columns_after_join); ASTs asts = select_query->groupBy()->children; + NameSet group_by_keys; if (select_query->group_by_with_grouping_sets) { for (const auto & ast : asts) @@ -1379,6 +1380,7 @@ bool SelectQueryExpressionAnalyzer::appendGroupBy(ExpressionActionsChain & chain for (const auto & ast_element : ast->children) { step.addRequiredOutput(ast_element->getColumnName()); + group_by_keys.insert(ast_element->getColumnName()); getRootActions(ast_element, only_types, step.actions()->dag); } } @@ -1388,12 +1390,16 @@ bool SelectQueryExpressionAnalyzer::appendGroupBy(ExpressionActionsChain & chain for (const auto & ast : asts) { step.addRequiredOutput(ast->getColumnName()); + group_by_keys.insert(ast->getColumnName()); getRootActions(ast, only_types, step.actions()->dag); } } for (const auto & result_column : step.getResultColumns()) - validateGroupByKeyType(result_column.type); + { + if (group_by_keys.contains(result_column.name)) + validateGroupByKeyType(result_column.type); + } if (optimize_aggregation_in_order) { @@ -1612,9 +1618,6 @@ ActionsAndProjectInputsFlagPtr SelectQueryExpressionAnalyzer::appendOrderBy( getRootActions(select_query->orderBy(), only_types, step.actions()->dag); - for (const auto & result_column : step.getResultColumns()) - validateOrderByKeyType(result_column.type); - bool with_fill = false; for (auto & child : select_query->orderBy()->children) @@ -1629,6 +1632,12 @@ ActionsAndProjectInputsFlagPtr SelectQueryExpressionAnalyzer::appendOrderBy( with_fill = true; } + for (const auto & result_column : step.getResultColumns()) + { + if (order_by_keys.contains(result_column.name)) + validateOrderByKeyType(result_column.type); + } + if (auto interpolate_list = select_query->interpolate()) { From 93620886f689d3b6ed59a9a36539c2902158e6b1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Sun, 6 Oct 2024 22:16:06 +0300 Subject: [PATCH 147/680] Revert part actual name to pass the check --- src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp index 3f5c70adb64..4a994bc38e2 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp @@ -844,8 +844,9 @@ std::pair, bool> ReplicatedMergeTreeSinkImpl:: } } - /// Save the current temporary path in case we need to revert the change to retry (ZK connection loss) + /// Save the current temporary path and name in case we need to revert the change to retry (ZK connection loss) or in case part is deduplicated. const String temporary_part_relative_path = part->getDataPartStorage().getPartDirectory(); + const String initial_part_name = part->name; /// Obtain incremental block number and lock it. The lock holds our intention to add the block to the filesystem. /// We remove the lock just after renaming the part. In case of exception, block number will be marked as abandoned. @@ -1024,6 +1025,7 @@ std::pair, bool> ReplicatedMergeTreeSinkImpl:: transaction.rollbackPartsToTemporaryState(); part->is_temp = true; + part->setName(initial_part_name); part->renameTo(temporary_part_relative_path, false); if constexpr (async_insert) From 91931b5b3cccabdc94231a8a07ad4d2e8de8d8b9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Sun, 6 Oct 2024 22:56:48 +0300 Subject: [PATCH 148/680] Fix style --- tests/integration/test_deduplicated_attached_part_rename/test.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/integration/test_deduplicated_attached_part_rename/test.py b/tests/integration/test_deduplicated_attached_part_rename/test.py index 7afd85c62dc..02fa2c9d4a3 100644 --- a/tests/integration/test_deduplicated_attached_part_rename/test.py +++ b/tests/integration/test_deduplicated_attached_part_rename/test.py @@ -1,4 +1,5 @@ import pytest + from helpers.cluster import ClickHouseCluster cluster = ClickHouseCluster(__file__) From 52484cbfec0c168bb440d623673aeb321e1c0211 Mon Sep 17 00:00:00 2001 From: Pavel Kruglov <48961922+Avogar@users.noreply.github.com> Date: Mon, 7 Oct 2024 12:45:23 +0800 Subject: [PATCH 149/680] Fix tests --- tests/queries/0_stateless/01825_new_type_json_ghdata.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/01825_new_type_json_ghdata.sh b/tests/queries/0_stateless/01825_new_type_json_ghdata.sh index 6a4fc7d5935..cabc3efcd8e 100755 --- a/tests/queries/0_stateless/01825_new_type_json_ghdata.sh +++ b/tests/queries/0_stateless/01825_new_type_json_ghdata.sh @@ -16,7 +16,7 @@ ${CLICKHOUSE_CLIENT} -q "SELECT count() FROM ghdata WHERE NOT ignore(*)" ${CLICKHOUSE_CLIENT} -q \ "SELECT data.repo.name, count() AS stars FROM ghdata \ - WHERE data.type = 'WatchEvent' GROUP BY data.repo.name ORDER BY stars DESC, data.repo.name LIMIT 5" + WHERE data.type = 'WatchEvent' GROUP BY data.repo.name ORDER BY stars DESC, data.repo.name LIMIT 5" --allow_suspicious_types_in_order_by 1 --allow_suspicious_types_in_group_by 1 ${CLICKHOUSE_CLIENT} --enable_analyzer=1 -q \ "SELECT data.payload.commits[].author.name AS name, count() AS c FROM ghdata \ From 7808f00857a157e2b49606df6de567a63462aa58 Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 7 Oct 2024 06:53:12 +0000 Subject: [PATCH 150/680] Support alter from String to JSON --- src/Columns/ColumnArray.h | 7 + src/Columns/ColumnDynamic.cpp | 9 ++ src/Columns/ColumnDynamic.h | 1 + src/Columns/ColumnMap.cpp | 7 + src/Columns/ColumnMap.h | 1 + src/Columns/ColumnObject.cpp | 25 ++++ src/Columns/ColumnObject.h | 2 + src/Columns/ColumnTuple.cpp | 20 +++ src/Columns/ColumnTuple.h | 1 + src/Columns/ColumnVariant.cpp | 17 +++ src/Columns/ColumnVariant.h | 1 + src/Columns/IColumn.h | 3 + .../Serializations/SerializationDynamic.cpp | 45 +++--- .../Serializations/SerializationDynamic.h | 20 ++- .../Serializations/SerializationObject.cpp | 22 ++- .../Serializations/SerializationObject.h | 16 ++- src/Functions/FunctionsConversion.cpp | 5 +- src/Storages/AlterCommands.cpp | 14 +- .../MergeTreeDataPartWriterCompact.cpp | 31 ++-- .../MergeTreeDataPartWriterCompact.h | 6 +- .../MergeTreeDataPartWriterOnDisk.cpp | 39 +++++ .../MergeTree/MergeTreeDataPartWriterOnDisk.h | 12 ++ .../MergeTree/MergeTreeDataPartWriterWide.cpp | 32 ++--- .../MergeTree/MergeTreeDataPartWriterWide.h | 8 +- .../03246_alter_from_string_to_json.reference | 134 ++++++++++++++++++ .../03246_alter_from_string_to_json.sql.j2 | 32 +++++ ...3247_ghdata_string_to_json_alter.reference | 12 ++ .../03247_ghdata_string_to_json_alter.sh | 30 ++++ .../03248_string_to_json_alter_fuzz.reference | 0 .../03248_string_to_json_alter_fuzz.sql | 17 +++ 30 files changed, 459 insertions(+), 110 deletions(-) create mode 100644 tests/queries/0_stateless/03246_alter_from_string_to_json.reference create mode 100644 tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 create mode 100644 tests/queries/0_stateless/03247_ghdata_string_to_json_alter.reference create mode 100755 tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh create mode 100644 tests/queries/0_stateless/03248_string_to_json_alter_fuzz.reference create mode 100644 tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql diff --git a/src/Columns/ColumnArray.h b/src/Columns/ColumnArray.h index f77268a8be6..df52880d6e4 100644 --- a/src/Columns/ColumnArray.h +++ b/src/Columns/ColumnArray.h @@ -192,6 +192,13 @@ public: bool hasDynamicStructure() const override { return getData().hasDynamicStructure(); } void takeDynamicStructureFromSourceColumns(const Columns & source_columns) override; + bool dynamicStructureEquals(const IColumn & rhs) const override + { + if (const auto * rhs_concrete = typeid_cast(&rhs)) + return data->dynamicStructureEquals(*rhs_concrete->data); + return false; + } + private: WrappedPtr data; WrappedPtr offsets; diff --git a/src/Columns/ColumnDynamic.cpp b/src/Columns/ColumnDynamic.cpp index 5a837a62761..09a05e52c90 100644 --- a/src/Columns/ColumnDynamic.cpp +++ b/src/Columns/ColumnDynamic.cpp @@ -1153,6 +1153,15 @@ void ColumnDynamic::prepareVariantsForSquashing(const Columns & source_columns) } } +bool ColumnDynamic::dynamicStructureEquals(const IColumn & rhs) const +{ + if (const auto * rhs_concrete = typeid_cast(&rhs)) + return max_dynamic_types == rhs_concrete->max_dynamic_types && global_max_dynamic_types == rhs_concrete->global_max_dynamic_types + && variant_info.variant_name == rhs_concrete->variant_info.variant_name + && variant_column->dynamicStructureEquals(*rhs_concrete->variant_column); + return false; +} + void ColumnDynamic::takeDynamicStructureFromSourceColumns(const Columns & source_columns) { if (!empty()) diff --git a/src/Columns/ColumnDynamic.h b/src/Columns/ColumnDynamic.h index 17b0d80e5eb..9e8b1f79321 100644 --- a/src/Columns/ColumnDynamic.h +++ b/src/Columns/ColumnDynamic.h @@ -367,6 +367,7 @@ public: bool addNewVariant(const DataTypePtr & new_variant) { return addNewVariant(new_variant, new_variant->getName()); } bool hasDynamicStructure() const override { return true; } + bool dynamicStructureEquals(const IColumn & rhs) const override; void takeDynamicStructureFromSourceColumns(const Columns & source_columns) override; const StatisticsPtr & getStatistics() const { return statistics; } diff --git a/src/Columns/ColumnMap.cpp b/src/Columns/ColumnMap.cpp index 536da4d06d0..4e81191939f 100644 --- a/src/Columns/ColumnMap.cpp +++ b/src/Columns/ColumnMap.cpp @@ -330,6 +330,13 @@ bool ColumnMap::structureEquals(const IColumn & rhs) const return false; } +bool ColumnMap::dynamicStructureEquals(const IColumn & rhs) const +{ + if (const auto * rhs_map = typeid_cast(&rhs)) + return nested->dynamicStructureEquals(*rhs_map->nested); + return false; +} + ColumnPtr ColumnMap::compress() const { auto compressed = nested->compress(); diff --git a/src/Columns/ColumnMap.h b/src/Columns/ColumnMap.h index 39d15a586b9..8cb0b1680a7 100644 --- a/src/Columns/ColumnMap.h +++ b/src/Columns/ColumnMap.h @@ -120,6 +120,7 @@ public: ColumnPtr compress() const override; bool hasDynamicStructure() const override { return nested->hasDynamicStructure(); } + bool dynamicStructureEquals(const IColumn & rhs) const override; void takeDynamicStructureFromSourceColumns(const Columns & source_columns) override; }; diff --git a/src/Columns/ColumnObject.cpp b/src/Columns/ColumnObject.cpp index 3577ab1ec82..8e0182c7276 100644 --- a/src/Columns/ColumnObject.cpp +++ b/src/Columns/ColumnObject.cpp @@ -1299,6 +1299,31 @@ void ColumnObject::prepareForSquashing(const std::vector & source_col } } +bool ColumnObject::dynamicStructureEquals(const IColumn & rhs) const +{ + const auto * rhs_object = typeid_cast(&rhs); + if (!rhs_object || typed_paths.size() != rhs_object->typed_paths.size() + || global_max_dynamic_paths != rhs_object->global_max_dynamic_paths || max_dynamic_types != rhs_object->max_dynamic_types + || dynamic_paths.size() != rhs_object->dynamic_paths.size()) + return false; + + for (const auto & [path, column] : typed_paths) + { + auto it = rhs_object->typed_paths.find(path); + if (it == rhs_object->typed_paths.end() || !it->second->dynamicStructureEquals(*column)) + return false; + } + + for (const auto & [path, column] : dynamic_paths) + { + auto it = rhs_object->dynamic_paths.find(path); + if (it == rhs_object->dynamic_paths.end() || !it->second->dynamicStructureEquals(*column)) + return false; + } + + return true; +} + void ColumnObject::takeDynamicStructureFromSourceColumns(const DB::Columns & source_columns) { if (!empty()) diff --git a/src/Columns/ColumnObject.h b/src/Columns/ColumnObject.h index c7f282d9079..d5370625115 100644 --- a/src/Columns/ColumnObject.h +++ b/src/Columns/ColumnObject.h @@ -172,6 +172,7 @@ public: bool isFinalized() const override; bool hasDynamicStructure() const override { return true; } + bool dynamicStructureEquals(const IColumn & rhs) const override; void takeDynamicStructureFromSourceColumns(const Columns & source_columns) override; const PathToColumnMap & getTypedPaths() const { return typed_paths; } @@ -221,6 +222,7 @@ public: void setDynamicPaths(const std::vector & paths); void setMaxDynamicPaths(size_t max_dynamic_paths_); + void setGlobalMaxDynamicPaths(size_t global_max_dynamic_paths_); void setStatistics(const StatisticsPtr & statistics_) { statistics = statistics_; } void serializePathAndValueIntoSharedData(ColumnString * shared_data_paths, ColumnString * shared_data_values, std::string_view path, const IColumn & column, size_t n); diff --git a/src/Columns/ColumnTuple.cpp b/src/Columns/ColumnTuple.cpp index e741eb51c68..42acfdc85be 100644 --- a/src/Columns/ColumnTuple.cpp +++ b/src/Columns/ColumnTuple.cpp @@ -727,6 +727,26 @@ bool ColumnTuple::hasDynamicStructure() const return false; } +bool ColumnTuple::dynamicStructureEquals(const IColumn & rhs) const +{ + if (const auto * rhs_tuple = typeid_cast(&rhs)) + { + const size_t tuple_size = columns.size(); + if (tuple_size != rhs_tuple->columns.size()) + return false; + + for (size_t i = 0; i < tuple_size; ++i) + if (!columns[i]->dynamicStructureEquals(*rhs_tuple->columns[i])) + return false; + + return true; + } + else + { + return false; + } +} + void ColumnTuple::takeDynamicStructureFromSourceColumns(const Columns & source_columns) { std::vector nested_source_columns; diff --git a/src/Columns/ColumnTuple.h b/src/Columns/ColumnTuple.h index 6968294aef9..2539c27c441 100644 --- a/src/Columns/ColumnTuple.h +++ b/src/Columns/ColumnTuple.h @@ -138,6 +138,7 @@ public: ColumnPtr & getColumnPtr(size_t idx) { return columns[idx]; } bool hasDynamicStructure() const override; + bool dynamicStructureEquals(const IColumn & rhs) const override; void takeDynamicStructureFromSourceColumns(const Columns & source_columns) override; /// Empty tuple needs a public method to manage its size. diff --git a/src/Columns/ColumnVariant.cpp b/src/Columns/ColumnVariant.cpp index c6511695f5c..a18dffd8360 100644 --- a/src/Columns/ColumnVariant.cpp +++ b/src/Columns/ColumnVariant.cpp @@ -1376,6 +1376,23 @@ bool ColumnVariant::structureEquals(const IColumn & rhs) const return true; } +bool ColumnVariant::dynamicStructureEquals(const IColumn & rhs) const +{ + const auto * rhs_variant = typeid_cast(&rhs); + if (!rhs_variant) + return false; + + const size_t num_variants = variants.size(); + if (num_variants != rhs_variant->variants.size()) + return false; + + for (size_t i = 0; i < num_variants; ++i) + if (!variants[i]->dynamicStructureEquals(rhs_variant->getVariantByGlobalDiscriminator(globalDiscriminatorByLocal(i)))) + return false; + + return true; +} + ColumnPtr ColumnVariant::compress() const { ColumnPtr local_discriminators_compressed = local_discriminators->compress(); diff --git a/src/Columns/ColumnVariant.h b/src/Columns/ColumnVariant.h index 925eab74af8..2084de4fae7 100644 --- a/src/Columns/ColumnVariant.h +++ b/src/Columns/ColumnVariant.h @@ -327,6 +327,7 @@ public: void extend(const std::vector & old_to_new_global_discriminators, std::vector> && new_variants_and_discriminators); bool hasDynamicStructure() const override; + bool dynamicStructureEquals(const IColumn & rhs) const override; void takeDynamicStructureFromSourceColumns(const Columns & source_columns) override; private: diff --git a/src/Columns/IColumn.h b/src/Columns/IColumn.h index e4fe233ffdf..7131765f99c 100644 --- a/src/Columns/IColumn.h +++ b/src/Columns/IColumn.h @@ -590,6 +590,9 @@ public: /// Checks if column has dynamic subcolumns. virtual bool hasDynamicStructure() const { return false; } + + /// For columns with dynamic subcolumns checks if columns have equal dynamic structure. + [[nodiscard]] virtual bool dynamicStructureEquals(const IColumn & rhs) const { return structureEquals(rhs); } /// For columns with dynamic subcolumns this method takes dynamic structure from source columns /// and creates proper resulting dynamic structure in advance for merge of these source columns. virtual void takeDynamicStructureFromSourceColumns(const std::vector & /*source_columns*/) {} diff --git a/src/DataTypes/Serializations/SerializationDynamic.cpp b/src/DataTypes/Serializations/SerializationDynamic.cpp index 18a75918499..b00668fa8a4 100644 --- a/src/DataTypes/Serializations/SerializationDynamic.cpp +++ b/src/DataTypes/Serializations/SerializationDynamic.cpp @@ -26,8 +26,8 @@ namespace ErrorCodes struct SerializeBinaryBulkStateDynamic : public ISerialization::SerializeBinaryBulkState { - SerializationDynamic::DynamicStructureSerializationVersion structure_version; - size_t max_dynamic_types; + SerializationDynamic::DynamicSerializationVersion structure_version; + size_t num_dynamic_types; DataTypePtr variant_type; Names variant_names; SerializationPtr variant_serialization; @@ -81,14 +81,14 @@ void SerializationDynamic::enumerateStreams( settings.path.pop_back(); } -SerializationDynamic::DynamicStructureSerializationVersion::DynamicStructureSerializationVersion(UInt64 version) : value(static_cast(version)) +SerializationDynamic::DynamicSerializationVersion::DynamicSerializationVersion(UInt64 version) : value(static_cast(version)) { checkVersion(version); } -void SerializationDynamic::DynamicStructureSerializationVersion::checkVersion(UInt64 version) +void SerializationDynamic::DynamicSerializationVersion::checkVersion(UInt64 version) { - if (version != VariantTypeName) + if (version != V1 && version != V2) throw Exception(ErrorCodes::INCORRECT_DATA, "Invalid version for Dynamic structure serialization."); } @@ -108,22 +108,17 @@ void SerializationDynamic::serializeBinaryBulkStatePrefix( throw Exception(ErrorCodes::LOGICAL_ERROR, "Missing stream for Dynamic column structure during serialization of binary bulk state prefix"); /// Write structure serialization version. - UInt64 structure_version = DynamicStructureSerializationVersion::Value::VariantTypeName; + UInt64 structure_version = DynamicSerializationVersion::Value::V2; writeBinaryLittleEndian(structure_version, *stream); auto dynamic_state = std::make_shared(structure_version); - dynamic_state->max_dynamic_types = column_dynamic.getMaxDynamicTypes(); - /// Write max_dynamic_types parameter, because it can differ from the max_dynamic_types - /// that is specified in the Dynamic type (we could decrease it before merge). - writeVarUInt(dynamic_state->max_dynamic_types, *stream); - dynamic_state->variant_type = variant_info.variant_type; dynamic_state->variant_names = variant_info.variant_names; const auto & variant_column = column_dynamic.getVariantColumn(); - /// Write information about variants. - size_t num_variants = dynamic_state->variant_names.size() - 1; /// Don't write shared variant, Dynamic column should always have it. - writeVarUInt(num_variants, *stream); + /// Write information about dynamic types. + dynamic_state->num_dynamic_types = dynamic_state->variant_names.size() - 1; /// -1 for SharedVariant + writeVarUInt(dynamic_state->num_dynamic_types, *stream); if (settings.data_types_binary_encoding) { const auto & variants = assert_cast(*dynamic_state->variant_type).getVariants(); @@ -251,22 +246,25 @@ ISerialization::DeserializeBinaryBulkStatePtr SerializationDynamic::deserializeD UInt64 structure_version; readBinaryLittleEndian(structure_version, *structure_stream); auto structure_state = std::make_shared(structure_version); - /// Read max_dynamic_types parameter. - readVarUInt(structure_state->max_dynamic_types, *structure_stream); + if (structure_state->structure_version.value == DynamicSerializationVersion::Value::V1) + { + /// Skip max_dynamic_types parameter in V1 serialization version. + size_t max_dynamic_types; + readVarUInt(max_dynamic_types, *structure_stream); + } /// Read information about variants. DataTypes variants; - size_t num_variants; - readVarUInt(num_variants, *structure_stream); - variants.reserve(num_variants + 1); /// +1 for shared variant. + readVarUInt(structure_state->num_dynamic_types, *structure_stream); + variants.reserve(structure_state->num_dynamic_types + 1); /// +1 for shared variant. if (settings.data_types_binary_encoding) { - for (size_t i = 0; i != num_variants; ++i) + for (size_t i = 0; i != structure_state->num_dynamic_types; ++i) variants.push_back(decodeDataType(*structure_stream)); } else { String data_type_name; - for (size_t i = 0; i != num_variants; ++i) + for (size_t i = 0; i != structure_state->num_dynamic_types; ++i) { readStringBinary(data_type_name, *structure_stream); variants.push_back(DataTypeFactory::instance().get(data_type_name)); @@ -364,9 +362,6 @@ void SerializationDynamic::serializeBinaryBulkWithMultipleStreamsAndCountTotalSi if (!variant_info.variant_type->equals(*dynamic_state->variant_type)) throw Exception(ErrorCodes::LOGICAL_ERROR, "Mismatch of internal columns of Dynamic. Expected: {}, Got: {}", dynamic_state->variant_type->getName(), variant_info.variant_type->getName()); - if (column_dynamic.getMaxDynamicTypes() != dynamic_state->max_dynamic_types) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Mismatch of max_dynamic_types parameter of Dynamic. Expected: {}, Got: {}", dynamic_state->max_dynamic_types, column_dynamic.getMaxDynamicTypes()); - settings.path.push_back(Substream::DynamicData); assert_cast(*dynamic_state->variant_serialization) .serializeBinaryBulkWithMultipleStreamsAndUpdateVariantStatistics( @@ -424,7 +419,7 @@ void SerializationDynamic::deserializeBinaryBulkWithMultipleStreams( if (mutable_column->empty()) { - column_dynamic.setMaxDynamicPaths(structure_state->max_dynamic_types); + column_dynamic.setMaxDynamicPaths(structure_state->num_dynamic_types); column_dynamic.setVariantType(structure_state->variant_type); column_dynamic.setStatistics(structure_state->statistics); } diff --git a/src/DataTypes/Serializations/SerializationDynamic.h b/src/DataTypes/Serializations/SerializationDynamic.h index f34b5d0e770..ac98bbbc8b5 100644 --- a/src/DataTypes/Serializations/SerializationDynamic.h +++ b/src/DataTypes/Serializations/SerializationDynamic.h @@ -16,18 +16,28 @@ public: { } - struct DynamicStructureSerializationVersion + struct DynamicSerializationVersion { enum Value { - VariantTypeName = 1, + /// V1 serialization: + /// - DynamicStructure stream: + /// + /// + /// + /// (only in MergeTree serialization) + /// (only in MergeTree serialization) + /// - DynamicData stream: contains the data of nested Variant column. + V1 = 1, + /// V2 serialization: the same as V1 but without max_dynamic_types parameter in DynamicStructure stream. + V2 = 2, }; Value value; static void checkVersion(UInt64 version); - explicit DynamicStructureSerializationVersion(UInt64 version); + explicit DynamicSerializationVersion(UInt64 version); }; void enumerateStreams( @@ -113,9 +123,9 @@ private: struct DeserializeBinaryBulkStateDynamicStructure : public ISerialization::DeserializeBinaryBulkState { - DynamicStructureSerializationVersion structure_version; + DynamicSerializationVersion structure_version; DataTypePtr variant_type; - size_t max_dynamic_types; + size_t num_dynamic_types; ColumnDynamic::StatisticsPtr statistics; explicit DeserializeBinaryBulkStateDynamicStructure(UInt64 structure_version_) diff --git a/src/DataTypes/Serializations/SerializationObject.cpp b/src/DataTypes/Serializations/SerializationObject.cpp index 760f6ce750d..b3ac2c52d70 100644 --- a/src/DataTypes/Serializations/SerializationObject.cpp +++ b/src/DataTypes/Serializations/SerializationObject.cpp @@ -68,14 +68,13 @@ SerializationObject::ObjectSerializationVersion::ObjectSerializationVersion(UInt void SerializationObject::ObjectSerializationVersion::checkVersion(UInt64 version) { - if (version != BASIC) + if (version != V1 && version != V2) throw Exception(ErrorCodes::INCORRECT_DATA, "Invalid version for Object structure serialization."); } struct SerializeBinaryBulkStateObject: public ISerialization::SerializeBinaryBulkState { SerializationObject::ObjectSerializationVersion serialization_version; - size_t max_dynamic_paths; std::vector sorted_dynamic_paths; std::unordered_map typed_path_states; std::unordered_map dynamic_path_states; @@ -193,13 +192,10 @@ void SerializationObject::serializeBinaryBulkStatePrefix( throw Exception(ErrorCodes::LOGICAL_ERROR, "Missing stream for Object column structure during serialization of binary bulk state prefix"); /// Write serialization version. - UInt64 serialization_version = ObjectSerializationVersion::Value::BASIC; + UInt64 serialization_version = ObjectSerializationVersion::Value::V2; writeBinaryLittleEndian(serialization_version, *stream); auto object_state = std::make_shared(serialization_version); - object_state->max_dynamic_paths = column_object.getMaxDynamicPaths(); - /// Write max_dynamic_paths parameter. - writeVarUInt(object_state->max_dynamic_paths, *stream); /// Write all dynamic paths in sorted order. object_state->sorted_dynamic_paths.reserve(dynamic_paths.size()); for (const auto & [path, _] : dynamic_paths) @@ -353,8 +349,13 @@ ISerialization::DeserializeBinaryBulkStatePtr SerializationObject::deserializeOb UInt64 serialization_version; readBinaryLittleEndian(serialization_version, *structure_stream); auto structure_state = std::make_shared(serialization_version); - /// Read max_dynamic_paths parameter. - readVarUInt(structure_state->max_dynamic_paths, *structure_stream); + if (structure_state->structure_version.value == ObjectSerializationVersion::Value::V1) + { + /// Skip max_dynamic_paths parameter in V1 serialization version. + size_t max_dynamic_paths; + readVarUInt(max_dynamic_paths, *structure_stream); + } + /// Read the sorted list of dynamic paths. size_t dynamic_paths_size; readVarUInt(dynamic_paths_size, *structure_stream); @@ -411,9 +412,6 @@ void SerializationObject::serializeBinaryBulkWithMultipleStreams( const auto & shared_data = column_object.getSharedDataPtr(); auto * object_state = checkAndGetState(state); - if (column_object.getMaxDynamicPaths() != object_state->max_dynamic_paths) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Mismatch of max_dynamic_paths parameter of Object. Expected: {}, Got: {}", object_state->max_dynamic_paths, column_object.getMaxDynamicPaths()); - if (column_object.getDynamicPaths().size() != object_state->sorted_dynamic_paths.size()) throw Exception(ErrorCodes::LOGICAL_ERROR, "Mismatch of number of dynamic paths in Object. Expected: {}, Got: {}", object_state->sorted_dynamic_paths.size(), column_object.getDynamicPaths().size()); @@ -538,7 +536,7 @@ void SerializationObject::deserializeBinaryBulkWithMultipleStreams( /// If it's a new object column, set dynamic paths and statistics. if (column_object.empty()) { - column_object.setMaxDynamicPaths(structure_state->max_dynamic_paths); + column_object.setMaxDynamicPaths(structure_state->sorted_dynamic_paths.size()); column_object.setDynamicPaths(structure_state->sorted_dynamic_paths); column_object.setStatistics(structure_state->statistics); } diff --git a/src/DataTypes/Serializations/SerializationObject.h b/src/DataTypes/Serializations/SerializationObject.h index 62ff9849f45..ba66dd6470e 100644 --- a/src/DataTypes/Serializations/SerializationObject.h +++ b/src/DataTypes/Serializations/SerializationObject.h @@ -19,7 +19,20 @@ public: { enum Value { - BASIC = 0, + /// V1 serialization: + /// - ObjectStructure stream: + /// + /// + /// + /// (only in MergeTree serialization) + /// (only in MergeTree serialization) + /// - ObjectData stream: + /// - ObjectTypedPath stream for each column in typed paths + /// - ObjectDynamicPath stream for each column in dynamic paths + /// - ObjectSharedData stream shared data column. + V1 = 0, + /// V2 serialization: the same as V1 but without max_dynamic_paths parameter in ObjectStructure stream. + V2 = 2, }; Value value; @@ -82,7 +95,6 @@ private: struct DeserializeBinaryBulkStateObjectStructure : public ISerialization::DeserializeBinaryBulkState { ObjectSerializationVersion structure_version; - size_t max_dynamic_paths; std::vector sorted_dynamic_paths; std::unordered_set dynamic_paths; /// Paths statistics. Map (dynamic path) -> (number of non-null values in this path). diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp index ed13e581759..a7098e85ea0 100644 --- a/src/Functions/FunctionsConversion.cpp +++ b/src/Functions/FunctionsConversion.cpp @@ -83,6 +83,7 @@ namespace Setting extern const SettingsBool input_format_ipv4_default_on_conversion_error; extern const SettingsBool input_format_ipv6_default_on_conversion_error; extern const SettingsBool precise_float_parsing; + extern const SettingsBool cast_to_json_disable_dynamic_subcolumns; } namespace ErrorCodes @@ -4056,9 +4057,7 @@ private: { return [this](ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * nullable_source, size_t input_rows_count) { - auto res = ConvertImplGenericFromString::execute(arguments, result_type, nullable_source, input_rows_count, context)->assumeMutable(); - res->finalize(); - return res; + return ConvertImplGenericFromString::execute(arguments, result_type, nullable_source, input_rows_count, context)->assumeMutable(); }; } diff --git a/src/Storages/AlterCommands.cpp b/src/Storages/AlterCommands.cpp index 460d74e68bf..0d7d3295e0a 100644 --- a/src/Storages/AlterCommands.cpp +++ b/src/Storages/AlterCommands.cpp @@ -1466,13 +1466,13 @@ void AlterCommands::validate(const StoragePtr & table, ContextPtr context) const "The change of data type {} of column {} to {} is not allowed. It has known bugs", old_data_type->getName(), backQuote(column_name), command.data_type->getName()); - bool has_object_type = isObject(command.data_type); - command.data_type->forEachChild([&](const IDataType & type){ has_object_type |= isObject(type); }); - if (has_object_type) - throw Exception( - ErrorCodes::BAD_ARGUMENTS, - "The change of data type {} of column {} to {} is not supported.", - old_data_type->getName(), backQuote(column_name), command.data_type->getName()); +// bool has_object_type = isObject(command.data_type); +// command.data_type->forEachChild([&](const IDataType & type){ has_object_type |= isObject(type); }); +// if (has_object_type) +// throw Exception( +// ErrorCodes::BAD_ARGUMENTS, +// "The change of data type {} of column {} to {} is not supported.", +// old_data_type->getName(), backQuote(column_name), command.data_type->getName()); } if (command.isRemovingProperty()) diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp index a859172023f..96623307c8f 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp @@ -61,22 +61,6 @@ MergeTreeDataPartWriterCompact::MergeTreeDataPartWriterCompact( } } -void MergeTreeDataPartWriterCompact::initDynamicStreamsIfNeeded(const Block & block) -{ - if (is_dynamic_streams_initialized) - return; - - is_dynamic_streams_initialized = true; - for (const auto & column : columns_list) - { - if (column.type->hasDynamicSubcolumns()) - { - auto compression = getCodecDescOrDefault(column.name, default_codec); - addStreams(column, block.getByName(column.name).column, compression); - } - } -} - void MergeTreeDataPartWriterCompact::addStreams(const NameAndTypePair & name_and_type, const ColumnPtr & column, const ASTPtr & effective_codec_desc) { ISerialization::StreamCallback callback = [&](const auto & substream_path) @@ -175,20 +159,25 @@ void writeColumnSingleGranule( void MergeTreeDataPartWriterCompact::write(const Block & block, const IColumn::Permutation * permutation) { - /// On first block of data initialize streams for dynamic subcolumns. - initDynamicStreamsIfNeeded(block); + Block result_block = block; + + /// During serialization columns with dynamic subcolumns (like JSON/Dynamic) must have the same dynamic structure. + /// But it may happen that they don't (for example during ALTER MODIFY COLUMN from some type to JSON/Dynamic). + /// In this case we use dynamic structure of the column from the first written block and adjust columns from + /// the next blocks so they match this dynamic structure. + initOrAdjustDynamicStructureIfNeeded(result_block); /// Fill index granularity for this block /// if it's unknown (in case of insert data or horizontal merge, /// but not in case of vertical merge) if (compute_granularity) { - size_t index_granularity_for_block = computeIndexGranularity(block); + size_t index_granularity_for_block = computeIndexGranularity(result_block); assert(index_granularity_for_block >= 1); - fillIndexGranularity(index_granularity_for_block, block.rows()); + fillIndexGranularity(index_granularity_for_block, result_block.rows()); } - Block result_block = permuteBlockIfNeeded(block, permutation); + result_block = permuteBlockIfNeeded(result_block, permutation); if (!header) header = result_block.cloneEmpty(); diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h index b440a37222d..03da9c5f754 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h @@ -48,9 +48,7 @@ private: void addToChecksums(MergeTreeDataPartChecksums & checksums); - void addStreams(const NameAndTypePair & name_and_type, const ColumnPtr & column, const ASTPtr & effective_codec_desc); - - void initDynamicStreamsIfNeeded(const Block & block); + void addStreams(const NameAndTypePair & name_and_type, const ColumnPtr & column, const ASTPtr & effective_codec_desc) override; Block header; @@ -104,8 +102,6 @@ private: /// then finally to 'marks_file'. std::unique_ptr marks_compressor; std::unique_ptr marks_source_hashing; - - bool is_dynamic_streams_initialized = false; }; } diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp index 35914d8c50a..fbf6ac769a0 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp @@ -557,6 +557,45 @@ Names MergeTreeDataPartWriterOnDisk::getSkipIndicesColumns() const return Names(skip_indexes_column_names_set.begin(), skip_indexes_column_names_set.end()); } +void MergeTreeDataPartWriterOnDisk::initOrAdjustDynamicStructureIfNeeded(Block & block) +{ + if (!is_dynamic_streams_initialized) + { + for (const auto & column : columns_list) + { + if (column.type->hasDynamicSubcolumns()) + { + /// Create all streams for dynamic subcolumns using dynamic structure from block. + auto compression = getCodecDescOrDefault(column.name, default_codec); + addStreams(column, block.getByName(column.name).column, compression); + } + } + is_dynamic_streams_initialized = true; + block_sample = block.cloneEmpty(); + } + else + { + size_t size = block.columns(); + for (size_t i = 0; i != size; ++i) + { + auto & column = block.getByPosition(i); + const auto & sample_column = block_sample.getByPosition(i); + /// Check if the dynamic structure of this column is different from the sample column. + if (column.type->hasDynamicSubcolumns() && !column.column->dynamicStructureEquals(*sample_column.column)) + { + /// We need to change the dynamic structure of the column so it matches the sample column. + /// To do it, we create empty column of this type, take dynamic structure from sample column + /// and insert data into it. Resulting column will have required dynamic structure and the content + /// of the column in current block. + auto new_column = sample_column.type->createColumn(); + new_column->takeDynamicStructureFromSourceColumns({sample_column.column}); + new_column->insertRangeFrom(*column.column, 0, column.column->size()); + column.column = std::move(new_column); + } + } + } +} + template struct MergeTreeDataPartWriterOnDisk::Stream; template struct MergeTreeDataPartWriterOnDisk::Stream; diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h index 8d84442981e..69a089eda1b 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h @@ -153,6 +153,14 @@ protected: /// Get unique non ordered skip indices column. Names getSkipIndicesColumns() const; + virtual void addStreams(const NameAndTypePair & name_and_type, const ColumnPtr & column, const ASTPtr & effective_codec_desc) = 0; + + /// On first block create all required streams for columns with dynamic subcolumns and remember the block sample. + /// On each next block check if dynamic structure of the columns equals to the dynamic structure of the same + /// columns in the sample block. If for some column dynamic structure is different, adjust it so it matches + /// the structure from the sample. + void initOrAdjustDynamicStructureIfNeeded(Block & block); + const MergeTreeIndices skip_indices; const ColumnsStatistics stats; @@ -187,6 +195,10 @@ protected: size_t current_mark = 0; GinIndexStoreFactory::GinIndexStores gin_index_stores; + + bool is_dynamic_streams_initialized = false; + Block block_sample; + private: void initSkipIndices(); void initPrimaryIndex(); diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp index 04e07a0588a..ba9d82fd097 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp @@ -106,23 +106,6 @@ MergeTreeDataPartWriterWide::MergeTreeDataPartWriterWide( } } -void MergeTreeDataPartWriterWide::initDynamicStreamsIfNeeded(const DB::Block & block) -{ - if (is_dynamic_streams_initialized) - return; - - is_dynamic_streams_initialized = true; - block_sample = block.cloneEmpty(); - for (const auto & column : columns_list) - { - if (column.type->hasDynamicSubcolumns()) - { - auto compression = getCodecDescOrDefault(column.name, default_codec); - addStreams(column, block_sample.getByName(column.name).column, compression); - } - } -} - void MergeTreeDataPartWriterWide::addStreams( const NameAndTypePair & name_and_type, const ColumnPtr & column, @@ -260,15 +243,20 @@ void MergeTreeDataPartWriterWide::shiftCurrentMark(const Granules & granules_wri void MergeTreeDataPartWriterWide::write(const Block & block, const IColumn::Permutation * permutation) { - /// On first block of data initialize streams for dynamic subcolumns. - initDynamicStreamsIfNeeded(block); + Block block_to_write = block; + + /// During serialization columns with dynamic subcolumns (like JSON/Dynamic) must have the same dynamic structure. + /// But it may happen that they don't (for example during ALTER MODIFY COLUMN from some type to JSON/Dynamic). + /// In this case we use dynamic structure of the column from the first written block and adjust columns from + /// the next blocks so they match this dynamic structure. + initOrAdjustDynamicStructureIfNeeded(block_to_write); /// Fill index granularity for this block /// if it's unknown (in case of insert data or horizontal merge, /// but not in case of vertical part of vertical merge) if (compute_granularity) { - size_t index_granularity_for_block = computeIndexGranularity(block); + size_t index_granularity_for_block = computeIndexGranularity(block_to_write); if (rows_written_in_last_mark > 0) { size_t rows_left_in_last_mark = index_granularity.getMarkRows(getCurrentMark()) - rows_written_in_last_mark; @@ -286,11 +274,9 @@ void MergeTreeDataPartWriterWide::write(const Block & block, const IColumn::Perm } } - fillIndexGranularity(index_granularity_for_block, block.rows()); + fillIndexGranularity(index_granularity_for_block, block_to_write.rows()); } - Block block_to_write = block; - auto granules_to_write = getGranulesToWrite(index_granularity, block_to_write.rows(), getCurrentMark(), rows_written_in_last_mark); auto offset_columns = written_offset_columns ? *written_offset_columns : WrittenOffsetColumns{}; diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h index ab86ed27c7e..78dfc93c4d2 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h @@ -91,9 +91,7 @@ private: void addStreams( const NameAndTypePair & name_and_type, const ColumnPtr & column, - const ASTPtr & effective_codec_desc); - - void initDynamicStreamsIfNeeded(const Block & block); + const ASTPtr & effective_codec_desc) override; /// Method for self check (used in debug-build only). Checks that written /// data and corresponding marks are consistent. Otherwise throws logical @@ -139,10 +137,6 @@ private: /// How many rows we have already written in the current mark. /// More than zero when incoming blocks are smaller then their granularity. size_t rows_written_in_last_mark = 0; - - Block block_sample; - - bool is_dynamic_streams_initialized = false; }; } diff --git a/tests/queries/0_stateless/03246_alter_from_string_to_json.reference b/tests/queries/0_stateless/03246_alter_from_string_to_json.reference new file mode 100644 index 00000000000..a2d3a799fff --- /dev/null +++ b/tests/queries/0_stateless/03246_alter_from_string_to_json.reference @@ -0,0 +1,134 @@ +All paths: +['key0','key1','key2','key3','key4','key5'] +Shared data paths: +key2 +key3 +key4 +key5 +{"key0":"value0"} +{"key1":"value1"} +{"key0":"value2"} +{"key1":"value3"} +{"key0":"value4"} +{"key1":"value5"} +{"key0":"value6"} +{"key1":"value7"} +{"key0":"value8"} +{"key1":"value9"} +{"key2":"value300000"} +{"key3":"value300001"} +{"key2":"value300002"} +{"key3":"value300003"} +{"key2":"value300004"} +{"key3":"value300005"} +{"key2":"value300006"} +{"key3":"value300007"} +{"key2":"value300008"} +{"key3":"value300009"} +{"key4":"value600000"} +{"key5":"value600001"} +{"key4":"value600002"} +{"key5":"value600003"} +{"key4":"value600004"} +{"key5":"value600005"} +{"key4":"value600006"} +{"key5":"value600007"} +{"key4":"value600008"} +{"key5":"value600009"} +value0 \N \N \N \N \N +\N value1 \N \N \N \N +value2 \N \N \N \N \N +\N value3 \N \N \N \N +value4 \N \N \N \N \N +\N value5 \N \N \N \N +value6 \N \N \N \N \N +\N value7 \N \N \N \N +value8 \N \N \N \N \N +\N value9 \N \N \N \N +\N \N value300000 \N \N \N +\N \N \N value300001 \N \N +\N \N value300002 \N \N \N +\N \N \N value300003 \N \N +\N \N value300004 \N \N \N +\N \N \N value300005 \N \N +\N \N value300006 \N \N \N +\N \N \N value300007 \N \N +\N \N value300008 \N \N \N +\N \N \N value300009 \N \N +\N \N \N \N value600000 \N +\N \N \N \N \N value600001 +\N \N \N \N value600002 \N +\N \N \N \N \N value600003 +\N \N \N \N value600004 \N +\N \N \N \N \N value600005 +\N \N \N \N value600006 \N +\N \N \N \N \N value600007 +\N \N \N \N value600008 \N +\N \N \N \N \N value600009 +All paths: +['key0','key1','key2','key3','key4','key5'] +Shared data paths: +key2 +key3 +key4 +key5 +{"key0":"value0"} +{"key1":"value1"} +{"key0":"value2"} +{"key1":"value3"} +{"key0":"value4"} +{"key1":"value5"} +{"key0":"value6"} +{"key1":"value7"} +{"key0":"value8"} +{"key1":"value9"} +{"key2":"value300000"} +{"key3":"value300001"} +{"key2":"value300002"} +{"key3":"value300003"} +{"key2":"value300004"} +{"key3":"value300005"} +{"key2":"value300006"} +{"key3":"value300007"} +{"key2":"value300008"} +{"key3":"value300009"} +{"key4":"value600000"} +{"key5":"value600001"} +{"key4":"value600002"} +{"key5":"value600003"} +{"key4":"value600004"} +{"key5":"value600005"} +{"key4":"value600006"} +{"key5":"value600007"} +{"key4":"value600008"} +{"key5":"value600009"} +value0 \N \N \N \N \N +\N value1 \N \N \N \N +value2 \N \N \N \N \N +\N value3 \N \N \N \N +value4 \N \N \N \N \N +\N value5 \N \N \N \N +value6 \N \N \N \N \N +\N value7 \N \N \N \N +value8 \N \N \N \N \N +\N value9 \N \N \N \N +\N \N value300000 \N \N \N +\N \N \N value300001 \N \N +\N \N value300002 \N \N \N +\N \N \N value300003 \N \N +\N \N value300004 \N \N \N +\N \N \N value300005 \N \N +\N \N value300006 \N \N \N +\N \N \N value300007 \N \N +\N \N value300008 \N \N \N +\N \N \N value300009 \N \N +\N \N \N \N value600000 \N +\N \N \N \N \N value600001 +\N \N \N \N value600002 \N +\N \N \N \N \N value600003 +\N \N \N \N value600004 \N +\N \N \N \N \N value600005 +\N \N \N \N value600006 \N +\N \N \N \N \N value600007 +\N \N \N \N value600008 \N +\N \N \N \N \N value600009 diff --git a/tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 b/tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 new file mode 100644 index 00000000000..a13867b145d --- /dev/null +++ b/tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 @@ -0,0 +1,32 @@ +set allow_experimental_json_type = 1; + +drop table if exists test; + +{% for create_command in ['create table test (x UInt64, json String) engine=MergeTree order by x settings min_rows_for_wide_part=100000000, min_bytes_for_wide_part=1000000000;', + 'create table test (x UInt64, json String) engine=MergeTree order by x settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1;'] -%} + +{{ create_command }} + +insert into test select number, toJSONString(map('key' || multiIf(number < 300000, number % 2, number < 600000, number % 2 + 2, number % 2 + 4), 'value' || number)) from numbers(1000000); + +alter table test modify column json JSON settings mutations_sync=1; + +select 'All paths:'; +select distinctJSONPaths(json) from test; +select 'Shared data paths:'; +select distinct (arrayJoin(JSONSharedDataPaths(json))) as path from test order by path; +select json from test order by x limit 10; +select json from test order by x limit 10 offset 300000; +select json from test order by x limit 10 offset 600000; +select json.key0, json.key1, json.key2, json.key3, json.key4, json.key5 from test order by x limit 10; +select json.key0, json.key1, json.key2, json.key3, json.key4, json.key5 from test order by x limit 10 offset 300000; +select json.key0, json.key1, json.key2, json.key3, json.key4, json.key5 from test order by x limit 10 offset 600000; + +select json from test format Null; +select json from test order by x format Null; +select json.key0, json.key1, json.key2, json.key3, json.key4, json.key5 from test format Null; +select json.key0, json.key1, json.key2, json.key3, json.key4, json.key5 from test order by x format Null; + +drop table test; + +{% endfor -%} diff --git a/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.reference b/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.reference new file mode 100644 index 00000000000..ca2fb7e8ff9 --- /dev/null +++ b/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.reference @@ -0,0 +1,12 @@ +5000 +leonardomso/33-js-concepts 3 +ytdl-org/youtube-dl 3 +Bogdanp/neko 2 +bminossi/AllVideoPocsFromHackerOne 2 +disclose/diodata 2 +Commit 182 +chipeo345 119 +phanwi346 114 +Nicholas Piggin 95 +direwolf-github 49 +2 diff --git a/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh b/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh new file mode 100755 index 00000000000..931d106120c --- /dev/null +++ b/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh @@ -0,0 +1,30 @@ +#!/usr/bin/env bash +# Tags: no-fasttest, no-s3-storage, long +# ^ no-s3-storage: too memory hungry + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS ghdata" +${CLICKHOUSE_CLIENT} -q "CREATE TABLE ghdata (data String) ENGINE = MergeTree ORDER BY tuple() SETTINGS index_granularity = 8192, index_granularity_bytes = '10Mi'" + +cat $CUR_DIR/data_json/ghdata_sample.json | ${CLICKHOUSE_CLIENT} \ + --max_memory_usage 10G --query "INSERT INTO ghdata FORMAT JSONAsString" + +${CLICKHOUSE_CLIENT} -q "ALTER TABLE ghdata MODIFY column data JSON SETTINGS mutations_sync=1" --allow_experimental_json_type 1 + +${CLICKHOUSE_CLIENT} -q "SELECT count() FROM ghdata WHERE NOT ignore(*)" + +${CLICKHOUSE_CLIENT} -q \ +"SELECT data.repo.name, count() AS stars FROM ghdata \ + WHERE data.type = 'WatchEvent' GROUP BY data.repo.name ORDER BY stars DESC, data.repo.name LIMIT 5" + +${CLICKHOUSE_CLIENT} --enable_analyzer=1 -q \ +"SELECT data.payload.commits[].author.name AS name, count() AS c FROM ghdata \ + ARRAY JOIN data.payload.commits[].author.name \ + GROUP BY name ORDER BY c DESC, name LIMIT 5" + +${CLICKHOUSE_CLIENT} -q "SELECT max(data.payload.pull_request.assignees[].size0) FROM ghdata" + +${CLICKHOUSE_CLIENT} -q "DROP TABLE IF EXISTS ghdata" diff --git a/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.reference b/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql b/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql new file mode 100644 index 00000000000..87e10df9cc8 --- /dev/null +++ b/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql @@ -0,0 +1,17 @@ +set allow_experimental_json_type=1; +set max_insert_block_size=10000; +set max_block_size=10000; + +drop table if exists test; +drop named collection if exists json_alter_fuzzer; + +create table test (json String) engine=MergeTree order by tuple(); +create named collection json_alter_fuzzer AS json_str='{}'; +insert into test select * from fuzzJSON(json_alter_fuzzer, reuse_output=true, max_output_length=128) limit 200000; +alter table test modify column json JSON settings mutations_sync=1; +select json from test format Null; +optimize table test final; +select json from test format Null; +drop named collection json_alter_fuzzer; +drop table test; + From a9fc07d9af728f56b7b43c53403e278ae69e8096 Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 7 Oct 2024 07:06:10 +0000 Subject: [PATCH 151/680] Remove unneded changes --- src/Storages/AlterCommands.cpp | 8 -------- 1 file changed, 8 deletions(-) diff --git a/src/Storages/AlterCommands.cpp b/src/Storages/AlterCommands.cpp index 0d7d3295e0a..9972b34ecc4 100644 --- a/src/Storages/AlterCommands.cpp +++ b/src/Storages/AlterCommands.cpp @@ -1465,14 +1465,6 @@ void AlterCommands::validate(const StoragePtr & table, ContextPtr context) const ErrorCodes::BAD_ARGUMENTS, "The change of data type {} of column {} to {} is not allowed. It has known bugs", old_data_type->getName(), backQuote(column_name), command.data_type->getName()); - -// bool has_object_type = isObject(command.data_type); -// command.data_type->forEachChild([&](const IDataType & type){ has_object_type |= isObject(type); }); -// if (has_object_type) -// throw Exception( -// ErrorCodes::BAD_ARGUMENTS, -// "The change of data type {} of column {} to {} is not supported.", -// old_data_type->getName(), backQuote(column_name), command.data_type->getName()); } if (command.isRemovingProperty()) From a10c2674fe15c977a51c1ae7054f9f8e9bc4f7a3 Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 7 Oct 2024 07:20:10 +0000 Subject: [PATCH 152/680] Add example in docs --- docs/en/sql-reference/data-types/newjson.md | 22 +++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/docs/en/sql-reference/data-types/newjson.md b/docs/en/sql-reference/data-types/newjson.md index 68952590eb9..f799072a02f 100644 --- a/docs/en/sql-reference/data-types/newjson.md +++ b/docs/en/sql-reference/data-types/newjson.md @@ -630,6 +630,28 @@ SELECT arrayJoin(distinctJSONPathsAndTypes(json)) FROM s3('s3://clickhouse-publi └─arrayJoin(distinctJSONPathsAndTypes(json))──────────────────┘ ``` +## ALTER MODIFY COLUMN to JSON type + +It's possible to alter an existing table and change the type of the column to the new `JSON` type. Right now only alter from `String` type is supported. + +**Example** + +```sql +CREATE TABLE test (json String) ENGINE=MergeTree ORDeR BY tuple(); +INSERT INTO test VALUES ('{"a" : 42}'), ('{"a" : 43, "b" : "Hello"}'), ('{"a" : 44, "b" : [1, 2, 3]}')), ('{"c" : "2020-01-01"}'); +ALTER TABLE test MODIFY COLUMN json JSON; +SELECT json, json.a, json.b, json.c FROM test; +``` + +```text + ┌─json─────────────────────────┬─json.a─┬─json.b──┬─json.c─────┐ +1. │ {"a":"42"} │ 42 │ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │ +2. │ {"a":"43","b":"Hello"} │ 43 │ Hello │ ᴺᵁᴸᴸ │ +3. │ {"a":"44","b":["1","2","3"]} │ 44 │ [1,2,3] │ ᴺᵁᴸᴸ │ +4. │ {"c":"2020-01-01"} │ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │ 2020-01-01 │ + └──────────────────────────────┴────────┴─────────┴────────────┘ +``` + ## Tips for better usage of the JSON type Before creating `JSON` column and loading data into it, consider the following tips: From a803c56ae2943bbb46a87572448d6434d1ef4337 Mon Sep 17 00:00:00 2001 From: vdimir Date: Mon, 7 Oct 2024 14:06:08 +0000 Subject: [PATCH 153/680] fix JoinStep permute columns --- src/Processors/QueryPlan/JoinStep.cpp | 106 ++++------- .../Transforms/ColumnPermuteTransform.cpp | 16 +- .../Transforms/ColumnPermuteTransform.h | 2 + .../01763_filter_push_down_bugs.reference | 2 +- ...emove_redundant_sorting_analyzer.reference | 4 +- ...move_redundant_distinct_analyzer.reference | 18 +- .../02514_analyzer_drop_join_on.reference | 44 ++--- ...oin_with_totals_and_subquery_bug.reference | 2 +- .../02835_join_step_explain.reference | 28 +-- ...filter_push_down_equivalent_sets.reference | 166 +++++++++--------- 10 files changed, 175 insertions(+), 213 deletions(-) diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index 3edc64ef967..dcedc57713d 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -19,7 +19,7 @@ namespace ErrorCodes namespace { -std::vector> describeJoinActions(const JoinPtr & join) +static std::vector> describeJoinActions(const JoinPtr & join) { std::vector> description; const auto & table_join = join->getTableJoin(); @@ -37,52 +37,37 @@ std::vector> describeJoinActions(const JoinPtr & join) return description; } -size_t getPrefixLength(const NameSet & prefix, const Names & names) +std::vector getPermutationForBlock( + const Block & block, + const Block & lhs_block, + const Block & rhs_block, + const NameSet & name_filter) { - size_t i = 0; - for (; i < names.size(); ++i) - { - if (!prefix.contains(names[i])) - break; - } - return i; -} + std::vector permutation; + permutation.reserve(block.columns()); + Block::NameMap name_map = block.getNamesToIndexesMap(); -std::vector getPermutationToRotate(size_t prefix_size, size_t total_size) -{ - std::vector permutation(total_size); - size_t i = prefix_size % total_size; - for (auto & elem : permutation) + bool is_trivial = true; + for (const auto & other_block : {lhs_block, rhs_block}) { - elem = i; - i = (i + 1) % total_size; + for (const auto & col : other_block) + { + if (!name_filter.contains(col.name)) + continue; + if (auto it = name_map.find(col.name); it != name_map.end()) + { + is_trivial = is_trivial && it->second == permutation.size(); + permutation.push_back(it->second); + } + } } + + if (is_trivial && permutation.size() == block.columns()) + return {}; + return permutation; } -Block rotateBlock(const Block & block, size_t prefix_size) -{ - auto columns = block.getColumnsWithTypeAndName(); - std::rotate(columns.begin(), columns.begin() + prefix_size, columns.end()); - auto res = Block(std::move(columns)); - return res; -} - -NameSet getNameSetFromBlock(const Block & block) -{ - NameSet names; - for (const auto & column : block) - names.insert(column.name); - return names; -} - -Block rotateBlock(const Block & block, const Block & prefix_block) -{ - NameSet prefix_names_set = getNameSetFromBlock(prefix_block); - size_t prefix_size = getPrefixLength(prefix_names_set, block.getNames()); - return rotateBlock(block, prefix_size); -} - } JoinStep::JoinStep( @@ -109,7 +94,8 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines if (pipelines.size() != 2) throw Exception(ErrorCodes::LOGICAL_ERROR, "JoinStep expect two input steps"); - NameSet rhs_names = getNameSetFromBlock(pipelines[1]->getHeader()); + Block lhs_header = pipelines[0]->getHeader(); + Block rhs_header = pipelines[1]->getHeader(); if (swap_streams) std::swap(pipelines[0], pipelines[1]); @@ -135,29 +121,15 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines if (!use_new_analyzer) return pipeline; - const auto & result_names = pipeline->getHeader().getNames(); - size_t prefix_size = getPrefixLength(rhs_names, result_names); - if (!columns_to_remove.empty() || (0 < prefix_size && prefix_size < result_names.size())) + auto column_permutation = getPermutationForBlock(pipeline->getHeader(), lhs_header, rhs_header, required_output); + if (!column_permutation.empty()) { - auto column_permutation = getPermutationToRotate(prefix_size, result_names.size()); - size_t n = 0; - auto it = columns_to_remove.begin(); - for (size_t i = 0; i < column_permutation.size(); ++i) - { - if (it != columns_to_remove.end() && *it == i) - ++it; - else - column_permutation[n++] = column_permutation[i]; - } - column_permutation.resize(n); - pipeline->addSimpleTransform([&column_permutation](const Block & header) { return std::make_shared(header, column_permutation); }); } - return pipeline; } @@ -177,12 +149,16 @@ void JoinStep::describeActions(FormatSettings & settings) const for (const auto & [name, value] : describeJoinActions(join)) settings.out << prefix << name << ": " << value << '\n'; + if (swap_streams) + settings.out << prefix << "Swapped: true\n"; } void JoinStep::describeActions(JSONBuilder::JSONMap & map) const { for (const auto & [name, value] : describeJoinActions(join)) map.add(name, value); + if (swap_streams) + map.add("Swapped", true); } void JoinStep::setJoin(JoinPtr join_, bool swap_streams_) @@ -210,20 +186,10 @@ void JoinStep::updateOutputStream() return; } - if (swap_streams) - result_header = rotateBlock(result_header, input_streams[1].header); + auto column_permutation = getPermutationForBlock(result_header, input_streams[0].header, input_streams[1].header, required_output); + if (!column_permutation.empty()) + result_header = ColumnPermuteTransform::permute(std::move(result_header), column_permutation); - columns_to_remove.clear(); - for (size_t i = 0; i < result_header.columns(); ++i) - { - if (!required_output.contains(result_header.getByPosition(i).name)) - columns_to_remove.insert(i); - } - /// Do not remove all columns, keep at least one - if (!columns_to_remove.empty() && columns_to_remove.size() == result_header.columns()) - columns_to_remove.erase(columns_to_remove.begin()); - - result_header.erase(columns_to_remove); output_stream = DataStream { .header = result_header }; } diff --git a/src/Processors/Transforms/ColumnPermuteTransform.cpp b/src/Processors/Transforms/ColumnPermuteTransform.cpp index eb2a691d6d1..67c7996cbe0 100644 --- a/src/Processors/Transforms/ColumnPermuteTransform.cpp +++ b/src/Processors/Transforms/ColumnPermuteTransform.cpp @@ -16,13 +16,6 @@ void applyPermutation(std::vector & data, const std::vector & permuta data = std::move(res); } -Block permuteBlock(const Block & block, const std::vector & permutation) -{ - auto columns = block.getColumnsWithTypeAndName(); - applyPermutation(columns, permutation); - return Block(columns); -} - void permuteChunk(Chunk & chunk, const std::vector & permutation) { size_t num_rows = chunk.getNumRows(); @@ -33,8 +26,15 @@ void permuteChunk(Chunk & chunk, const std::vector & permutation) } +Block ColumnPermuteTransform::permute(const Block & block, const std::vector & permutation) +{ + auto columns = block.getColumnsWithTypeAndName(); + applyPermutation(columns, permutation); + return Block(columns); +} + ColumnPermuteTransform::ColumnPermuteTransform(const Block & header_, const std::vector & permutation_) - : ISimpleTransform(header_, permuteBlock(header_, permutation_), false) + : ISimpleTransform(header_, permute(header_, permutation_), false) , permutation(permutation_) { } diff --git a/src/Processors/Transforms/ColumnPermuteTransform.h b/src/Processors/Transforms/ColumnPermuteTransform.h index f4d68850193..25f3a8d0825 100644 --- a/src/Processors/Transforms/ColumnPermuteTransform.h +++ b/src/Processors/Transforms/ColumnPermuteTransform.h @@ -19,6 +19,8 @@ public: void transform(Chunk & chunk) override; + static Block permute(const Block & block, const std::vector & permutation); + private: Names column_names; std::vector permutation; diff --git a/tests/queries/0_stateless/01763_filter_push_down_bugs.reference b/tests/queries/0_stateless/01763_filter_push_down_bugs.reference index 19018a610b7..229ac6eae09 100644 --- a/tests/queries/0_stateless/01763_filter_push_down_bugs.reference +++ b/tests/queries/0_stateless/01763_filter_push_down_bugs.reference @@ -26,7 +26,7 @@ Expression ((Projection + Before ORDER BY)) Parts: 1/1 Granules: 1/1 Expression ((Project names + Projection)) - Filter ((WHERE + DROP unused columns after JOIN)) + Filter (WHERE) Join (JOIN FillRightFirst) Expression ReadFromMergeTree (default.t1) diff --git a/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference b/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference index 3c68d14fdf2..c9bf36f88ea 100644 --- a/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference +++ b/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference @@ -117,7 +117,7 @@ ORDER BY t1.number, t2.number -- explain Expression (Project names) Sorting (Sorting for ORDER BY) - Expression ((Before ORDER BY + (Projection + DROP unused columns after JOIN))) + Expression ((Before ORDER BY + Projection)) Join (JOIN FillRightFirst) Expression ((Change column names to column identifiers + (Project names + (Before ORDER BY + (Projection + (Change column names to column identifiers + (Project names + (Before ORDER BY + (Projection + Change column names to column identifiers))))))))) ReadFromSystemNumbers @@ -161,7 +161,7 @@ ORDER BY t1.number, t2.number -- explain Expression (Project names) Sorting (Sorting for ORDER BY) - Expression ((Before ORDER BY + (Projection + DROP unused columns after JOIN))) + Expression ((Before ORDER BY + Projection)) Join (JOIN FillRightFirst) Expression ((Change column names to column identifiers + (Project names + (Before ORDER BY + (Projection + (Change column names to column identifiers + (Project names + (Before ORDER BY + (Projection + Change column names to column identifiers))))))))) ReadFromSystemNumbers diff --git a/tests/queries/0_stateless/02500_remove_redundant_distinct_analyzer.reference b/tests/queries/0_stateless/02500_remove_redundant_distinct_analyzer.reference index 867ae394c1f..baa2be9dfdb 100644 --- a/tests/queries/0_stateless/02500_remove_redundant_distinct_analyzer.reference +++ b/tests/queries/0_stateless/02500_remove_redundant_distinct_analyzer.reference @@ -79,7 +79,7 @@ Expression (Project names) Sorting (Sorting for ORDER BY) Expression (Before ORDER BY) Distinct (Preliminary DISTINCT) - Expression ((Projection + DROP unused columns after JOIN)) + Expression (Projection) Join (JOIN FillRightFirst) Expression ((Change column names to column identifiers + Project names)) Distinct (DISTINCT) @@ -244,7 +244,7 @@ Expression ((Project names + (Projection + (Change column names to column identi Sorting (Sorting for ORDER BY) Expression ((Before ORDER BY + Projection)) Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -280,7 +280,7 @@ Expression (Project names) Sorting (Sorting for ORDER BY) Expression ((Before ORDER BY + Projection)) Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -315,7 +315,7 @@ Expression (Project names) Expression ((Before ORDER BY + Projection)) Rollup Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -348,7 +348,7 @@ Expression ((Project names + (Projection + (Change column names to column identi Expression ((Before ORDER BY + Projection)) Rollup Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -386,7 +386,7 @@ Expression (Project names) Expression ((Before ORDER BY + Projection)) Cube Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -419,7 +419,7 @@ Expression ((Project names + (Projection + (Change column names to column identi Expression ((Before ORDER BY + Projection)) Cube Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -457,7 +457,7 @@ Expression (Project names) Expression ((Before ORDER BY + Projection)) TotalsHaving Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -491,7 +491,7 @@ Expression ((Project names + (Projection + (Change column names to column identi Expression ((Before ORDER BY + Projection)) TotalsHaving Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers diff --git a/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference b/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference index d407a4c7985..bbfdf1ad5f4 100644 --- a/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference +++ b/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference @@ -8,17 +8,17 @@ Header: count() UInt64 Aggregating Header: __table1.a2 String count() UInt64 - Expression ((Before GROUP BY + DROP unused columns after JOIN)) + Expression (Before GROUP BY) Header: __table1.a2 String Join (JOIN FillRightFirst) Header: __table1.a2 String - Expression ((JOIN actions + DROP unused columns after JOIN)) + Expression (JOIN actions) Header: __table1.a2 String __table3.c1 UInt64 Join (JOIN FillRightFirst) Header: __table1.a2 String __table3.c1 UInt64 - Expression ((JOIN actions + DROP unused columns after JOIN)) + Expression (JOIN actions) Header: __table1.a2 String __table2.b1 UInt64 Join (JOIN FillRightFirst) @@ -45,38 +45,32 @@ Header: count() UInt64 EXPLAIN PLAN header = 1 SELECT a.a2, d.d2 FROM a JOIN b USING (k) JOIN c USING (k) JOIN d USING (k) ; -Expression ((Project names + (Projection + DROP unused columns after JOIN))) +Expression ((Project names + Projection)) Header: a2 String d2 String Join (JOIN FillRightFirst) Header: __table1.a2 String __table4.d2 String - Expression (DROP unused columns after JOIN) + Join (JOIN FillRightFirst) Header: __table1.a2 String __table1.k UInt64 Join (JOIN FillRightFirst) Header: __table1.a2 String __table1.k UInt64 - Expression (DROP unused columns after JOIN) + Expression (Change column names to column identifiers) Header: __table1.a2 String __table1.k UInt64 - Join (JOIN FillRightFirst) - Header: __table1.a2 String - __table1.k UInt64 - Expression (Change column names to column identifiers) - Header: __table1.a2 String - __table1.k UInt64 - ReadFromMemoryStorage - Header: a2 String - k UInt64 - Expression (Change column names to column identifiers) - Header: __table2.k UInt64 - ReadFromMemoryStorage - Header: k UInt64 + ReadFromMemoryStorage + Header: a2 String + k UInt64 Expression (Change column names to column identifiers) - Header: __table3.k UInt64 + Header: __table2.k UInt64 ReadFromMemoryStorage Header: k UInt64 + Expression (Change column names to column identifiers) + Header: __table3.k UInt64 + ReadFromMemoryStorage + Header: k UInt64 Expression (Change column names to column identifiers) Header: __table4.d2 String __table4.k UInt64 @@ -105,21 +99,21 @@ Header: bx String Expression Header: __table1.a2 String __table2.bx String - __table4.c2 String __table4.c1 UInt64 + __table4.c2 String Join (JOIN FillRightFirst) Header: __table1.a2 String __table2.bx String - __table4.c2 String __table4.c1 UInt64 - Expression ((JOIN actions + DROP unused columns after JOIN)) + __table4.c2 String + Expression (JOIN actions) Header: __table1.a2 String - __table2.bx String __table2.b1 UInt64 + __table2.bx String Join (JOIN FillRightFirst) Header: __table1.a2 String - __table2.bx String __table2.b1 UInt64 + __table2.bx String Expression ((JOIN actions + Change column names to column identifiers)) Header: __table1.a1 UInt64 __table1.a2 String diff --git a/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.reference b/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.reference index 86e7e2a6a49..116c78a15e4 100644 --- a/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.reference +++ b/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.reference @@ -5,7 +5,7 @@ 1 1 -1 +0 \N 100000000000000000000 diff --git a/tests/queries/0_stateless/02835_join_step_explain.reference b/tests/queries/0_stateless/02835_join_step_explain.reference index 2f641d4aa44..bdbc019d4f8 100644 --- a/tests/queries/0_stateless/02835_join_step_explain.reference +++ b/tests/queries/0_stateless/02835_join_step_explain.reference @@ -1,22 +1,22 @@ -Expression ((Project names + (Projection + DROP unused columns after JOIN))) +Expression ((Project names + Projection)) Header: id UInt64 value_1 String rhs.id UInt64 rhs.value_1 String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value_1 String : 1 - INPUT : 2 -> __table2.value_1 String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value_1 String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value_1 :: 1 -> value_1 String : 0 - ALIAS __table2.value_1 :: 2 -> rhs.value_1 String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 0 2 1 + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value_1 :: 3 -> rhs.value_1 String : 2 +Positions: 4 0 1 2 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value_1 String - __table2.value_1 String __table2.id UInt64 + __table2.value_1 String Type: INNER Strictness: ALL Algorithm: HashJoin @@ -50,25 +50,25 @@ Positions: 4 0 2 1 Parts: 1 Granules: 1 -- -Expression ((Project names + (Projection + DROP unused columns after JOIN))) +Expression ((Project names + Projection)) Header: id UInt64 value_1 String rhs.id UInt64 rhs.value_1 String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value_1 String : 1 - INPUT : 2 -> __table2.value_1 String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value_1 String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value_1 :: 1 -> value_1 String : 0 - ALIAS __table2.value_1 :: 2 -> rhs.value_1 String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 0 2 1 + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value_1 :: 3 -> rhs.value_1 String : 2 +Positions: 4 0 1 2 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value_1 String - __table2.value_1 String __table2.id UInt64 + __table2.value_1 String Type: INNER Strictness: ASOF Algorithm: HashJoin diff --git a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference index c98a98b236c..b7718d926c6 100644 --- a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference +++ b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference @@ -12,18 +12,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 2 0 1 + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 1 0 2 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: INNER Strictness: ALL Algorithm: HashJoin @@ -81,18 +81,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 2 0 1 + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 1 0 2 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: INNER Strictness: ALL Algorithm: HashJoin @@ -150,18 +150,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 2 0 1 + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 1 0 2 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: INNER Strictness: ALL Algorithm: HashJoin @@ -222,18 +222,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 2 0 1 + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 1 0 2 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: LEFT Strictness: ALL Algorithm: HashJoin @@ -291,31 +291,31 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 2 0 1 - Filter ((WHERE + DROP unused columns after JOIN)) + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 1 0 2 + Filter (WHERE) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Filter column: equals(__table2.id, 5_UInt8) (removed) Actions: INPUT :: 0 -> __table1.id UInt64 : 0 INPUT :: 1 -> __table1.value String : 1 - INPUT :: 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT :: 3 -> __table2.value String : 3 COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 4 - FUNCTION equals(__table2.id : 3, 5_UInt8 :: 4) -> equals(__table2.id, 5_UInt8) UInt8 : 5 + FUNCTION equals(__table2.id : 2, 5_UInt8 :: 4) -> equals(__table2.id, 5_UInt8) UInt8 : 5 Positions: 5 0 1 2 3 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: LEFT Strictness: ALL Algorithm: HashJoin @@ -367,31 +367,31 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 2 0 1 - Filter ((WHERE + DROP unused columns after JOIN)) + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 1 0 2 + Filter (WHERE) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Filter column: equals(__table1.id, 5_UInt8) (removed) Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT :: 1 -> __table1.value String : 1 - INPUT :: 2 -> __table2.value String : 2 - INPUT :: 3 -> __table2.id UInt64 : 3 + INPUT :: 2 -> __table2.id UInt64 : 2 + INPUT :: 3 -> __table2.value String : 3 COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 4 FUNCTION equals(__table1.id : 0, 5_UInt8 :: 4) -> equals(__table1.id, 5_UInt8) UInt8 : 5 Positions: 5 0 1 2 3 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: RIGHT Strictness: ALL Algorithm: HashJoin @@ -443,18 +443,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 2 0 1 + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 1 0 2 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: RIGHT Strictness: ALL Algorithm: HashJoin @@ -512,31 +512,31 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 2 0 1 - Filter ((WHERE + DROP unused columns after JOIN)) + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 1 0 2 + Filter (WHERE) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Filter column: equals(__table1.id, 5_UInt8) (removed) Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT :: 1 -> __table1.value String : 1 - INPUT :: 2 -> __table2.value String : 2 - INPUT :: 3 -> __table2.id UInt64 : 3 + INPUT :: 2 -> __table2.id UInt64 : 2 + INPUT :: 3 -> __table2.value String : 3 COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 4 FUNCTION equals(__table1.id : 0, 5_UInt8 :: 4) -> equals(__table1.id, 5_UInt8) UInt8 : 5 Positions: 5 0 1 2 3 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: FULL Strictness: ALL Algorithm: HashJoin @@ -588,31 +588,31 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 2 0 1 - Filter ((WHERE + DROP unused columns after JOIN)) + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 1 0 2 + Filter (WHERE) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Filter column: equals(__table2.id, 5_UInt8) (removed) Actions: INPUT :: 0 -> __table1.id UInt64 : 0 INPUT :: 1 -> __table1.value String : 1 - INPUT :: 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT :: 3 -> __table2.value String : 3 COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 4 - FUNCTION equals(__table2.id : 3, 5_UInt8 :: 4) -> equals(__table2.id, 5_UInt8) UInt8 : 5 + FUNCTION equals(__table2.id : 2, 5_UInt8 :: 4) -> equals(__table2.id, 5_UInt8) UInt8 : 5 Positions: 5 0 1 2 3 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: FULL Strictness: ALL Algorithm: HashJoin @@ -664,34 +664,34 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 2 0 1 - Filter ((WHERE + DROP unused columns after JOIN)) + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 1 0 2 + Filter (WHERE) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Filter column: and(equals(__table1.id, 5_UInt8), equals(__table2.id, 6_UInt8)) (removed) Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT :: 1 -> __table1.value String : 1 - INPUT :: 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT :: 3 -> __table2.value String : 3 COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 4 COLUMN Const(UInt8) -> 6_UInt8 UInt8 : 5 FUNCTION equals(__table1.id : 0, 5_UInt8 :: 4) -> equals(__table1.id, 5_UInt8) UInt8 : 6 - FUNCTION equals(__table2.id : 3, 6_UInt8 :: 5) -> equals(__table2.id, 6_UInt8) UInt8 : 4 + FUNCTION equals(__table2.id : 2, 6_UInt8 :: 5) -> equals(__table2.id, 6_UInt8) UInt8 : 4 FUNCTION and(equals(__table1.id, 5_UInt8) :: 6, equals(__table2.id, 6_UInt8) :: 4) -> and(equals(__table1.id, 5_UInt8), equals(__table2.id, 6_UInt8)) UInt8 : 5 Positions: 5 0 1 2 3 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: FULL Strictness: ALL Algorithm: HashJoin From a019dd0410adff0b0e64eb0d818b5f25056fc764 Mon Sep 17 00:00:00 2001 From: vdimir Date: Mon, 7 Oct 2024 17:12:28 +0000 Subject: [PATCH 154/680] fix clang tidy --- src/Processors/QueryPlan/JoinStep.cpp | 4 ++-- src/Processors/Transforms/ColumnPermuteTransform.cpp | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index dcedc57713d..9cb06042cf6 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -19,7 +19,7 @@ namespace ErrorCodes namespace { -static std::vector> describeJoinActions(const JoinPtr & join) +std::vector> describeJoinActions(const JoinPtr & join) { std::vector> description; const auto & table_join = join->getTableJoin(); @@ -188,7 +188,7 @@ void JoinStep::updateOutputStream() auto column_permutation = getPermutationForBlock(result_header, input_streams[0].header, input_streams[1].header, required_output); if (!column_permutation.empty()) - result_header = ColumnPermuteTransform::permute(std::move(result_header), column_permutation); + result_header = ColumnPermuteTransform::permute(result_header, column_permutation); output_stream = DataStream { .header = result_header }; } diff --git a/src/Processors/Transforms/ColumnPermuteTransform.cpp b/src/Processors/Transforms/ColumnPermuteTransform.cpp index 67c7996cbe0..f371689814c 100644 --- a/src/Processors/Transforms/ColumnPermuteTransform.cpp +++ b/src/Processors/Transforms/ColumnPermuteTransform.cpp @@ -12,7 +12,7 @@ void applyPermutation(std::vector & data, const std::vector & permuta std::vector res; res.reserve(permutation.size()); for (size_t i : permutation) - res.emplace_back(std::move(data[i])); + res.push_back(data[i]); data = std::move(res); } From 07da0c99b8318cd368c52a0d573e598599207196 Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 8 Oct 2024 05:52:25 +0000 Subject: [PATCH 155/680] Fix tests --- .../03225_alter_to_json_not_supported.reference | 0 .../03225_alter_to_json_not_supported.sql | 15 --------------- .../03248_string_to_json_alter_fuzz.sql | 4 ++-- 3 files changed, 2 insertions(+), 17 deletions(-) delete mode 100644 tests/queries/0_stateless/03225_alter_to_json_not_supported.reference delete mode 100644 tests/queries/0_stateless/03225_alter_to_json_not_supported.sql diff --git a/tests/queries/0_stateless/03225_alter_to_json_not_supported.reference b/tests/queries/0_stateless/03225_alter_to_json_not_supported.reference deleted file mode 100644 index e69de29bb2d..00000000000 diff --git a/tests/queries/0_stateless/03225_alter_to_json_not_supported.sql b/tests/queries/0_stateless/03225_alter_to_json_not_supported.sql deleted file mode 100644 index 398494d56de..00000000000 --- a/tests/queries/0_stateless/03225_alter_to_json_not_supported.sql +++ /dev/null @@ -1,15 +0,0 @@ -set allow_experimental_json_type = 1; - -drop table if exists test; -create table test (s String) engine=MergeTree order by tuple(); -alter table test modify column s JSON; -- { serverError BAD_ARGUMENTS } -drop table test; - -create table test (s Array(String)) engine=MergeTree order by tuple(); -alter table test modify column s Array(JSON); -- { serverError BAD_ARGUMENTS } -drop table test; - -create table test (s Tuple(String, String)) engine=MergeTree order by tuple(); -alter table test modify column s Tuple(JSON, String); -- { serverError BAD_ARGUMENTS } -drop table test; - diff --git a/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql b/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql index 87e10df9cc8..d4d775732e8 100644 --- a/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql +++ b/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql @@ -7,8 +7,8 @@ drop named collection if exists json_alter_fuzzer; create table test (json String) engine=MergeTree order by tuple(); create named collection json_alter_fuzzer AS json_str='{}'; -insert into test select * from fuzzJSON(json_alter_fuzzer, reuse_output=true, max_output_length=128) limit 200000; -alter table test modify column json JSON settings mutations_sync=1; +insert into test select * from fuzzJSON(json_alter_fuzzer, reuse_output=true, max_output_length=64) limit 200000; +alter table test modify column json JSON(max_dynamic_paths=100) settings mutations_sync=1; select json from test format Null; optimize table test final; select json from test format Null; From c6b58f4db2461bcdc09929b67a84b9d061ddefd5 Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 8 Oct 2024 08:01:45 +0000 Subject: [PATCH 156/680] Better docs --- docs/en/sql-reference/data-types/newjson.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/docs/en/sql-reference/data-types/newjson.md b/docs/en/sql-reference/data-types/newjson.md index f799072a02f..8e9eeb43c72 100644 --- a/docs/en/sql-reference/data-types/newjson.md +++ b/docs/en/sql-reference/data-types/newjson.md @@ -644,12 +644,12 @@ SELECT json, json.a, json.b, json.c FROM test; ``` ```text - ┌─json─────────────────────────┬─json.a─┬─json.b──┬─json.c─────┐ -1. │ {"a":"42"} │ 42 │ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │ -2. │ {"a":"43","b":"Hello"} │ 43 │ Hello │ ᴺᵁᴸᴸ │ -3. │ {"a":"44","b":["1","2","3"]} │ 44 │ [1,2,3] │ ᴺᵁᴸᴸ │ -4. │ {"c":"2020-01-01"} │ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │ 2020-01-01 │ - └──────────────────────────────┴────────┴─────────┴────────────┘ +┌─json─────────────────────────┬─json.a─┬─json.b──┬─json.c─────┐ +│ {"a":"42"} │ 42 │ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │ +│ {"a":"43","b":"Hello"} │ 43 │ Hello │ ᴺᵁᴸᴸ │ +│ {"a":"44","b":["1","2","3"]} │ 44 │ [1,2,3] │ ᴺᵁᴸᴸ │ +│ {"c":"2020-01-01"} │ ᴺᵁᴸᴸ │ ᴺᵁᴸᴸ │ 2020-01-01 │ +└──────────────────────────────┴────────┴─────────┴────────────┘ ``` ## Tips for better usage of the JSON type From 41588b05cf1c8104a1e2e344b043a4eec5db5f10 Mon Sep 17 00:00:00 2001 From: avogar Date: Tue, 8 Oct 2024 08:10:21 +0000 Subject: [PATCH 157/680] Fix test --- ...mic_variant_in_order_by_group_by.reference | 188 +++++++++--------- ...1_dynamic_variant_in_order_by_group_by.sql | 32 +-- 2 files changed, 110 insertions(+), 110 deletions(-) diff --git a/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference index 5c7b4cb0bea..5983dd15f5b 100644 --- a/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference +++ b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.reference @@ -20,98 +20,6 @@ 4 0 1 -4 -3 -2 -0 -1 -4 -3 -2 -[4] -[3] -[2] -[0] -[1] -{'str':0} -{'str':1} -{'str':4} -{'str':3} -{'str':2} -0 -1 -2 -3 -4 -\N -0 -1 -2 -3 -4 -0 -1 -2 -3 -4 -0 -1 -2 -3 -4 -0 -1 -2 -3 -4 -0 -1 -4 -3 -2 -0 -1 -4 -3 -2 -[4] -[3] -[2] -[0] -[1] -{'str':0} -{'str':1} -{'str':4} -{'str':3} -{'str':2} -0 -1 -2 -3 -4 -\N -0 -1 -2 -3 -4 -0 -1 -2 -3 -4 -0 -1 -2 -3 -4 -0 -1 -2 -3 -4 -0 -1 2 3 4 @@ -120,11 +28,11 @@ 2 3 4 -[4] [0] [1] [2] [3] +[4] {'str':0} {'str':1} {'str':2} @@ -166,11 +74,103 @@ 2 3 4 -[4] [0] [1] [2] [3] +[4] +{'str':0} +{'str':1} +{'str':2} +{'str':3} +{'str':4} +0 +1 +2 +3 +4 +\N +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +[0] +[1] +[2] +[3] +[4] +{'str':0} +{'str':1} +{'str':2} +{'str':3} +{'str':4} +0 +1 +2 +3 +4 +\N +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +0 +1 +2 +3 +4 +[0] +[1] +[2] +[3] +[4] {'str':0} {'str':1} {'str':2} diff --git a/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql index 6e4a39c7234..a53b02e8e41 100644 --- a/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql +++ b/tests/queries/0_stateless/03231_dynamic_variant_in_order_by_group_by.sql @@ -53,10 +53,10 @@ select * from test order by tuple(d); select * from test order by array(d); select * from test order by map('str', d); -select * from test group by d; -select * from test group by tuple(d); -select array(d) from test group by array(d); -select map('str', d) from test group by map('str', d); +select * from test group by d order by all; +select * from test group by tuple(d) order by all; +select array(d) from test group by array(d) order by all; +select map('str', d) from test group by map('str', d) order by all; select * from test group by grouping sets ((d), ('str')) order by all; set allow_experimental_analyzer=0; @@ -86,10 +86,10 @@ select * from test order by tuple(d); select * from test order by array(d); select * from test order by map('str', d); -select * from test group by d; -select * from test group by tuple(d); -select array(d) from test group by array(d); -select map('str', d) from test group by map('str', d); +select * from test group by d order by all; +select * from test group by tuple(d) order by all; +select array(d) from test group by array(d) order by all; +select map('str', d) from test group by map('str', d) order by all; select * from test group by grouping sets ((d), ('str')) order by all; drop table test; @@ -124,10 +124,10 @@ select * from test order by tuple(d); select * from test order by array(d); select * from test order by map('str', d); -select * from test group by d; -select * from test group by tuple(d); -select array(d) from test group by array(d); -select map('str', d) from test group by map('str', d); +select * from test group by d order by all; +select * from test group by tuple(d) order by all; +select array(d) from test group by array(d) order by all; +select map('str', d) from test group by map('str', d) order by all; select * from test group by grouping sets ((d), ('str')) order by all; set allow_experimental_analyzer=0; @@ -157,10 +157,10 @@ select * from test order by tuple(d); select * from test order by array(d); select * from test order by map('str', d); -select * from test group by d; -select * from test group by tuple(d); -select array(d) from test group by array(d); -select map('str', d) from test group by map('str', d); +select * from test group by d order by all; +select * from test group by tuple(d) order by all; +select array(d) from test group by array(d) order by all; +select map('str', d) from test group by map('str', d) order by all; select * from test group by grouping sets ((d), ('str')) order by all; drop table test; From c4cc4cca91ee5191cdc37ef3de14ea3cd70514d6 Mon Sep 17 00:00:00 2001 From: avogar Date: Wed, 9 Oct 2024 03:14:48 +0000 Subject: [PATCH 158/680] Fix tests and builds --- .../MergeTreeDataPartWriterCompact.cpp | 2 +- .../MergeTree/MergeTreeDataPartWriterWide.cpp | 2 +- .../03246_alter_from_string_to_json.reference | 160 +++++++++--------- .../03246_alter_from_string_to_json.sql.j2 | 11 +- .../03248_string_to_json_alter_fuzz.reference | 0 .../03248_string_to_json_alter_fuzz.sql | 17 -- 6 files changed, 88 insertions(+), 104 deletions(-) delete mode 100644 tests/queries/0_stateless/03248_string_to_json_alter_fuzz.reference delete mode 100644 tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp index 96623307c8f..377677c5244 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp @@ -57,7 +57,7 @@ MergeTreeDataPartWriterCompact::MergeTreeDataPartWriterCompact( for (const auto & column : columns_list) { auto compression = getCodecDescOrDefault(column.name, default_codec); - addStreams(column, nullptr, compression); + MergeTreeDataPartWriterCompact::addStreams(column, nullptr, compression); } } diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp index ba9d82fd097..f015fcb0d10 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp @@ -102,7 +102,7 @@ MergeTreeDataPartWriterWide::MergeTreeDataPartWriterWide( for (const auto & column : columns_list) { auto compression = getCodecDescOrDefault(column.name, default_codec); - addStreams(column, nullptr, compression); + MergeTreeDataPartWriterWide::addStreams(column, nullptr, compression); } } diff --git a/tests/queries/0_stateless/03246_alter_from_string_to_json.reference b/tests/queries/0_stateless/03246_alter_from_string_to_json.reference index a2d3a799fff..8253c4fef48 100644 --- a/tests/queries/0_stateless/03246_alter_from_string_to_json.reference +++ b/tests/queries/0_stateless/03246_alter_from_string_to_json.reference @@ -15,26 +15,26 @@ key5 {"key1":"value7"} {"key0":"value8"} {"key1":"value9"} -{"key2":"value300000"} -{"key3":"value300001"} -{"key2":"value300002"} -{"key3":"value300003"} -{"key2":"value300004"} -{"key3":"value300005"} -{"key2":"value300006"} -{"key3":"value300007"} -{"key2":"value300008"} -{"key3":"value300009"} -{"key4":"value600000"} -{"key5":"value600001"} -{"key4":"value600002"} -{"key5":"value600003"} -{"key4":"value600004"} -{"key5":"value600005"} -{"key4":"value600006"} -{"key5":"value600007"} -{"key4":"value600008"} -{"key5":"value600009"} +{"key2":"value60000"} +{"key3":"value60001"} +{"key2":"value60002"} +{"key3":"value60003"} +{"key2":"value60004"} +{"key3":"value60005"} +{"key2":"value60006"} +{"key3":"value60007"} +{"key2":"value60008"} +{"key3":"value60009"} +{"key4":"value120000"} +{"key5":"value120001"} +{"key4":"value120002"} +{"key5":"value120003"} +{"key4":"value120004"} +{"key5":"value120005"} +{"key4":"value120006"} +{"key5":"value120007"} +{"key4":"value120008"} +{"key5":"value120009"} value0 \N \N \N \N \N \N value1 \N \N \N \N value2 \N \N \N \N \N @@ -45,26 +45,26 @@ value6 \N \N \N \N \N \N value7 \N \N \N \N value8 \N \N \N \N \N \N value9 \N \N \N \N -\N \N value300000 \N \N \N -\N \N \N value300001 \N \N -\N \N value300002 \N \N \N -\N \N \N value300003 \N \N -\N \N value300004 \N \N \N -\N \N \N value300005 \N \N -\N \N value300006 \N \N \N -\N \N \N value300007 \N \N -\N \N value300008 \N \N \N -\N \N \N value300009 \N \N -\N \N \N \N value600000 \N -\N \N \N \N \N value600001 -\N \N \N \N value600002 \N -\N \N \N \N \N value600003 -\N \N \N \N value600004 \N -\N \N \N \N \N value600005 -\N \N \N \N value600006 \N -\N \N \N \N \N value600007 -\N \N \N \N value600008 \N -\N \N \N \N \N value600009 +\N \N value60000 \N \N \N +\N \N \N value60001 \N \N +\N \N value60002 \N \N \N +\N \N \N value60003 \N \N +\N \N value60004 \N \N \N +\N \N \N value60005 \N \N +\N \N value60006 \N \N \N +\N \N \N value60007 \N \N +\N \N value60008 \N \N \N +\N \N \N value60009 \N \N +\N \N \N \N value120000 \N +\N \N \N \N \N value120001 +\N \N \N \N value120002 \N +\N \N \N \N \N value120003 +\N \N \N \N value120004 \N +\N \N \N \N \N value120005 +\N \N \N \N value120006 \N +\N \N \N \N \N value120007 +\N \N \N \N value120008 \N +\N \N \N \N \N value120009 All paths: ['key0','key1','key2','key3','key4','key5'] Shared data paths: @@ -82,26 +82,26 @@ key5 {"key1":"value7"} {"key0":"value8"} {"key1":"value9"} -{"key2":"value300000"} -{"key3":"value300001"} -{"key2":"value300002"} -{"key3":"value300003"} -{"key2":"value300004"} -{"key3":"value300005"} -{"key2":"value300006"} -{"key3":"value300007"} -{"key2":"value300008"} -{"key3":"value300009"} -{"key4":"value600000"} -{"key5":"value600001"} -{"key4":"value600002"} -{"key5":"value600003"} -{"key4":"value600004"} -{"key5":"value600005"} -{"key4":"value600006"} -{"key5":"value600007"} -{"key4":"value600008"} -{"key5":"value600009"} +{"key2":"value60000"} +{"key3":"value60001"} +{"key2":"value60002"} +{"key3":"value60003"} +{"key2":"value60004"} +{"key3":"value60005"} +{"key2":"value60006"} +{"key3":"value60007"} +{"key2":"value60008"} +{"key3":"value60009"} +{"key4":"value120000"} +{"key5":"value120001"} +{"key4":"value120002"} +{"key5":"value120003"} +{"key4":"value120004"} +{"key5":"value120005"} +{"key4":"value120006"} +{"key5":"value120007"} +{"key4":"value120008"} +{"key5":"value120009"} value0 \N \N \N \N \N \N value1 \N \N \N \N value2 \N \N \N \N \N @@ -112,23 +112,23 @@ value6 \N \N \N \N \N \N value7 \N \N \N \N value8 \N \N \N \N \N \N value9 \N \N \N \N -\N \N value300000 \N \N \N -\N \N \N value300001 \N \N -\N \N value300002 \N \N \N -\N \N \N value300003 \N \N -\N \N value300004 \N \N \N -\N \N \N value300005 \N \N -\N \N value300006 \N \N \N -\N \N \N value300007 \N \N -\N \N value300008 \N \N \N -\N \N \N value300009 \N \N -\N \N \N \N value600000 \N -\N \N \N \N \N value600001 -\N \N \N \N value600002 \N -\N \N \N \N \N value600003 -\N \N \N \N value600004 \N -\N \N \N \N \N value600005 -\N \N \N \N value600006 \N -\N \N \N \N \N value600007 -\N \N \N \N value600008 \N -\N \N \N \N \N value600009 +\N \N value60000 \N \N \N +\N \N \N value60001 \N \N +\N \N value60002 \N \N \N +\N \N \N value60003 \N \N +\N \N value60004 \N \N \N +\N \N \N value60005 \N \N +\N \N value60006 \N \N \N +\N \N \N value60007 \N \N +\N \N value60008 \N \N \N +\N \N \N value60009 \N \N +\N \N \N \N value120000 \N +\N \N \N \N \N value120001 +\N \N \N \N value120002 \N +\N \N \N \N \N value120003 +\N \N \N \N value120004 \N +\N \N \N \N \N value120005 +\N \N \N \N value120006 \N +\N \N \N \N \N value120007 +\N \N \N \N value120008 \N +\N \N \N \N \N value120009 diff --git a/tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 b/tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 index a13867b145d..e8760b659dc 100644 --- a/tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 +++ b/tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 @@ -1,4 +1,5 @@ set allow_experimental_json_type = 1; +set max_block_size = 20000; drop table if exists test; @@ -7,7 +8,7 @@ drop table if exists test; {{ create_command }} -insert into test select number, toJSONString(map('key' || multiIf(number < 300000, number % 2, number < 600000, number % 2 + 2, number % 2 + 4), 'value' || number)) from numbers(1000000); +insert into test select number, toJSONString(map('key' || multiIf(number < 60000, number % 2, number < 120000, number % 2 + 2, number % 2 + 4), 'value' || number)) from numbers(200000); alter table test modify column json JSON settings mutations_sync=1; @@ -16,11 +17,11 @@ select distinctJSONPaths(json) from test; select 'Shared data paths:'; select distinct (arrayJoin(JSONSharedDataPaths(json))) as path from test order by path; select json from test order by x limit 10; -select json from test order by x limit 10 offset 300000; -select json from test order by x limit 10 offset 600000; +select json from test order by x limit 10 offset 60000; +select json from test order by x limit 10 offset 120000; select json.key0, json.key1, json.key2, json.key3, json.key4, json.key5 from test order by x limit 10; -select json.key0, json.key1, json.key2, json.key3, json.key4, json.key5 from test order by x limit 10 offset 300000; -select json.key0, json.key1, json.key2, json.key3, json.key4, json.key5 from test order by x limit 10 offset 600000; +select json.key0, json.key1, json.key2, json.key3, json.key4, json.key5 from test order by x limit 10 offset 60000; +select json.key0, json.key1, json.key2, json.key3, json.key4, json.key5 from test order by x limit 10 offset 120000; select json from test format Null; select json from test order by x format Null; diff --git a/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.reference b/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.reference deleted file mode 100644 index e69de29bb2d..00000000000 diff --git a/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql b/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql deleted file mode 100644 index d4d775732e8..00000000000 --- a/tests/queries/0_stateless/03248_string_to_json_alter_fuzz.sql +++ /dev/null @@ -1,17 +0,0 @@ -set allow_experimental_json_type=1; -set max_insert_block_size=10000; -set max_block_size=10000; - -drop table if exists test; -drop named collection if exists json_alter_fuzzer; - -create table test (json String) engine=MergeTree order by tuple(); -create named collection json_alter_fuzzer AS json_str='{}'; -insert into test select * from fuzzJSON(json_alter_fuzzer, reuse_output=true, max_output_length=64) limit 200000; -alter table test modify column json JSON(max_dynamic_paths=100) settings mutations_sync=1; -select json from test format Null; -optimize table test final; -select json from test format Null; -drop named collection json_alter_fuzzer; -drop table test; - From b86f3481d1ebf82601b38a12343fb4b055765cda Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 10 Oct 2024 00:45:45 +0000 Subject: [PATCH 159/680] exclude jobs option for fuzzers --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index b3c19fbb0a4..e4a8c691ded 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -122,7 +122,7 @@ def run_fuzzer(fuzzer: str, timeout: int): if parser.has_section("libfuzzer"): custom_libfuzzer_options = " ".join( - f"-{key}={value}" for key, value in parser["libfuzzer"].items() + f"-{key}={value}" for key, value in parser["libfuzzer"].items() if key != "jobs" ) if parser.has_section("fuzzer_arguments"): From c6d6ee27f4e7feaa2dbcedcf2a3c98faef041345 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 10 Oct 2024 00:52:58 +0000 Subject: [PATCH 160/680] Automatic style fix --- tests/fuzz/runner.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index e4a8c691ded..f398b33308e 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -122,7 +122,9 @@ def run_fuzzer(fuzzer: str, timeout: int): if parser.has_section("libfuzzer"): custom_libfuzzer_options = " ".join( - f"-{key}={value}" for key, value in parser["libfuzzer"].items() if key != "jobs" + f"-{key}={value}" + for key, value in parser["libfuzzer"].items() + if key != "jobs" ) if parser.has_section("fuzzer_arguments"): From df77c6f120beddfe97ff4c8c247473db56c587d7 Mon Sep 17 00:00:00 2001 From: Pavel Kruglov <48961922+Avogar@users.noreply.github.com> Date: Thu, 10 Oct 2024 11:24:47 +0800 Subject: [PATCH 161/680] Print invalid version in exception message --- src/DataTypes/Serializations/SerializationDynamic.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/DataTypes/Serializations/SerializationDynamic.cpp b/src/DataTypes/Serializations/SerializationDynamic.cpp index b00668fa8a4..0e6e866e454 100644 --- a/src/DataTypes/Serializations/SerializationDynamic.cpp +++ b/src/DataTypes/Serializations/SerializationDynamic.cpp @@ -89,7 +89,7 @@ SerializationDynamic::DynamicSerializationVersion::DynamicSerializationVersion(U void SerializationDynamic::DynamicSerializationVersion::checkVersion(UInt64 version) { if (version != V1 && version != V2) - throw Exception(ErrorCodes::INCORRECT_DATA, "Invalid version for Dynamic structure serialization."); + throw Exception(ErrorCodes::INCORRECT_DATA, "Invalid version for Dynamic structure serialization: {}", version); } void SerializationDynamic::serializeBinaryBulkStatePrefix( From a5853ee23022969e8250b4a83ed2ba9a02bcaf77 Mon Sep 17 00:00:00 2001 From: vdimir Date: Thu, 10 Oct 2024 11:52:34 +0000 Subject: [PATCH 162/680] fix empty outer_scope_columns in JoinStep --- src/Planner/PlannerJoinTree.cpp | 12 ++++++- ...convert_outer_join_to_inner_join.reference | 36 +++++++++---------- 2 files changed, 29 insertions(+), 19 deletions(-) diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp index 0e82215c12a..ee0b68f4b63 100644 --- a/src/Planner/PlannerJoinTree.cpp +++ b/src/Planner/PlannerJoinTree.cpp @@ -1684,13 +1684,23 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ } auto join_pipeline_type = join_algorithm->pipelineType(); + + ColumnIdentifierSet outer_scope_columns_nonempty; + if (outer_scope_columns.empty()) + { + if (left_header.columns() > 1) + outer_scope_columns_nonempty.insert(left_header.getByPosition(0).name); + else if (right_header.columns() > 1) + outer_scope_columns_nonempty.insert(right_header.getByPosition(0).name); + } + auto join_step = std::make_unique( left_plan.getCurrentDataStream(), right_plan.getCurrentDataStream(), std::move(join_algorithm), settings[Setting::max_block_size], settings[Setting::max_threads], - outer_scope_columns, + outer_scope_columns.empty() ? outer_scope_columns_nonempty : outer_scope_columns, false /*optimize_read_in_order*/, true /*optimize_skip_unused_shards*/); join_step->inner_table_selection_mode = settings[Setting::query_plan_join_inner_table_selection]; diff --git a/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.reference b/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.reference index d35bdeff98b..5fde4f80c5d 100644 --- a/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.reference +++ b/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.reference @@ -5,18 +5,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 0 2 1 + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 0 1 2 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: INNER Strictness: ALL Algorithm: HashJoin @@ -75,18 +75,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 0 2 1 + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 0 1 2 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: INNER Strictness: ALL Algorithm: HashJoin @@ -145,18 +145,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 + INPUT : 2 -> __table2.id UInt64 : 2 + INPUT : 3 -> __table2.value String : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.value :: 2 -> rhs.value String : 1 - ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 -Positions: 4 0 2 1 + ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 + ALIAS __table2.value :: 3 -> rhs.value String : 2 +Positions: 4 0 1 2 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.value String __table2.id UInt64 + __table2.value String Type: INNER Strictness: ALL Algorithm: HashJoin From 845c4a543c091f5951b5e5b2063531ad264da6d1 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 10 Oct 2024 18:59:48 +0000 Subject: [PATCH 163/680] add test for libfuzzer --- utils/CMakeLists.txt | 4 ++++ utils/libfuzzer-test/CMakeLists.txt | 1 + utils/libfuzzer-test/README.md | 1 + utils/libfuzzer-test/test_basic_fuzzer/CMakeLists.txt | 1 + utils/libfuzzer-test/test_basic_fuzzer/main.cpp | 11 +++++++++++ 5 files changed, 18 insertions(+) create mode 100644 utils/libfuzzer-test/CMakeLists.txt create mode 100644 utils/libfuzzer-test/README.md create mode 100644 utils/libfuzzer-test/test_basic_fuzzer/CMakeLists.txt create mode 100644 utils/libfuzzer-test/test_basic_fuzzer/main.cpp diff --git a/utils/CMakeLists.txt b/utils/CMakeLists.txt index ec44a1e1de9..8c706ee6b67 100644 --- a/utils/CMakeLists.txt +++ b/utils/CMakeLists.txt @@ -23,3 +23,7 @@ if (ENABLE_UTILS) add_subdirectory (keeper-data-dumper) add_subdirectory (memcpy-bench) endif () + +if (ENABLE_FUZZING) + add_subdirectory (libfuzzer-test) +endif () diff --git a/utils/libfuzzer-test/CMakeLists.txt b/utils/libfuzzer-test/CMakeLists.txt new file mode 100644 index 00000000000..8765787ff8a --- /dev/null +++ b/utils/libfuzzer-test/CMakeLists.txt @@ -0,0 +1 @@ +add_subdirectory (test_basic_fuzzer) diff --git a/utils/libfuzzer-test/README.md b/utils/libfuzzer-test/README.md new file mode 100644 index 00000000000..5598cbdb961 --- /dev/null +++ b/utils/libfuzzer-test/README.md @@ -0,0 +1 @@ +This folder contains various stuff intended to test libfuzzer functionality. diff --git a/utils/libfuzzer-test/test_basic_fuzzer/CMakeLists.txt b/utils/libfuzzer-test/test_basic_fuzzer/CMakeLists.txt new file mode 100644 index 00000000000..dc927f35a4b --- /dev/null +++ b/utils/libfuzzer-test/test_basic_fuzzer/CMakeLists.txt @@ -0,0 +1 @@ +add_executable (test_basic_fuzzer main.cpp) diff --git a/utils/libfuzzer-test/test_basic_fuzzer/main.cpp b/utils/libfuzzer-test/test_basic_fuzzer/main.cpp new file mode 100644 index 00000000000..7ccad63273d --- /dev/null +++ b/utils/libfuzzer-test/test_basic_fuzzer/main.cpp @@ -0,0 +1,11 @@ +#include +#include + +extern "C" int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) +{ + if (size > 0 && data[0] == 'H') + if (size > 1 && data[1] == 'I') + if (size > 2 && data[2] == '!') + __builtin_trap(); + return 0; +} From 6d8125d520a1c00efde8377f27a096aec56a41db Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Thu, 10 Oct 2024 15:38:22 -0400 Subject: [PATCH 164/680] trigger build From b064d757ca0af321e1a4929d6be1fe3b12dd200f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Thu, 10 Oct 2024 15:48:33 -0400 Subject: [PATCH 165/680] trigger build From ca5f3c50d2e9a74a0a5a7cf9b5ef7f42e171fba7 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Thu, 10 Oct 2024 16:10:02 -0400 Subject: [PATCH 166/680] trigger build --- src/DataTypes/fuzzers/CMakeLists.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/src/DataTypes/fuzzers/CMakeLists.txt b/src/DataTypes/fuzzers/CMakeLists.txt index 8dedd3470e2..8940586fc70 100644 --- a/src/DataTypes/fuzzers/CMakeLists.txt +++ b/src/DataTypes/fuzzers/CMakeLists.txt @@ -1,2 +1,3 @@ clickhouse_add_executable(data_type_deserialization_fuzzer data_type_deserialization_fuzzer.cpp ${SRCS}) + target_link_libraries(data_type_deserialization_fuzzer PRIVATE clickhouse_aggregate_functions dbms) From 8f9ccdf69c983440d698deb0497250a92dcf76ec Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 10 Oct 2024 23:08:52 +0000 Subject: [PATCH 167/680] fix parser --- tests/fuzz/runner.py | 30 +++++++++++++----------------- 1 file changed, 13 insertions(+), 17 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index f398b33308e..c6c978c3508 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -17,8 +17,7 @@ FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") def report(source: str, reason: str, call_stack: list, test_unit: str): print(f"########### REPORT: {source} {reason} {test_unit}") - for line in call_stack: - print(f" {line}") + print("".join(call_stack)) print("########### END OF REPORT ###########") @@ -31,31 +30,28 @@ def process_error(error: str): ERROR = r"^==\d+==\s?ERROR: (\S+): (.*)" error_source = "" error_reason = "" - TEST_UNIT_LINE = r"artifact_prefix='.*/'; Test unit written to (.*)" - call_stack = [] - is_call_stack = False + test_unit = "" + TEST_UNIT_LINE = r"artifact_prefix='.*\/'; Test unit written to (.*)" + error_info = [] + is_error = False # pylint: disable=unused-variable - for line_num, line in enumerate(error.splitlines(), 1): - if is_call_stack: - if re.search(r"^==\d+==", line): - is_call_stack = False - continue - call_stack.append(line) - continue - - if call_stack: + for line_num, line in enumerate(sys.stdin, 1): + if is_error: + error_info.append(line) match = re.search(TEST_UNIT_LINE, line) if match: - report(error_source, error_reason, call_stack, match.group(1)) - call_stack.clear() + test_unit = match.group(1) continue match = re.search(ERROR, line) if match: + error_info.append(line) error_source = match.group(1) error_reason = match.group(2) - is_call_stack = True + is_error = True + + report(error_source, error_reason, error_info, test_unit) def kill_fuzzer(fuzzer: str): From 85a6bb1d1fc4024d57139008953fb35b5be51288 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 11 Oct 2024 03:11:39 +0000 Subject: [PATCH 168/680] fix parser --- tests/fuzz/runner.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index c6c978c3508..3a462d11172 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -16,9 +16,9 @@ FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") def report(source: str, reason: str, call_stack: list, test_unit: str): - print(f"########### REPORT: {source} {reason} {test_unit}") - print("".join(call_stack)) - print("########### END OF REPORT ###########") + logging.info("########### REPORT: %s %s %s", source, reason, test_unit) + logging.info("".join(call_stack)) + logging.info("########### END OF REPORT ###########") # pylint: disable=unused-argument @@ -157,7 +157,7 @@ def run_fuzzer(fuzzer: str, timeout: int): ) except subprocess.CalledProcessError as e: # print("Command failed with error:", e) - print("Stderr output: ", e.stderr) + logging.info("Stderr output: %s", e.stderr) process_error(e.stderr) except subprocess.TimeoutExpired as e: logging.info("Timeout for %s", cmd_line) From 5e99f63e7e5b825813f01ac56a0094d6c95c276a Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 11 Oct 2024 04:05:08 +0000 Subject: [PATCH 169/680] fix parser --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 3a462d11172..1d3829598c3 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -36,7 +36,7 @@ def process_error(error: str): is_error = False # pylint: disable=unused-variable - for line_num, line in enumerate(sys.stdin, 1): + for line_num, line in enumerate(error.splitlines(), 1): if is_error: error_info.append(line) match = re.search(TEST_UNIT_LINE, line) From 1bd4be3df127fdc42e4df01dd3c3da938ce6d327 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 01:10:57 +0000 Subject: [PATCH 170/680] prepare for database upload --- tests/fuzz/runner.py | 44 +++++++++++++++++++++++++++++++++++++++----- 1 file changed, 39 insertions(+), 5 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 1d3829598c3..bc6d3864810 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -8,6 +8,7 @@ import signal import subprocess from pathlib import Path from time import sleep +from typing import List from botocore.exceptions import ClientError @@ -26,7 +27,7 @@ def process_fuzzer_output(output: str): pass -def process_error(error: str): +def process_error(error: str) -> list: ERROR = r"^==\d+==\s?ERROR: (\S+): (.*)" error_source = "" error_reason = "" @@ -52,6 +53,7 @@ def process_error(error: str): is_error = True report(error_source, error_reason, error_info, test_unit) + return error_info def kill_fuzzer(fuzzer: str): @@ -64,7 +66,7 @@ def kill_fuzzer(fuzzer: str): os.kill(pid, signal.SIGKILL) -def run_fuzzer(fuzzer: str, timeout: int): +def run_fuzzer(fuzzer: str, timeout: int) -> TestResult: s3 = S3Helper() logging.info("Running fuzzer %s...", fuzzer) @@ -142,8 +144,9 @@ def run_fuzzer(fuzzer: str, timeout: int): cmd_line += " < /dev/null" logging.info("...will execute: %s", cmd_line) - # subprocess.check_call(cmd_line, shell=True) + test_result = TestResult(fuzzer, "OK") + stopwatch = Stopwatch() try: result = subprocess.run( cmd_line, @@ -158,19 +161,36 @@ def run_fuzzer(fuzzer: str, timeout: int): except subprocess.CalledProcessError as e: # print("Command failed with error:", e) logging.info("Stderr output: %s", e.stderr) - process_error(e.stderr) + test_result = TestResult( + fuzzer, + "FAIL", + stopwatch.duration_seconds, + "", + "\n".join(process_error(e.stderr)), + ) except subprocess.TimeoutExpired as e: logging.info("Timeout for %s", cmd_line) kill_fuzzer(fuzzer) sleep(10) process_fuzzer_output(e.stderr) + test_result = TestResult( + fuzzer, + "Timeout", + stopwatch.duration_seconds, + "", + "", + ) else: process_fuzzer_output(result.stderr) + test_result.time = stopwatch.duration_seconds s3.upload_build_directory_to_s3( Path(new_corpus_dir), f"fuzzer/corpus/{fuzzer}", False ) + logging.info("test_result: %s", test_result) + return test_result + def main(): logging.basicConfig(level=logging.INFO) @@ -183,10 +203,17 @@ def main(): if match: timeout += int(match.group(2)) + test_results = [] + stopwatch = Stopwatch() with Path() as current: for fuzzer in current.iterdir(): if (current / fuzzer).is_file() and os.access(current / fuzzer, os.X_OK): - run_fuzzer(fuzzer.name, timeout) + test_results.append(run_fuzzer(fuzzer.name, timeout)) + + prepared_results = prepare_tests_results_for_clickhouse(PRInfo(), test_results, "failure", stopwatch.duration_seconds, stopwatch.start_time_str, "", "libFuzzer") + # ch_helper = ClickHouseHelper() + # ch_helper.insert_events_into(db="default", table="checks", events=prepared_results) + logging.info("prepared_results: %s", prepared_results) if __name__ == "__main__": @@ -198,5 +225,12 @@ if __name__ == "__main__": S3_BUILDS_BUCKET, ) from s3_helper import S3Helper # pylint: disable=import-error,no-name-in-module + from clickhouse_helper import ( # pylint: disable=import-error,no-name-in-module + ClickHouseHelper, + prepare_tests_results_for_clickhouse, + ) + from pr_info import PRInfo # pylint: disable=import-error,no-name-in-module + from stopwatch import Stopwatch # pylint: disable=import-error,no-name-in-module + from report import TestResult # pylint: disable=import-error,no-name-in-module main() From e590d036fed24a126a63c226d4ee6e01d7a66957 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 01:26:24 +0000 Subject: [PATCH 171/680] fix style --- tests/fuzz/runner.py | 23 +++++++++++++++-------- 1 file changed, 15 insertions(+), 8 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index bc6d3864810..313b38d2d86 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -8,7 +8,6 @@ import signal import subprocess from pathlib import Path from time import sleep -from typing import List from botocore.exceptions import ClientError @@ -210,7 +209,15 @@ def main(): if (current / fuzzer).is_file() and os.access(current / fuzzer, os.X_OK): test_results.append(run_fuzzer(fuzzer.name, timeout)) - prepared_results = prepare_tests_results_for_clickhouse(PRInfo(), test_results, "failure", stopwatch.duration_seconds, stopwatch.start_time_str, "", "libFuzzer") + prepared_results = prepare_tests_results_for_clickhouse( + PRInfo(), + test_results, + "failure", + stopwatch.duration_seconds, + stopwatch.start_time_str, + "", + "libFuzzer", + ) # ch_helper = ClickHouseHelper() # ch_helper.insert_events_into(db="default", table="checks", events=prepared_results) logging.info("prepared_results: %s", prepared_results) @@ -221,16 +228,16 @@ if __name__ == "__main__": ACTIVE_DIR = path.dirname(path.abspath(__file__)) sys.path.append((Path(path.dirname(ACTIVE_DIR)) / "ci").as_posix()) - from env_helper import ( # pylint: disable=import-error,no-name-in-module - S3_BUILDS_BUCKET, - ) - from s3_helper import S3Helper # pylint: disable=import-error,no-name-in-module - from clickhouse_helper import ( # pylint: disable=import-error,no-name-in-module + from clickhouse_helper import ( # pylint: disable=import-error,no-name-in-module,unused-import ClickHouseHelper, prepare_tests_results_for_clickhouse, ) + from env_helper import ( # pylint: disable=import-error,no-name-in-module + S3_BUILDS_BUCKET, + ) from pr_info import PRInfo # pylint: disable=import-error,no-name-in-module - from stopwatch import Stopwatch # pylint: disable=import-error,no-name-in-module from report import TestResult # pylint: disable=import-error,no-name-in-module + from s3_helper import S3Helper # pylint: disable=import-error,no-name-in-module + from stopwatch import Stopwatch # pylint: disable=import-error,no-name-in-module main() From 9c790785d63695e16773192c4cdad3ddd27f2a3e Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 02:15:04 +0000 Subject: [PATCH 172/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 313b38d2d86..8dd510a8f6e 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -65,7 +65,7 @@ def kill_fuzzer(fuzzer: str): os.kill(pid, signal.SIGKILL) -def run_fuzzer(fuzzer: str, timeout: int) -> TestResult: +def run_fuzzer(fuzzer: str, timeout: int): s3 = S3Helper() logging.info("Running fuzzer %s...", fuzzer) From fbbac87299ed8a6cec447786eed5afb628c48b66 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 02:57:58 +0000 Subject: [PATCH 173/680] add requests --- docker/test/libfuzzer/requirements.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/docker/test/libfuzzer/requirements.txt b/docker/test/libfuzzer/requirements.txt index 74147513e76..fd19ad04d8f 100644 --- a/docker/test/libfuzzer/requirements.txt +++ b/docker/test/libfuzzer/requirements.txt @@ -26,3 +26,4 @@ wadllib==1.3.6 wheel==0.37.1 zipp==1.0.0 boto3 +requests From 7ed274559330501da9f3d570cc7460ec22926e79 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 03:57:15 +0000 Subject: [PATCH 174/680] add github --- docker/test/libfuzzer/requirements.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/docker/test/libfuzzer/requirements.txt b/docker/test/libfuzzer/requirements.txt index fd19ad04d8f..bebf26db0bf 100644 --- a/docker/test/libfuzzer/requirements.txt +++ b/docker/test/libfuzzer/requirements.txt @@ -27,3 +27,4 @@ wheel==0.37.1 zipp==1.0.0 boto3 requests +github From c1956d4458b9722371610047fda01cccc7278fbb Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 04:49:53 +0000 Subject: [PATCH 175/680] add pygithub --- docker/test/libfuzzer/requirements.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docker/test/libfuzzer/requirements.txt b/docker/test/libfuzzer/requirements.txt index bebf26db0bf..d73af2861e6 100644 --- a/docker/test/libfuzzer/requirements.txt +++ b/docker/test/libfuzzer/requirements.txt @@ -27,4 +27,4 @@ wheel==0.37.1 zipp==1.0.0 boto3 requests -github +pygithub From 9ebd2fc4dbd3c6407b9bfb1cc9ce9b0c4708cb0f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 05:42:19 +0000 Subject: [PATCH 176/680] add unidiff --- docker/test/libfuzzer/requirements.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/docker/test/libfuzzer/requirements.txt b/docker/test/libfuzzer/requirements.txt index d73af2861e6..3fd33058a6b 100644 --- a/docker/test/libfuzzer/requirements.txt +++ b/docker/test/libfuzzer/requirements.txt @@ -28,3 +28,4 @@ zipp==1.0.0 boto3 requests pygithub +unidiff From 7981e99bee1c0f4a6f79ddcace1c53183c883d18 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 14:18:19 +0000 Subject: [PATCH 177/680] use func-tester --- tests/ci/ci_config.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/ci_config.py b/tests/ci/ci_config.py index a34ef624ce3..7637c096474 100644 --- a/tests/ci/ci_config.py +++ b/tests/ci/ci_config.py @@ -523,7 +523,7 @@ class CI: run_by_labels=[Tags.libFuzzer], timeout=10800, run_command='libfuzzer_test_check.py "$CHECK_NAME"', - runner_type=Runners.STYLE_CHECKER, + runner_type=Runners.FUNC_TESTER, ), JobNames.DOCKER_SERVER: CommonJobConfigs.DOCKER_SERVER.with_properties( required_builds=[BuildNames.PACKAGE_RELEASE, BuildNames.PACKAGE_AARCH64] From f5a99dde8651fcbdcfdc8ba26c7cde4fa86d37c1 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 21:23:04 +0000 Subject: [PATCH 178/680] test results to output directory --- docker/test/libfuzzer/requirements.txt | 3 -- tests/fuzz/runner.py | 57 ++++++++------------------ 2 files changed, 18 insertions(+), 42 deletions(-) diff --git a/docker/test/libfuzzer/requirements.txt b/docker/test/libfuzzer/requirements.txt index 3fd33058a6b..74147513e76 100644 --- a/docker/test/libfuzzer/requirements.txt +++ b/docker/test/libfuzzer/requirements.txt @@ -26,6 +26,3 @@ wadllib==1.3.6 wheel==0.37.1 zipp==1.0.0 boto3 -requests -pygithub -unidiff diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 8dd510a8f6e..a8d48d7c5f3 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -13,6 +13,7 @@ from botocore.exceptions import ClientError DEBUGGER = os.getenv("DEBUGGER", "") FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") +OUTPUT = "/test_output" def report(source: str, reason: str, call_stack: list, test_unit: str): @@ -121,7 +122,7 @@ def run_fuzzer(fuzzer: str, timeout: int): custom_libfuzzer_options = " ".join( f"-{key}={value}" for key, value in parser["libfuzzer"].items() - if key != "jobs" + if key != "jobs" and key != "exact_artifact_path" ) if parser.has_section("fuzzer_arguments"): @@ -130,8 +131,14 @@ def run_fuzzer(fuzzer: str, timeout: int): for key, value in parser["fuzzer_arguments"].items() ) + exact_artifact_path = f"{OUTPUT}/{fuzzer}.unit" + status_path = f"{OUTPUT}/{fuzzer}.status" + out_path = f"{OUTPUT}/{fuzzer}.out" + cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {new_corpus_dir} {active_corpus_dir} {seed_corpus_dir}" + cmd_line += f" -exact_artifact_path={exact_artifact_path}" + if custom_libfuzzer_options: cmd_line += f" {custom_libfuzzer_options}" if fuzzer_arguments: @@ -144,12 +151,11 @@ def run_fuzzer(fuzzer: str, timeout: int): logging.info("...will execute: %s", cmd_line) - test_result = TestResult(fuzzer, "OK") stopwatch = Stopwatch() try: result = subprocess.run( cmd_line, - stderr=subprocess.PIPE, + stderr=open(out_path, "w"), stdout=subprocess.DEVNULL, text=True, check=True, @@ -160,36 +166,24 @@ def run_fuzzer(fuzzer: str, timeout: int): except subprocess.CalledProcessError as e: # print("Command failed with error:", e) logging.info("Stderr output: %s", e.stderr) - test_result = TestResult( - fuzzer, - "FAIL", - stopwatch.duration_seconds, - "", - "\n".join(process_error(e.stderr)), - ) + with open(status_path, "w") as status: + status.write(f"FAIL\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n") except subprocess.TimeoutExpired as e: logging.info("Timeout for %s", cmd_line) kill_fuzzer(fuzzer) sleep(10) process_fuzzer_output(e.stderr) - test_result = TestResult( - fuzzer, - "Timeout", - stopwatch.duration_seconds, - "", - "", - ) + with open(status_path,"w") as status: + status.write(f"Timeout\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n") else: process_fuzzer_output(result.stderr) - test_result.time = stopwatch.duration_seconds + with open(status_path,"w") as status: + status.write(f"OK\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n") s3.upload_build_directory_to_s3( Path(new_corpus_dir), f"fuzzer/corpus/{fuzzer}", False ) - logging.info("test_result: %s", test_result) - return test_result - def main(): logging.basicConfig(level=logging.INFO) @@ -202,25 +196,16 @@ def main(): if match: timeout += int(match.group(2)) - test_results = [] stopwatch = Stopwatch() with Path() as current: for fuzzer in current.iterdir(): if (current / fuzzer).is_file() and os.access(current / fuzzer, os.X_OK): - test_results.append(run_fuzzer(fuzzer.name, timeout)) + run_fuzzer(fuzzer.name, timeout) + + subprocess.check_call(f"ls -al {OUTPUT}", shell=True) - prepared_results = prepare_tests_results_for_clickhouse( - PRInfo(), - test_results, - "failure", - stopwatch.duration_seconds, - stopwatch.start_time_str, - "", - "libFuzzer", - ) # ch_helper = ClickHouseHelper() # ch_helper.insert_events_into(db="default", table="checks", events=prepared_results) - logging.info("prepared_results: %s", prepared_results) if __name__ == "__main__": @@ -228,15 +213,9 @@ if __name__ == "__main__": ACTIVE_DIR = path.dirname(path.abspath(__file__)) sys.path.append((Path(path.dirname(ACTIVE_DIR)) / "ci").as_posix()) - from clickhouse_helper import ( # pylint: disable=import-error,no-name-in-module,unused-import - ClickHouseHelper, - prepare_tests_results_for_clickhouse, - ) from env_helper import ( # pylint: disable=import-error,no-name-in-module S3_BUILDS_BUCKET, ) - from pr_info import PRInfo # pylint: disable=import-error,no-name-in-module - from report import TestResult # pylint: disable=import-error,no-name-in-module from s3_helper import S3Helper # pylint: disable=import-error,no-name-in-module from stopwatch import Stopwatch # pylint: disable=import-error,no-name-in-module From ae71f1070fdc459809553a38b289f83b16dcc71f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 21:39:03 +0000 Subject: [PATCH 179/680] fix style --- tests/fuzz/runner.py | 41 ++++++++++++++++++++++++----------------- 1 file changed, 24 insertions(+), 17 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index a8d48d7c5f3..f8d318b174a 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -122,7 +122,7 @@ def run_fuzzer(fuzzer: str, timeout: int): custom_libfuzzer_options = " ".join( f"-{key}={value}" for key, value in parser["libfuzzer"].items() - if key != "jobs" and key != "exact_artifact_path" + if key not in ('jobs', 'exact_artifact_path') ) if parser.has_section("fuzzer_arguments"): @@ -153,32 +153,39 @@ def run_fuzzer(fuzzer: str, timeout: int): stopwatch = Stopwatch() try: - result = subprocess.run( - cmd_line, - stderr=open(out_path, "w"), - stdout=subprocess.DEVNULL, - text=True, - check=True, - shell=True, - errors="replace", - timeout=timeout, - ) + with open(out_path, "wb") as out: + result = subprocess.run( + cmd_line, + stderr=out, + stdout=subprocess.DEVNULL, + text=True, + check=True, + shell=True, + errors="replace", + timeout=timeout, + ) except subprocess.CalledProcessError as e: # print("Command failed with error:", e) logging.info("Stderr output: %s", e.stderr) - with open(status_path, "w") as status: - status.write(f"FAIL\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n") + with open(status_path, "wb") as status: + status.write( + f"FAIL\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" + ) except subprocess.TimeoutExpired as e: logging.info("Timeout for %s", cmd_line) kill_fuzzer(fuzzer) sleep(10) process_fuzzer_output(e.stderr) - with open(status_path,"w") as status: - status.write(f"Timeout\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n") + with open(status_path,"wb") as status: + status.write( + f"Timeout\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" + ) else: process_fuzzer_output(result.stderr) - with open(status_path,"w") as status: - status.write(f"OK\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n") + with open(status_path,"wb") as status: + status.write( + f"OK\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" + ) s3.upload_build_directory_to_s3( Path(new_corpus_dir), f"fuzzer/corpus/{fuzzer}", False From 55a24facd29cb8cb68992334b6a68456fa966c19 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 21:39:48 +0000 Subject: [PATCH 180/680] fix style --- tests/fuzz/runner.py | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index f8d318b174a..e933c94f2a8 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -203,7 +203,6 @@ def main(): if match: timeout += int(match.group(2)) - stopwatch = Stopwatch() with Path() as current: for fuzzer in current.iterdir(): if (current / fuzzer).is_file() and os.access(current / fuzzer, os.X_OK): From 7a096859a2c1056df602e5d0d4555bff5641a1db Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Wed, 16 Oct 2024 21:47:00 +0000 Subject: [PATCH 181/680] Automatic style fix --- tests/fuzz/runner.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index e933c94f2a8..a8ca8246ed2 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -122,7 +122,7 @@ def run_fuzzer(fuzzer: str, timeout: int): custom_libfuzzer_options = " ".join( f"-{key}={value}" for key, value in parser["libfuzzer"].items() - if key not in ('jobs', 'exact_artifact_path') + if key not in ("jobs", "exact_artifact_path") ) if parser.has_section("fuzzer_arguments"): @@ -176,13 +176,13 @@ def run_fuzzer(fuzzer: str, timeout: int): kill_fuzzer(fuzzer) sleep(10) process_fuzzer_output(e.stderr) - with open(status_path,"wb") as status: + with open(status_path, "wb") as status: status.write( f"Timeout\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) else: process_fuzzer_output(result.stderr) - with open(status_path,"wb") as status: + with open(status_path, "wb") as status: status.write( f"OK\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) From eb7bf08da5c3a693c8ebb8617eb35d36a029e01f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 16 Oct 2024 22:34:40 +0000 Subject: [PATCH 182/680] fix --- tests/fuzz/runner.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index a8ca8246ed2..f483608605b 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -167,7 +167,7 @@ def run_fuzzer(fuzzer: str, timeout: int): except subprocess.CalledProcessError as e: # print("Command failed with error:", e) logging.info("Stderr output: %s", e.stderr) - with open(status_path, "wb") as status: + with open(status_path, "w", encoding="utf-8") as status: status.write( f"FAIL\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) @@ -176,13 +176,13 @@ def run_fuzzer(fuzzer: str, timeout: int): kill_fuzzer(fuzzer) sleep(10) process_fuzzer_output(e.stderr) - with open(status_path, "wb") as status: + with open(status_path, "w", encoding="utf-8") as status: status.write( f"Timeout\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) else: process_fuzzer_output(result.stderr) - with open(status_path, "wb") as status: + with open(status_path, "w", encoding="utf-8") as status: status.write( f"OK\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) From 84c664dadaa5bac20fe3afac3e386befcc22fb6a Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 01:00:27 +0000 Subject: [PATCH 183/680] move all s3 stuff to check script --- docker/test/libfuzzer/requirements.txt | 1 - tests/ci/libfuzzer_test_check.py | 39 ++++++++++++++++++- tests/fuzz/runner.py | 53 ++++++++++++-------------- 3 files changed, 61 insertions(+), 32 deletions(-) diff --git a/docker/test/libfuzzer/requirements.txt b/docker/test/libfuzzer/requirements.txt index 74147513e76..3dce93e023b 100644 --- a/docker/test/libfuzzer/requirements.txt +++ b/docker/test/libfuzzer/requirements.txt @@ -25,4 +25,3 @@ six==1.16.0 wadllib==1.3.6 wheel==0.37.1 zipp==1.0.0 -boto3 diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 5de28d5641a..a4f31b1663d 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -11,12 +11,17 @@ from typing import List from build_download_helper import download_fuzzers from clickhouse_helper import CiLogsCredentials from docker_images_helper import DockerImage, get_docker_image, pull_image -from env_helper import REPO_COPY, REPORT_PATH, TEMP_PATH +from env_helper import REPO_COPY, REPORT_PATH, S3_BUILDS_BUCKET, TEMP_PATH from pr_info import PRInfo +from s3_helper import S3Helper from stopwatch import Stopwatch from tee_popen import TeePopen +from botocore.exceptions import ClientError + + NO_CHANGES_MSG = "Nothing to run" +s3 = S3Helper() def get_additional_envs(check_name, run_by_hash_num, run_by_hash_total): @@ -85,6 +90,34 @@ def parse_args(): return parser.parse_args() +def download_corpus(corpus_path: str, fuzzer_name: str): + logging.info("Download corpus for %s ...", fuzzer_name) + + units = [] + + try: + units = s3.download_files( + bucket=S3_BUILDS_BUCKET, + s3_path=f"fuzzer/corpus/{fuzzer_name}/", + file_suffix="", + local_directory=corpus_path, + ) + except ClientError as e: + if e.response["Error"]["Code"] == "NoSuchKey": + logging.debug("No active corpus exists for %s", fuzzer_name) + else: + raise + + logging.info("...downloaded %d units", len(units)) + + +def upload_corpus(fuzzers_path: str): + for file in os.listdir(f"{fuzzers_path}/corpus/"): + s3.upload_build_directory_to_s3( + Path(f"{fuzzers_path}/corpus/{file}"), f"fuzzer/corpus/{file}", False + ) + + def main(): logging.basicConfig(level=logging.INFO) @@ -119,6 +152,7 @@ def main(): for file in os.listdir(fuzzers_path): if file.endswith("_fuzzer"): os.chmod(fuzzers_path / file, 0o777) + download_corpus(f"{fuzzers_path}/{file}.corpus", file) elif file.endswith("_seed_corpus.zip"): corpus_path = fuzzers_path / (file.removesuffix("_seed_corpus.zip") + ".in") with zipfile.ZipFile(fuzzers_path / file, "r") as zfd: @@ -133,7 +167,7 @@ def main(): check_name, run_by_hash_num, run_by_hash_total ) - additional_envs.append("CI=1") + # additional_envs.append("CI=1") ci_logs_credentials = CiLogsCredentials(Path(temp_path) / "export-logs-config.sh") ci_logs_args = ci_logs_credentials.get_docker_arguments( @@ -154,6 +188,7 @@ def main(): retcode = process.wait() if retcode == 0: logging.info("Run successfully") + upload_corpus(fuzzers_path) else: logging.info("Run failed") diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index f483608605b..b4c174de6b1 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -1,6 +1,7 @@ #!/usr/bin/env python3 import configparser +import datetime import logging import os import re @@ -16,6 +17,23 @@ FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") OUTPUT = "/test_output" +class Stopwatch: + def __init__(self): + self.reset() + + @property + def duration_seconds(self) -> float: + return (datetime.datetime.utcnow() - self.start_time).total_seconds() + + @property + def start_time_str(self) -> str: + return self.start_time_str_value + + def reset(self) -> None: + self.start_time = datetime.datetime.utcnow() + self.start_time_str_value = self.start_time.strftime("%Y-%m-%d %H:%M:%S") + + def report(source: str, reason: str, call_stack: list, test_unit: str): logging.info("########### REPORT: %s %s %s", source, reason, test_unit) logging.info("".join(call_stack)) @@ -67,8 +85,6 @@ def kill_fuzzer(fuzzer: str): def run_fuzzer(fuzzer: str, timeout: int): - s3 = S3Helper() - logging.info("Running fuzzer %s...", fuzzer) seed_corpus_dir = f"{fuzzer}.in" @@ -77,20 +93,7 @@ def run_fuzzer(fuzzer: str, timeout: int): seed_corpus_dir = "" active_corpus_dir = f"{fuzzer}.corpus" - try: - s3.download_files( - bucket=S3_BUILDS_BUCKET, - s3_path=f"fuzzer/corpus/{fuzzer}/", - file_suffix="", - local_directory=active_corpus_dir, - ) - except ClientError as e: - if e.response["Error"]["Code"] == "NoSuchKey": - logging.debug("No active corpus exists for %s", fuzzer) - else: - raise - - new_corpus_dir = f"{fuzzer}.corpus_new" + new_corpus_dir = f"{OUTPUT}/corpus/{fuzzer}" if not os.path.exists(new_corpus_dir): os.makedirs(new_corpus_dir) @@ -180,16 +183,18 @@ def run_fuzzer(fuzzer: str, timeout: int): status.write( f"Timeout\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) + os.remove(out_path) else: process_fuzzer_output(result.stderr) with open(status_path, "w", encoding="utf-8") as status: status.write( f"OK\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) + os.remove(out_path) - s3.upload_build_directory_to_s3( - Path(new_corpus_dir), f"fuzzer/corpus/{fuzzer}", False - ) + # s3.upload_build_directory_to_s3( + # Path(new_corpus_dir), f"fuzzer/corpus/{fuzzer}", False + # ) def main(): @@ -215,14 +220,4 @@ def main(): if __name__ == "__main__": - from os import path, sys - - ACTIVE_DIR = path.dirname(path.abspath(__file__)) - sys.path.append((Path(path.dirname(ACTIVE_DIR)) / "ci").as_posix()) - from env_helper import ( # pylint: disable=import-error,no-name-in-module - S3_BUILDS_BUCKET, - ) - from s3_helper import S3Helper # pylint: disable=import-error,no-name-in-module - from stopwatch import Stopwatch # pylint: disable=import-error,no-name-in-module - main() From 0b82913507801367caa147627c8b97f99d3871df Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 01:11:45 +0000 Subject: [PATCH 184/680] fix style --- tests/ci/libfuzzer_test_check.py | 5 ++--- tests/fuzz/runner.py | 2 -- 2 files changed, 2 insertions(+), 5 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index a4f31b1663d..b0cb375bc56 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -8,6 +8,8 @@ import zipfile from pathlib import Path from typing import List +from botocore.exceptions import ClientError + from build_download_helper import download_fuzzers from clickhouse_helper import CiLogsCredentials from docker_images_helper import DockerImage, get_docker_image, pull_image @@ -17,9 +19,6 @@ from s3_helper import S3Helper from stopwatch import Stopwatch from tee_popen import TeePopen -from botocore.exceptions import ClientError - - NO_CHANGES_MSG = "Nothing to run" s3 = S3Helper() diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index b4c174de6b1..3a91d8f62f8 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -10,8 +10,6 @@ import subprocess from pathlib import Path from time import sleep -from botocore.exceptions import ClientError - DEBUGGER = os.getenv("DEBUGGER", "") FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") OUTPUT = "/test_output" From b8f095b6260d647ba50a15094f98651161f2358c Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 02:23:38 +0000 Subject: [PATCH 185/680] fix upload corpus, fix s3 helper to allow listing more than 1000 --- tests/ci/libfuzzer_test_check.py | 8 ++++---- tests/ci/s3_helper.py | 22 +++++++++++++--------- 2 files changed, 17 insertions(+), 13 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index b0cb375bc56..19e72b82712 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -110,10 +110,10 @@ def download_corpus(corpus_path: str, fuzzer_name: str): logging.info("...downloaded %d units", len(units)) -def upload_corpus(fuzzers_path: str): - for file in os.listdir(f"{fuzzers_path}/corpus/"): +def upload_corpus(result_path: str): + for file in os.listdir(f"{result_path}/corpus/"): s3.upload_build_directory_to_s3( - Path(f"{fuzzers_path}/corpus/{file}"), f"fuzzer/corpus/{file}", False + Path(f"{result_path}/corpus/{file}"), f"fuzzer/corpus/{file}", False ) @@ -187,7 +187,7 @@ def main(): retcode = process.wait() if retcode == 0: logging.info("Run successfully") - upload_corpus(fuzzers_path) + upload_corpus(result_path) else: logging.info("Run failed") diff --git a/tests/ci/s3_helper.py b/tests/ci/s3_helper.py index 9a40ad1277f..7d5b68f0222 100644 --- a/tests/ci/s3_helper.py +++ b/tests/ci/s3_helper.py @@ -311,23 +311,27 @@ class S3Helper: def list_prefix( self, s3_prefix_path: str, bucket: str = S3_BUILDS_BUCKET ) -> List[str]: - objects = self.client.list_objects_v2(Bucket=bucket, Prefix=s3_prefix_path) + paginator = self.client.get_paginator('list_objects_v2') + pages = paginator.paginate(Bucket=bucket, Prefix=s3_prefix_path) result = [] - if "Contents" in objects: - for obj in objects["Contents"]: - result.append(obj["Key"]) + for page in pages: + if "Contents" in page: + for obj in page["Contents"]: + result.append(obj["Key"]) return result def list_prefix_non_recursive( self, s3_prefix_path: str, bucket: str = S3_BUILDS_BUCKET ) -> List[str]: - objects = self.client.list_objects_v2(Bucket=bucket, Prefix=s3_prefix_path) + paginator = self.client.get_paginator('list_objects_v2') + pages = paginator.paginate(Bucket=bucket, Prefix=s3_prefix_path) result = [] - if "Contents" in objects: - for obj in objects["Contents"]: - if "/" not in obj["Key"][len(s3_prefix_path) + 1 :]: - result.append(obj["Key"]) + for page in pages: + if "Contents" in page: + for obj in page["Contents"]: + if "/" not in obj["Key"][len(s3_prefix_path) + 1 :]: + result.append(obj["Key"]) return result From 4ba099cd7dd06ef180d5cec57c40597bf69b7051 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 17 Oct 2024 02:29:36 +0000 Subject: [PATCH 186/680] Automatic style fix --- tests/ci/s3_helper.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/ci/s3_helper.py b/tests/ci/s3_helper.py index 7d5b68f0222..46c206f0540 100644 --- a/tests/ci/s3_helper.py +++ b/tests/ci/s3_helper.py @@ -311,7 +311,7 @@ class S3Helper: def list_prefix( self, s3_prefix_path: str, bucket: str = S3_BUILDS_BUCKET ) -> List[str]: - paginator = self.client.get_paginator('list_objects_v2') + paginator = self.client.get_paginator("list_objects_v2") pages = paginator.paginate(Bucket=bucket, Prefix=s3_prefix_path) result = [] for page in pages: @@ -324,7 +324,7 @@ class S3Helper: def list_prefix_non_recursive( self, s3_prefix_path: str, bucket: str = S3_BUILDS_BUCKET ) -> List[str]: - paginator = self.client.get_paginator('list_objects_v2') + paginator = self.client.get_paginator("list_objects_v2") pages = paginator.paginate(Bucket=bucket, Prefix=s3_prefix_path) result = [] for page in pages: From 55d7563c48d4ce467badeaf55796bf8e83cd8173 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 12:42:51 +0000 Subject: [PATCH 187/680] zip corpus --- tests/ci/libfuzzer_test_check.py | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 19e72b82712..bfd3e5c4373 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -23,6 +23,15 @@ NO_CHANGES_MSG = "Nothing to run" s3 = S3Helper() +def zipdir(path, ziph): + # ziph is zipfile handle + for root, dirs, files in os.walk(path): + for file in files: + ziph.write(os.path.join(root, file), + os.path.relpath(os.path.join(root, file), + os.path.join(path, '..'))) + + def get_additional_envs(check_name, run_by_hash_num, run_by_hash_total): result = [] if "DatabaseReplicated" in check_name: @@ -111,10 +120,15 @@ def download_corpus(corpus_path: str, fuzzer_name: str): def upload_corpus(result_path: str): - for file in os.listdir(f"{result_path}/corpus/"): - s3.upload_build_directory_to_s3( - Path(f"{result_path}/corpus/{file}"), f"fuzzer/corpus/{file}", False - ) + with zipfile.ZipFile(f"{result_path}/corpus.zip", "w", zipfile.ZIP_DEFLATED) as zipf: + zipdir(f"{result_path}/corpus/", zipf) + s3.upload_file( + bucket=S3_BUILDS_BUCKET, file_path=f"{result_path}/corpus.zip", s3_path="fuzzer/corpus.zip" + ) + # for file in os.listdir(f"{result_path}/corpus/"): + # s3.upload_build_directory_to_s3( + # Path(f"{result_path}/corpus/{file}"), f"fuzzer/corpus/{file}", False + # ) def main(): From 846d3835f6b6dd5a8f226e36e572f1dd05190669 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 13:00:31 +0000 Subject: [PATCH 188/680] fix style --- tests/ci/libfuzzer_test_check.py | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index bfd3e5c4373..5bf03f269cb 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -27,9 +27,10 @@ def zipdir(path, ziph): # ziph is zipfile handle for root, dirs, files in os.walk(path): for file in files: - ziph.write(os.path.join(root, file), - os.path.relpath(os.path.join(root, file), - os.path.join(path, '..'))) + ziph.write( + os.path.join(root, file), + os.path.relpath(os.path.join(root, file), os.path.join(path, '..')), + ) def get_additional_envs(check_name, run_by_hash_num, run_by_hash_total): @@ -120,10 +121,14 @@ def download_corpus(corpus_path: str, fuzzer_name: str): def upload_corpus(result_path: str): - with zipfile.ZipFile(f"{result_path}/corpus.zip", "w", zipfile.ZIP_DEFLATED) as zipf: + with zipfile.ZipFile( + f"{result_path}/corpus.zip", "w", zipfile.ZIP_DEFLATED + ) as zipf: zipdir(f"{result_path}/corpus/", zipf) s3.upload_file( - bucket=S3_BUILDS_BUCKET, file_path=f"{result_path}/corpus.zip", s3_path="fuzzer/corpus.zip" + bucket=S3_BUILDS_BUCKET, + file_path=f"{result_path}/corpus.zip", + s3_path="fuzzer/corpus.zip", ) # for file in os.listdir(f"{result_path}/corpus/"): # s3.upload_build_directory_to_s3( From 8016e92ccce3306c5aea036594eaa8df9fa03487 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 13:12:04 +0000 Subject: [PATCH 189/680] fix style --- tests/ci/libfuzzer_test_check.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 5bf03f269cb..2e1a540b6a9 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -29,7 +29,7 @@ def zipdir(path, ziph): for file in files: ziph.write( os.path.join(root, file), - os.path.relpath(os.path.join(root, file), os.path.join(path, '..')), + os.path.relpath(os.path.join(root, file), os.path.join(path, "..")), ) From 034c5456a0764bc5b14ca149b1f98b9d57635520 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 13:23:23 +0000 Subject: [PATCH 190/680] fix style --- tests/ci/libfuzzer_test_check.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 2e1a540b6a9..513a1cfa353 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -28,7 +28,7 @@ def zipdir(path, ziph): for root, dirs, files in os.walk(path): for file in files: ziph.write( - os.path.join(root, file), + os.path.join(root, file), os.path.relpath(os.path.join(root, file), os.path.join(path, "..")), ) From 794c38ac4da3bed0d85131f31f3548c2a5ca0ea0 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 13:35:41 +0000 Subject: [PATCH 191/680] fix style --- tests/ci/libfuzzer_test_check.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 513a1cfa353..df46bb0daad 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -25,7 +25,7 @@ s3 = S3Helper() def zipdir(path, ziph): # ziph is zipfile handle - for root, dirs, files in os.walk(path): + for root, _, files in os.walk(path): for file in files: ziph.write( os.path.join(root, file), From ac3ee0477bcb3f0c42f11eccad78919ae5df22e9 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 16:29:19 +0000 Subject: [PATCH 192/680] fix --- tests/ci/libfuzzer_test_check.py | 14 ++++++++------ tests/fuzz/runner.py | 11 ++++++----- 2 files changed, 14 insertions(+), 11 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index df46bb0daad..bed52d2a608 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -120,14 +120,14 @@ def download_corpus(corpus_path: str, fuzzer_name: str): logging.info("...downloaded %d units", len(units)) -def upload_corpus(result_path: str): +def upload_corpus(path: str): with zipfile.ZipFile( - f"{result_path}/corpus.zip", "w", zipfile.ZIP_DEFLATED + f"{path}/corpus.zip", "w", zipfile.ZIP_DEFLATED ) as zipf: - zipdir(f"{result_path}/corpus/", zipf) + zipdir(f"{path}/corpus/", zipf) s3.upload_file( bucket=S3_BUILDS_BUCKET, - file_path=f"{result_path}/corpus.zip", + file_path=f"{path}/corpus.zip", s3_path="fuzzer/corpus.zip", ) # for file in os.listdir(f"{result_path}/corpus/"): @@ -164,13 +164,15 @@ def main(): fuzzers_path = temp_path / "fuzzers" fuzzers_path.mkdir(parents=True, exist_ok=True) + corpus_path = fuzzers_path / "corpus" + corpus_path.mkdir(parents=True, exist_ok=True) download_fuzzers(check_name, reports_path, fuzzers_path) for file in os.listdir(fuzzers_path): if file.endswith("_fuzzer"): os.chmod(fuzzers_path / file, 0o777) - download_corpus(f"{fuzzers_path}/{file}.corpus", file) + download_corpus(f"{corpus_path}/{file}", file) elif file.endswith("_seed_corpus.zip"): corpus_path = fuzzers_path / (file.removesuffix("_seed_corpus.zip") + ".in") with zipfile.ZipFile(fuzzers_path / file, "r") as zfd: @@ -206,7 +208,7 @@ def main(): retcode = process.wait() if retcode == 0: logging.info("Run successfully") - upload_corpus(result_path) + upload_corpus(fuzzers_path) else: logging.info("Run failed") diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 3a91d8f62f8..1b2ae7b98d1 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -90,10 +90,10 @@ def run_fuzzer(fuzzer: str, timeout: int): if not path.exists() or not path.is_dir(): seed_corpus_dir = "" - active_corpus_dir = f"{fuzzer}.corpus" - new_corpus_dir = f"{OUTPUT}/corpus/{fuzzer}" - if not os.path.exists(new_corpus_dir): - os.makedirs(new_corpus_dir) + active_corpus_dir = f"corpus/{fuzzer}" + # new_corpus_dir = f"{OUTPUT}/corpus/{fuzzer}" + # if not os.path.exists(new_corpus_dir): + # os.makedirs(new_corpus_dir) options_file = f"{fuzzer}.options" custom_libfuzzer_options = "" @@ -136,7 +136,8 @@ def run_fuzzer(fuzzer: str, timeout: int): status_path = f"{OUTPUT}/{fuzzer}.status" out_path = f"{OUTPUT}/{fuzzer}.out" - cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {new_corpus_dir} {active_corpus_dir} {seed_corpus_dir}" + cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {active_corpus_dir} {seed_corpus_dir}" + # cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {new_corpus_dir} {active_corpus_dir} {seed_corpus_dir}" cmd_line += f" -exact_artifact_path={exact_artifact_path}" From 73438587f280911aec6f91662aa523419a2d710d Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 17 Oct 2024 16:35:58 +0000 Subject: [PATCH 193/680] Automatic style fix --- tests/ci/libfuzzer_test_check.py | 4 +--- tests/fuzz/runner.py | 4 +++- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index bed52d2a608..c2ceea872a7 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -121,9 +121,7 @@ def download_corpus(corpus_path: str, fuzzer_name: str): def upload_corpus(path: str): - with zipfile.ZipFile( - f"{path}/corpus.zip", "w", zipfile.ZIP_DEFLATED - ) as zipf: + with zipfile.ZipFile(f"{path}/corpus.zip", "w", zipfile.ZIP_DEFLATED) as zipf: zipdir(f"{path}/corpus/", zipf) s3.upload_file( bucket=S3_BUILDS_BUCKET, diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 1b2ae7b98d1..c23f4cbc31c 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -136,7 +136,9 @@ def run_fuzzer(fuzzer: str, timeout: int): status_path = f"{OUTPUT}/{fuzzer}.status" out_path = f"{OUTPUT}/{fuzzer}.out" - cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {active_corpus_dir} {seed_corpus_dir}" + cmd_line = ( + f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {active_corpus_dir} {seed_corpus_dir}" + ) # cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {new_corpus_dir} {active_corpus_dir} {seed_corpus_dir}" cmd_line += f" -exact_artifact_path={exact_artifact_path}" From 66bbf11e074855f9e758be76ec45eb002fe67505 Mon Sep 17 00:00:00 2001 From: kssenii Date: Thu, 17 Oct 2024 13:51:08 +0200 Subject: [PATCH 194/680] Allow to disable background cache download for reading metadata files --- src/Common/ProfileEvents.cpp | 1 + src/Core/Settings.cpp | 3 +++ src/Core/SettingsChangesHistory.cpp | 1 + src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp | 2 +- src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp | 4 ++-- src/IO/ReadSettings.h | 1 + src/Interpreters/Cache/FileSegment.cpp | 10 +++++----- src/Interpreters/Cache/FileSegment.h | 6 +++--- src/Interpreters/Context.cpp | 2 ++ 9 files changed, 19 insertions(+), 11 deletions(-) diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp index ec10e25f74e..b6b669943e2 100644 --- a/src/Common/ProfileEvents.cpp +++ b/src/Common/ProfileEvents.cpp @@ -546,6 +546,7 @@ The server successfully detected this situation and will download merged part fr M(FilesystemCacheLoadMetadataMicroseconds, "Time spent loading filesystem cache metadata", ValueType::Microseconds) \ M(FilesystemCacheEvictedBytes, "Number of bytes evicted from filesystem cache", ValueType::Bytes) \ M(FilesystemCacheEvictedFileSegments, "Number of file segments evicted from filesystem cache", ValueType::Number) \ + M(FilesystemCacheBackgroundDownloadQueuePush, "Number of file segments sent for background download in filesystem cache", ValueType::Number) \ M(FilesystemCacheEvictionSkippedFileSegments, "Number of file segments skipped for eviction because of being in unreleasable state", ValueType::Number) \ M(FilesystemCacheEvictionSkippedEvictingFileSegments, "Number of file segments skipped for eviction because of being in evicting state", ValueType::Number) \ M(FilesystemCacheEvictionTries, "Number of filesystem cache eviction attempts", ValueType::Number) \ diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index cdaa305e804..b656c297288 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -4842,6 +4842,9 @@ Limit on size of a single batch of file segments that a read buffer can request )", 0) \ M(UInt64, filesystem_cache_reserve_space_wait_lock_timeout_milliseconds, 1000, R"( Wait time to lock cache for space reservation in filesystem cache +)", 0) \ + M(Bool, filesystem_cache_enable_background_download_for_metadata_files, true, R"( +Enable background download for metadata files in filesystem cache (related to background_download_threads cache settings) )", 0) \ M(UInt64, temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds, (10 * 60 * 1000), R"( Wait time to lock cache for space reservation for temporary data in filesystem cache diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index ad9499c6d86..46b491b3afc 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -104,6 +104,7 @@ static std::initializer_list completed_range.right); cache_file_reader.reset(); - file_segments->popFront(); + file_segments->completeAndPopFront(settings.filesystem_cache_allow_background_download); if (file_segments->empty() && !nextFileSegmentsBatch()) return false; diff --git a/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp b/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp index 6aedc1f5d04..df6fb871772 100644 --- a/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp +++ b/src/Disks/IO/CachedOnDiskWriteBufferFromFile.cpp @@ -196,7 +196,7 @@ void FileSegmentRangeWriter::completeFileSegment() if (file_segment.isDetached() || file_segment.isCompleted()) return; - file_segment.complete(); + file_segment.complete(false); appendFilesystemCacheLog(file_segment); } @@ -210,7 +210,7 @@ void FileSegmentRangeWriter::jumpToPosition(size_t position) if (position < current_write_offset) throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot jump backwards: {} < {}", position, current_write_offset); - file_segment.complete(); + file_segment.complete(false); file_segments.reset(); } expected_write_offset = position; diff --git a/src/IO/ReadSettings.h b/src/IO/ReadSettings.h index 7d6b9f10931..ac3d7fc9faf 100644 --- a/src/IO/ReadSettings.h +++ b/src/IO/ReadSettings.h @@ -106,6 +106,7 @@ struct ReadSettings bool enable_filesystem_cache_log = false; size_t filesystem_cache_segments_batch_size = 20; size_t filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = 1000; + bool filesystem_cache_allow_background_download = true; bool use_page_cache_for_disks_without_file_cache = false; bool read_from_page_cache_if_exists_otherwise_bypass_cache = false; diff --git a/src/Interpreters/Cache/FileSegment.cpp b/src/Interpreters/Cache/FileSegment.cpp index c356800fa57..944d685d2c1 100644 --- a/src/Interpreters/Cache/FileSegment.cpp +++ b/src/Interpreters/Cache/FileSegment.cpp @@ -627,7 +627,7 @@ void FileSegment::completePartAndResetDownloader() LOG_TEST(log, "Complete batch. ({})", getInfoForLogUnlocked(lk)); } -void FileSegment::complete() +void FileSegment::complete(bool allow_background_download) { ProfileEventTimeIncrement watch(ProfileEvents::FileSegmentCompleteMicroseconds); @@ -704,7 +704,7 @@ void FileSegment::complete() if (is_last_holder) { bool added_to_download_queue = false; - if (background_download_enabled && remote_file_reader) + if (allow_background_download && background_download_enabled && remote_file_reader) { added_to_download_queue = locked_key->addToDownloadQueue(offset(), segment_lock); /// Finish download in background. } @@ -1001,7 +1001,7 @@ void FileSegmentsHolder::reset() ProfileEvents::increment(ProfileEvents::FilesystemCacheUnusedHoldFileSegments, file_segments.size()); for (auto file_segment_it = file_segments.begin(); file_segment_it != file_segments.end();) - file_segment_it = completeAndPopFrontImpl(); + file_segment_it = completeAndPopFrontImpl(false); file_segments.clear(); } @@ -1010,9 +1010,9 @@ FileSegmentsHolder::~FileSegmentsHolder() reset(); } -FileSegments::iterator FileSegmentsHolder::completeAndPopFrontImpl() +FileSegments::iterator FileSegmentsHolder::completeAndPopFrontImpl(bool allow_background_download) { - front().complete(); + front().complete(allow_background_download); CurrentMetrics::sub(CurrentMetrics::FilesystemCacheHoldFileSegments); return file_segments.erase(file_segments.begin()); } diff --git a/src/Interpreters/Cache/FileSegment.h b/src/Interpreters/Cache/FileSegment.h index ee9aee1e354..9d796111659 100644 --- a/src/Interpreters/Cache/FileSegment.h +++ b/src/Interpreters/Cache/FileSegment.h @@ -189,7 +189,7 @@ public: * ========== Methods that must do cv.notify() ================== */ - void complete(); + void complete(bool allow_background_download); void completePartAndResetDownloader(); @@ -297,7 +297,7 @@ struct FileSegmentsHolder final : private boost::noncopyable String toString(bool with_state = false) const; - void popFront() { completeAndPopFrontImpl(); } + void completeAndPopFront(bool allow_background_download) { completeAndPopFrontImpl(allow_background_download); } FileSegment & front() { return *file_segments.front(); } const FileSegment & front() const { return *file_segments.front(); } @@ -319,7 +319,7 @@ struct FileSegmentsHolder final : private boost::noncopyable private: FileSegments file_segments{}; - FileSegments::iterator completeAndPopFrontImpl(); + FileSegments::iterator completeAndPopFrontImpl(bool allow_background_download); }; using FileSegmentsHolderPtr = std::unique_ptr; diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp index 8962be59f86..edffa6cc469 100644 --- a/src/Interpreters/Context.cpp +++ b/src/Interpreters/Context.cpp @@ -239,6 +239,7 @@ namespace Setting extern const SettingsUInt64 use_structure_from_insertion_table_in_table_functions; extern const SettingsString workload; extern const SettingsString compatibility; + extern const SettingsBool filesystem_cache_enable_background_download_for_metadata_files; } namespace MergeTreeSetting @@ -5687,6 +5688,7 @@ ReadSettings Context::getReadSettings() const res.filesystem_cache_segments_batch_size = settings_ref[Setting::filesystem_cache_segments_batch_size]; res.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = settings_ref[Setting::filesystem_cache_reserve_space_wait_lock_timeout_milliseconds]; + res.filesystem_cache_allow_background_download = settings_ref[Setting::filesystem_cache_enable_background_download_for_metadata_files]; res.filesystem_cache_max_download_size = settings_ref[Setting::filesystem_cache_max_download_size]; res.skip_download_if_exceeds_query_cache = settings_ref[Setting::skip_download_if_exceeds_query_cache]; From 8b1608ee21c2c92a16d627e9917317599b4664f2 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 19:01:41 +0000 Subject: [PATCH 195/680] test --- tests/ci/build_download_helper.py | 3 ++- tests/ci/libfuzzer_test_check.py | 3 +++ 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/tests/ci/build_download_helper.py b/tests/ci/build_download_helper.py index 8482abb26e0..1d95aa3f547 100644 --- a/tests/ci/build_download_helper.py +++ b/tests/ci/build_download_helper.py @@ -275,5 +275,6 @@ def download_fuzzers( check_name, reports_path, result_path, - lambda x: x.endswith(("_fuzzer", ".dict", ".options", "_seed_corpus.zip")), + lambda x: x.endswith(("double_delta_decompress_fuzzer", ".dict", ".options", "_seed_corpus.zip")), + # lambda x: x.endswith(("_fuzzer", ".dict", ".options", "_seed_corpus.zip")), ) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index c2ceea872a7..4d9291ffc57 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -3,6 +3,7 @@ import argparse import logging import os +import subprocess import sys import zipfile from pathlib import Path @@ -121,6 +122,8 @@ def download_corpus(corpus_path: str, fuzzer_name: str): def upload_corpus(path: str): + logging.info("Upload corpus from path %s", path) + subprocess.check_call(f"ls -al {path}", shell=True) with zipfile.ZipFile(f"{path}/corpus.zip", "w", zipfile.ZIP_DEFLATED) as zipf: zipdir(f"{path}/corpus/", zipf) s3.upload_file( From 7ad42664da11c0af82469b26369e47357e9a4e54 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 17 Oct 2024 19:11:34 +0000 Subject: [PATCH 196/680] Automatic style fix --- tests/ci/build_download_helper.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/ci/build_download_helper.py b/tests/ci/build_download_helper.py index 1d95aa3f547..2532ad5e64e 100644 --- a/tests/ci/build_download_helper.py +++ b/tests/ci/build_download_helper.py @@ -275,6 +275,8 @@ def download_fuzzers( check_name, reports_path, result_path, - lambda x: x.endswith(("double_delta_decompress_fuzzer", ".dict", ".options", "_seed_corpus.zip")), + lambda x: x.endswith( + ("double_delta_decompress_fuzzer", ".dict", ".options", "_seed_corpus.zip") + ), # lambda x: x.endswith(("_fuzzer", ".dict", ".options", "_seed_corpus.zip")), ) From 10d346a1d4627857ecb61f3e3d913aa7ab0fafcd Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 19:35:06 +0000 Subject: [PATCH 197/680] test --- tests/ci/libfuzzer_test_check.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 4d9291ffc57..a559ba9ad6a 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -124,6 +124,7 @@ def download_corpus(corpus_path: str, fuzzer_name: str): def upload_corpus(path: str): logging.info("Upload corpus from path %s", path) subprocess.check_call(f"ls -al {path}", shell=True) + subprocess.check_call(f"ls -Ral {path}/corpus/", shell=True) with zipfile.ZipFile(f"{path}/corpus.zip", "w", zipfile.ZIP_DEFLATED) as zipf: zipdir(f"{path}/corpus/", zipf) s3.upload_file( From 6e334a2d635d8f26626c95cd60e12cb0489d3ed6 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 19:55:25 +0000 Subject: [PATCH 198/680] test --- tests/ci/libfuzzer_test_check.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index a559ba9ad6a..a78c33e0f72 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -174,7 +174,9 @@ def main(): for file in os.listdir(fuzzers_path): if file.endswith("_fuzzer"): os.chmod(fuzzers_path / file, 0o777) - download_corpus(f"{corpus_path}/{file}", file) + fuzzer_corpus_path = corpus_path / file + fuzzer_corpus_path.mkdir(parents=True, exist_ok=True) + download_corpus(fuzzer_corpus_path, file) elif file.endswith("_seed_corpus.zip"): corpus_path = fuzzers_path / (file.removesuffix("_seed_corpus.zip") + ".in") with zipfile.ZipFile(fuzzers_path / file, "r") as zfd: From 9f55730b6f3bd962e4f55fbe1c17bb451382d69f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 20:15:34 +0000 Subject: [PATCH 199/680] test --- tests/ci/libfuzzer_test_check.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index a78c33e0f72..3dcf36fdaa9 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -176,7 +176,9 @@ def main(): os.chmod(fuzzers_path / file, 0o777) fuzzer_corpus_path = corpus_path / file fuzzer_corpus_path.mkdir(parents=True, exist_ok=True) + subprocess.check_call(f"ls -Ral {corpus_path}", shell=True) download_corpus(fuzzer_corpus_path, file) + subprocess.check_call(f"ls -Ral {fuzzer_corpus_path}", shell=True) elif file.endswith("_seed_corpus.zip"): corpus_path = fuzzers_path / (file.removesuffix("_seed_corpus.zip") + ".in") with zipfile.ZipFile(fuzzers_path / file, "r") as zfd: From debc90d3f0dab0d71bf7c995322509ace394626f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 20:50:21 +0000 Subject: [PATCH 200/680] test --- tests/ci/libfuzzer_test_check.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 3dcf36fdaa9..fa0103deba0 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -180,9 +180,9 @@ def main(): download_corpus(fuzzer_corpus_path, file) subprocess.check_call(f"ls -Ral {fuzzer_corpus_path}", shell=True) elif file.endswith("_seed_corpus.zip"): - corpus_path = fuzzers_path / (file.removesuffix("_seed_corpus.zip") + ".in") + seed_corpus_path = fuzzers_path / (file.removesuffix("_seed_corpus.zip") + ".in") with zipfile.ZipFile(fuzzers_path / file, "r") as zfd: - zfd.extractall(corpus_path) + zfd.extractall(seed_corpus_path) result_path = temp_path / "result_path" result_path.mkdir(parents=True, exist_ok=True) From 01d147eadad27c2ab3e112f4d4f0d166e54cb67f Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 17 Oct 2024 20:56:26 +0000 Subject: [PATCH 201/680] Automatic style fix --- tests/ci/libfuzzer_test_check.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index fa0103deba0..1603e540f00 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -180,7 +180,9 @@ def main(): download_corpus(fuzzer_corpus_path, file) subprocess.check_call(f"ls -Ral {fuzzer_corpus_path}", shell=True) elif file.endswith("_seed_corpus.zip"): - seed_corpus_path = fuzzers_path / (file.removesuffix("_seed_corpus.zip") + ".in") + seed_corpus_path = fuzzers_path / ( + file.removesuffix("_seed_corpus.zip") + ".in" + ) with zipfile.ZipFile(fuzzers_path / file, "r") as zfd: zfd.extractall(seed_corpus_path) From e85ce99262db93753e59e636c69709770e38b3ed Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 21:09:56 +0000 Subject: [PATCH 202/680] test --- tests/ci/libfuzzer_test_check.py | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 1603e540f00..e8c43070e4f 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -175,7 +175,6 @@ def main(): if file.endswith("_fuzzer"): os.chmod(fuzzers_path / file, 0o777) fuzzer_corpus_path = corpus_path / file - fuzzer_corpus_path.mkdir(parents=True, exist_ok=True) subprocess.check_call(f"ls -Ral {corpus_path}", shell=True) download_corpus(fuzzer_corpus_path, file) subprocess.check_call(f"ls -Ral {fuzzer_corpus_path}", shell=True) From 1624dc3e677d8613ff93c227399ba39c7fbb2407 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 21:25:08 +0000 Subject: [PATCH 203/680] zip corpus --- tests/ci/build_download_helper.py | 5 +---- tests/ci/libfuzzer_test_check.py | 5 ----- 2 files changed, 1 insertion(+), 9 deletions(-) diff --git a/tests/ci/build_download_helper.py b/tests/ci/build_download_helper.py index 2532ad5e64e..8482abb26e0 100644 --- a/tests/ci/build_download_helper.py +++ b/tests/ci/build_download_helper.py @@ -275,8 +275,5 @@ def download_fuzzers( check_name, reports_path, result_path, - lambda x: x.endswith( - ("double_delta_decompress_fuzzer", ".dict", ".options", "_seed_corpus.zip") - ), - # lambda x: x.endswith(("_fuzzer", ".dict", ".options", "_seed_corpus.zip")), + lambda x: x.endswith(("_fuzzer", ".dict", ".options", "_seed_corpus.zip")), ) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index e8c43070e4f..fbf0bd87fd7 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -122,9 +122,6 @@ def download_corpus(corpus_path: str, fuzzer_name: str): def upload_corpus(path: str): - logging.info("Upload corpus from path %s", path) - subprocess.check_call(f"ls -al {path}", shell=True) - subprocess.check_call(f"ls -Ral {path}/corpus/", shell=True) with zipfile.ZipFile(f"{path}/corpus.zip", "w", zipfile.ZIP_DEFLATED) as zipf: zipdir(f"{path}/corpus/", zipf) s3.upload_file( @@ -175,9 +172,7 @@ def main(): if file.endswith("_fuzzer"): os.chmod(fuzzers_path / file, 0o777) fuzzer_corpus_path = corpus_path / file - subprocess.check_call(f"ls -Ral {corpus_path}", shell=True) download_corpus(fuzzer_corpus_path, file) - subprocess.check_call(f"ls -Ral {fuzzer_corpus_path}", shell=True) elif file.endswith("_seed_corpus.zip"): seed_corpus_path = fuzzers_path / ( file.removesuffix("_seed_corpus.zip") + ".in" From c67b20b80a55e1c678c7e699f26172326c30d58e Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 17 Oct 2024 21:35:48 +0000 Subject: [PATCH 204/680] fix style --- tests/ci/libfuzzer_test_check.py | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index fbf0bd87fd7..c4c6ca0cdf2 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -3,7 +3,6 @@ import argparse import logging import os -import subprocess import sys import zipfile from pathlib import Path From 5ee699d0597804a6d66c161ab3bba9d282fe7519 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 18 Oct 2024 00:42:37 +0000 Subject: [PATCH 205/680] download corpus zip --- tests/ci/libfuzzer_test_check.py | 36 ++++++++++++++------------------ 1 file changed, 16 insertions(+), 20 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index c4c6ca0cdf2..bb2eb726341 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -99,25 +99,30 @@ def parse_args(): return parser.parse_args() -def download_corpus(corpus_path: str, fuzzer_name: str): - logging.info("Download corpus for %s ...", fuzzer_name) - - units = [] +def download_corpus(path: str): + logging.info("Download corpus...") try: - units = s3.download_files( + s3.download_file( bucket=S3_BUILDS_BUCKET, - s3_path=f"fuzzer/corpus/{fuzzer_name}/", - file_suffix="", - local_directory=corpus_path, + s3_path=f"fuzzer/corpus.zip", + local_file_path=path, ) except ClientError as e: if e.response["Error"]["Code"] == "NoSuchKey": - logging.debug("No active corpus exists for %s", fuzzer_name) + logging.debug("No active corpus exists") else: raise - logging.info("...downloaded %d units", len(units)) + with zipfile.ZipFile(f"{path}/corpus.zip", "r") as zipf: + zipf.extractall(path) + os.remove(f"{path}/corpus.zip") + + units = 0 + for _, _, files in os.walk(path): + units += len(files) + + logging.info("...downloaded %d units", units) def upload_corpus(path: str): @@ -128,10 +133,6 @@ def upload_corpus(path: str): file_path=f"{path}/corpus.zip", s3_path="fuzzer/corpus.zip", ) - # for file in os.listdir(f"{result_path}/corpus/"): - # s3.upload_build_directory_to_s3( - # Path(f"{result_path}/corpus/{file}"), f"fuzzer/corpus/{file}", False - # ) def main(): @@ -162,16 +163,13 @@ def main(): fuzzers_path = temp_path / "fuzzers" fuzzers_path.mkdir(parents=True, exist_ok=True) - corpus_path = fuzzers_path / "corpus" - corpus_path.mkdir(parents=True, exist_ok=True) + download_corpus(fuzzers_path) download_fuzzers(check_name, reports_path, fuzzers_path) for file in os.listdir(fuzzers_path): if file.endswith("_fuzzer"): os.chmod(fuzzers_path / file, 0o777) - fuzzer_corpus_path = corpus_path / file - download_corpus(fuzzer_corpus_path, file) elif file.endswith("_seed_corpus.zip"): seed_corpus_path = fuzzers_path / ( file.removesuffix("_seed_corpus.zip") + ".in" @@ -188,8 +186,6 @@ def main(): check_name, run_by_hash_num, run_by_hash_total ) - # additional_envs.append("CI=1") - ci_logs_credentials = CiLogsCredentials(Path(temp_path) / "export-logs-config.sh") ci_logs_args = ci_logs_credentials.get_docker_arguments( pr_info, stopwatch.start_time_str, check_name From 105f673522eea74a58d833abd57666ca7f52c11f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Fri, 18 Oct 2024 00:54:20 +0000 Subject: [PATCH 206/680] fix --- tests/ci/libfuzzer_test_check.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index bb2eb726341..b7f62836dea 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -105,7 +105,7 @@ def download_corpus(path: str): try: s3.download_file( bucket=S3_BUILDS_BUCKET, - s3_path=f"fuzzer/corpus.zip", + s3_path="fuzzer/corpus.zip", local_file_path=path, ) except ClientError as e: From 5c422be620c9c05495f2dbecf7662804487c8492 Mon Sep 17 00:00:00 2001 From: kssenii Date: Fri, 18 Oct 2024 12:05:48 +0200 Subject: [PATCH 207/680] Remove part of the changes, to be moved to Sync --- src/Core/Settings.cpp | 3 --- src/Core/SettingsChangesHistory.cpp | 1 - src/Interpreters/Cache/FileSegment.cpp | 2 ++ src/Interpreters/Context.cpp | 2 -- 4 files changed, 2 insertions(+), 6 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index b656c297288..cdaa305e804 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -4842,9 +4842,6 @@ Limit on size of a single batch of file segments that a read buffer can request )", 0) \ M(UInt64, filesystem_cache_reserve_space_wait_lock_timeout_milliseconds, 1000, R"( Wait time to lock cache for space reservation in filesystem cache -)", 0) \ - M(Bool, filesystem_cache_enable_background_download_for_metadata_files, true, R"( -Enable background download for metadata files in filesystem cache (related to background_download_threads cache settings) )", 0) \ M(UInt64, temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds, (10 * 60 * 1000), R"( Wait time to lock cache for space reservation for temporary data in filesystem cache diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 46b491b3afc..ad9499c6d86 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -104,7 +104,6 @@ static std::initializer_listaddToDownloadQueue(offset(), segment_lock); /// Finish download in background. } diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp index edffa6cc469..8962be59f86 100644 --- a/src/Interpreters/Context.cpp +++ b/src/Interpreters/Context.cpp @@ -239,7 +239,6 @@ namespace Setting extern const SettingsUInt64 use_structure_from_insertion_table_in_table_functions; extern const SettingsString workload; extern const SettingsString compatibility; - extern const SettingsBool filesystem_cache_enable_background_download_for_metadata_files; } namespace MergeTreeSetting @@ -5688,7 +5687,6 @@ ReadSettings Context::getReadSettings() const res.filesystem_cache_segments_batch_size = settings_ref[Setting::filesystem_cache_segments_batch_size]; res.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = settings_ref[Setting::filesystem_cache_reserve_space_wait_lock_timeout_milliseconds]; - res.filesystem_cache_allow_background_download = settings_ref[Setting::filesystem_cache_enable_background_download_for_metadata_files]; res.filesystem_cache_max_download_size = settings_ref[Setting::filesystem_cache_max_download_size]; res.skip_download_if_exceeds_query_cache = settings_ref[Setting::skip_download_if_exceeds_query_cache]; From c97c6250fcdcd6059753738bd12928a1e3fb2ac7 Mon Sep 17 00:00:00 2001 From: kssenii Date: Fri, 18 Oct 2024 14:06:33 +0200 Subject: [PATCH 208/680] Fix unit test --- src/Interpreters/tests/gtest_filecache.cpp | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/Interpreters/tests/gtest_filecache.cpp b/src/Interpreters/tests/gtest_filecache.cpp index 007b31d9fdc..de767947428 100644 --- a/src/Interpreters/tests/gtest_filecache.cpp +++ b/src/Interpreters/tests/gtest_filecache.cpp @@ -253,7 +253,7 @@ void download(FileSegment & file_segment) download(cache_base_path, file_segment); ASSERT_EQ(file_segment.state(), State::DOWNLOADING); - file_segment.complete(); + file_segment.complete(false); ASSERT_EQ(file_segment.state(), State::DOWNLOADED); } @@ -263,7 +263,7 @@ void assertDownloadFails(FileSegment & file_segment) ASSERT_EQ(file_segment.getDownloadedSize(), 0); std::string failure_reason; ASSERT_FALSE(file_segment.reserve(file_segment.range().size(), 1000, failure_reason)); - file_segment.complete(); + file_segment.complete(false); } void download(const HolderPtr & holder) @@ -971,7 +971,7 @@ TEST_F(FileCacheTest, temporaryData) ASSERT_TRUE(segment->getOrSetDownloader() == DB::FileSegment::getCallerId()); ASSERT_TRUE(segment->reserve(segment->range().size(), 1000, failure_reason)); download(*segment); - segment->complete(); + segment->complete(false); } } From 25ab525c0906a4b6fd3c5cf83f29b1d53f327400 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 04:28:48 +0000 Subject: [PATCH 209/680] job report --- tests/ci/libfuzzer_test_check.py | 79 +++++++++++++++++++++++++++++++- 1 file changed, 78 insertions(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index b7f62836dea..bab624fb144 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -3,6 +3,7 @@ import argparse import logging import os +import re import sys import zipfile from pathlib import Path @@ -15,6 +16,7 @@ from clickhouse_helper import CiLogsCredentials from docker_images_helper import DockerImage, get_docker_image, pull_image from env_helper import REPO_COPY, REPORT_PATH, S3_BUILDS_BUCKET, TEMP_PATH from pr_info import PRInfo +from report import JobReport, TestResult from s3_helper import S3Helper from stopwatch import Stopwatch from tee_popen import TeePopen @@ -135,6 +137,67 @@ def upload_corpus(path: str): ) +def process_error(path: Path) -> list: + ERROR = r"^==\d+==\s?ERROR: (\S+): (.*)" + # error_source = "" + # error_reason = "" + # test_unit = "" + TEST_UNIT_LINE = r"artifact_prefix='.*\/'; Test unit written to (.*)" + error_info = [] + is_error = False + + with open(path, "r") as file: + for line in file: + if is_error: + error_info.append(line) + # match = re.search(TEST_UNIT_LINE, line) + # if match: + # test_unit = match.group(1) + continue + + match = re.search(ERROR, line) + if match: + error_info.append(line) + # error_source = match.group(1) + # error_reason = match.group(2) + is_error = True + + return error_info + + +def read_status(status_path: Path): + result = [] + with open(status_path, "r") as file: + for line in file: + result.append(line) + return result + + +def process_results(result_path: Path): + test_results = [] + oks = 0 + timeouts = 0 + fails = 0 + for file in result_path.glob("*.status"): + fuzzer = file.stem + file_path = file.parent.with_stem(fuzzer) + file_path_unit = file_path.with_suffix(".unit") + file_path_out = file_path.with_suffix(".out") + status = read_status(file) + if status[0] == "OK": + oks += 1 + elif status[0] == "Timeout": + timeouts += 1 + else: + fails += 1 + result = TestResult(fuzzer, status[0], status[2]) + if file_path_unit.exists: + result.set_raw_logs("\n".join(process_error(file_path_out))) + test_results.append(result) + + return [oks, timeouts, fails, test_results] + + def main(): logging.basicConfig(level=logging.INFO) @@ -209,7 +272,21 @@ def main(): else: logging.info("Run failed") - sys.exit(0) + results = process_results(reports_path) + + success = results[1] == 0 and results[2] == 0 + + JobReport( + description=f"OK: {results[0]}, Timeout: {results[1]}, FAIL: {results[2]}", + test_results=results[3], + status= "OK" if success else "FAILURE", + start_time=stopwatch.start_time_str, + duration=stopwatch.duration_seconds, + additional_files=[], + ).dump() + + if not success: + sys.exit(1) if __name__ == "__main__": From 14166b377035febb53c6e0054e1b3664c71cee58 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 04:41:36 +0000 Subject: [PATCH 210/680] fix style --- tests/ci/libfuzzer_test_check.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index bab624fb144..fc1e1f940f2 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -142,11 +142,11 @@ def process_error(path: Path) -> list: # error_source = "" # error_reason = "" # test_unit = "" - TEST_UNIT_LINE = r"artifact_prefix='.*\/'; Test unit written to (.*)" + # TEST_UNIT_LINE = r"artifact_prefix='.*\/'; Test unit written to (.*)" error_info = [] is_error = False - with open(path, "r") as file: + with open(path, "r", encoding="utf-8") as file: for line in file: if is_error: error_info.append(line) @@ -167,7 +167,7 @@ def process_error(path: Path) -> list: def read_status(status_path: Path): result = [] - with open(status_path, "r") as file: + with open(status_path, "r", encoding="utf-8") as file: for line in file: result.append(line) return result @@ -279,7 +279,7 @@ def main(): JobReport( description=f"OK: {results[0]}, Timeout: {results[1]}, FAIL: {results[2]}", test_results=results[3], - status= "OK" if success else "FAILURE", + status="OK" if success else "FAILURE", start_time=stopwatch.start_time_str, duration=stopwatch.duration_seconds, additional_files=[], From 4edc84d262b3ae97b52f244d0376afa1f5e4c497 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 05:39:36 +0000 Subject: [PATCH 211/680] fix --- tests/ci/libfuzzer_test_check.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index fc1e1f940f2..fb91a4e50a2 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -279,7 +279,7 @@ def main(): JobReport( description=f"OK: {results[0]}, Timeout: {results[1]}, FAIL: {results[2]}", test_results=results[3], - status="OK" if success else "FAILURE", + status="SUCCESS" if success else "FAILURE", start_time=stopwatch.start_time_str, duration=stopwatch.duration_seconds, additional_files=[], From ca6ff66591055308319361b21fa2a5b3a0035463 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 06:36:10 +0000 Subject: [PATCH 212/680] fix --- tests/ci/libfuzzer_test_check.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index fb91a4e50a2..e0a985ac7b5 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -16,7 +16,7 @@ from clickhouse_helper import CiLogsCredentials from docker_images_helper import DockerImage, get_docker_image, pull_image from env_helper import REPO_COPY, REPORT_PATH, S3_BUILDS_BUCKET, TEMP_PATH from pr_info import PRInfo -from report import JobReport, TestResult +from report import FAILURE, SUCCESS, JobReport, TestResult from s3_helper import S3Helper from stopwatch import Stopwatch from tee_popen import TeePopen @@ -279,7 +279,7 @@ def main(): JobReport( description=f"OK: {results[0]}, Timeout: {results[1]}, FAIL: {results[2]}", test_results=results[3], - status="SUCCESS" if success else "FAILURE", + status=SUCCESS if success else FAILURE, start_time=stopwatch.start_time_str, duration=stopwatch.duration_seconds, additional_files=[], From ccbd9559adc0262c1ed5b1840350ecc047c61451 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 14:09:40 +0000 Subject: [PATCH 213/680] test --- tests/ci/build_download_helper.py | 3 ++- tests/ci/ci.py | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/ci/build_download_helper.py b/tests/ci/build_download_helper.py index 8482abb26e0..47ea772b502 100644 --- a/tests/ci/build_download_helper.py +++ b/tests/ci/build_download_helper.py @@ -275,5 +275,6 @@ def download_fuzzers( check_name, reports_path, result_path, - lambda x: x.endswith(("_fuzzer", ".dict", ".options", "_seed_corpus.zip")), + lambda x: x.endswith(("test_basic_fuzzer", ".dict", ".options", "_seed_corpus.zip")), + # lambda x: x.endswith(("_fuzzer", ".dict", ".options", "_seed_corpus.zip")), ) diff --git a/tests/ci/ci.py b/tests/ci/ci.py index 10431ce038f..e820f445e7a 100644 --- a/tests/ci/ci.py +++ b/tests/ci/ci.py @@ -1284,6 +1284,7 @@ def main() -> int: dump_to_file=True, ) print(f"Job report url: [{check_url}]") + print(job_report) prepared_events = prepare_tests_results_for_clickhouse( pr_info, job_report.test_results, From 31bf93c58f8dd198b396a3b888beb1e9f7557890 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 14:21:19 +0000 Subject: [PATCH 214/680] test --- tests/ci/ci.py | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/ci/ci.py b/tests/ci/ci.py index e820f445e7a..10431ce038f 100644 --- a/tests/ci/ci.py +++ b/tests/ci/ci.py @@ -1284,7 +1284,6 @@ def main() -> int: dump_to_file=True, ) print(f"Job report url: [{check_url}]") - print(job_report) prepared_events = prepare_tests_results_for_clickhouse( pr_info, job_report.test_results, From 3f0eacb47e80a5054f01374c00bd142303e679a7 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Sat, 19 Oct 2024 14:44:03 +0000 Subject: [PATCH 215/680] Automatic style fix --- tests/ci/build_download_helper.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/ci/build_download_helper.py b/tests/ci/build_download_helper.py index 47ea772b502..d7123564890 100644 --- a/tests/ci/build_download_helper.py +++ b/tests/ci/build_download_helper.py @@ -275,6 +275,8 @@ def download_fuzzers( check_name, reports_path, result_path, - lambda x: x.endswith(("test_basic_fuzzer", ".dict", ".options", "_seed_corpus.zip")), + lambda x: x.endswith( + ("test_basic_fuzzer", ".dict", ".options", "_seed_corpus.zip") + ), # lambda x: x.endswith(("_fuzzer", ".dict", ".options", "_seed_corpus.zip")), ) From daa32561c9a4f36352a9a65499568ee92d41eff9 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 16:32:36 +0000 Subject: [PATCH 216/680] test --- tests/ci/ci.py | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/ci/ci.py b/tests/ci/ci.py index 10431ce038f..e820f445e7a 100644 --- a/tests/ci/ci.py +++ b/tests/ci/ci.py @@ -1284,6 +1284,7 @@ def main() -> int: dump_to_file=True, ) print(f"Job report url: [{check_url}]") + print(job_report) prepared_events = prepare_tests_results_for_clickhouse( pr_info, job_report.test_results, From 8df6911a8375f072a0592b86c4015041d749f87c Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 17:13:47 +0000 Subject: [PATCH 217/680] fix --- tests/ci/libfuzzer_test_check.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index e0a985ac7b5..4a6f2875a4c 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -272,7 +272,7 @@ def main(): else: logging.info("Run failed") - results = process_results(reports_path) + results = process_results(result_path) success = results[1] == 0 and results[2] == 0 From 767daedd0d02adb36e3c9b8980a2d2effcb1f1ba Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 17:38:50 +0000 Subject: [PATCH 218/680] fix --- tests/ci/libfuzzer_test_check.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 4a6f2875a4c..e9f62c26cff 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -180,7 +180,7 @@ def process_results(result_path: Path): fails = 0 for file in result_path.glob("*.status"): fuzzer = file.stem - file_path = file.parent.with_stem(fuzzer) + file_path = file.parent / fuzzer file_path_unit = file_path.with_suffix(".unit") file_path_out = file_path.with_suffix(".out") status = read_status(file) From af8c50deeb3b93a753e8974960d06f57424f37fa Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 18:09:12 +0000 Subject: [PATCH 219/680] fix --- tests/ci/libfuzzer_test_check.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index e9f62c26cff..92f1336aa4b 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -148,6 +148,7 @@ def process_error(path: Path) -> list: with open(path, "r", encoding="utf-8") as file: for line in file: + line = line.rstrip("\n") if is_error: error_info.append(line) # match = re.search(TEST_UNIT_LINE, line) @@ -169,7 +170,7 @@ def read_status(status_path: Path): result = [] with open(status_path, "r", encoding="utf-8") as file: for line in file: - result.append(line) + result.append(line.rstrip("\n")) return result From 610630e20d2698a9ece08ec3d5bc94ec6b8ed735 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 18:51:25 +0000 Subject: [PATCH 220/680] fix --- tests/ci/libfuzzer_test_check.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 92f1336aa4b..6005a3bdc47 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -191,7 +191,7 @@ def process_results(result_path: Path): timeouts += 1 else: fails += 1 - result = TestResult(fuzzer, status[0], status[2]) + result = TestResult(fuzzer, status[0], float(status[2])) if file_path_unit.exists: result.set_raw_logs("\n".join(process_error(file_path_out))) test_results.append(result) From 8c14c33e5c1a2765049c4a9f21e2f1e1f671fc16 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 19:17:13 +0000 Subject: [PATCH 221/680] test --- tests/ci/build_download_helper.py | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/tests/ci/build_download_helper.py b/tests/ci/build_download_helper.py index d7123564890..8482abb26e0 100644 --- a/tests/ci/build_download_helper.py +++ b/tests/ci/build_download_helper.py @@ -275,8 +275,5 @@ def download_fuzzers( check_name, reports_path, result_path, - lambda x: x.endswith( - ("test_basic_fuzzer", ".dict", ".options", "_seed_corpus.zip") - ), - # lambda x: x.endswith(("_fuzzer", ".dict", ".options", "_seed_corpus.zip")), + lambda x: x.endswith(("_fuzzer", ".dict", ".options", "_seed_corpus.zip")), ) From 0a1f24e364a2e22e7235472dd5ef9d2f47fddc87 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 21:51:59 +0000 Subject: [PATCH 222/680] fix --- tests/ci/libfuzzer_test_check.py | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 6005a3bdc47..33b598ef0a6 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -185,15 +185,19 @@ def process_results(result_path: Path): file_path_unit = file_path.with_suffix(".unit") file_path_out = file_path.with_suffix(".out") status = read_status(file) + result = TestResult(fuzzer, status[0], float(status[2])) if status[0] == "OK": oks += 1 elif status[0] == "Timeout": timeouts += 1 + if file_path_out.exists(): + result.set_log_files([file_path_out]) else: fails += 1 - result = TestResult(fuzzer, status[0], float(status[2])) - if file_path_unit.exists: - result.set_raw_logs("\n".join(process_error(file_path_out))) + if file_path_out.exists(): + result.set_raw_logs("\n".join(process_error(file_path_out))) + if file_path_unit.exists: + result.set_log_files([file_path_unit]) test_results.append(result) return [oks, timeouts, fails, test_results] From ee989751aa1ef8c0de1352c4257ccbf09b3afbf8 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sat, 19 Oct 2024 23:46:02 +0000 Subject: [PATCH 223/680] fix --- tests/ci/libfuzzer_test_check.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 33b598ef0a6..703ff861eb7 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -191,13 +191,13 @@ def process_results(result_path: Path): elif status[0] == "Timeout": timeouts += 1 if file_path_out.exists(): - result.set_log_files([file_path_out]) + result.set_log_files([str(file_path_out)]) else: fails += 1 if file_path_out.exists(): result.set_raw_logs("\n".join(process_error(file_path_out))) if file_path_unit.exists: - result.set_log_files([file_path_unit]) + result.set_log_files([str(file_path_unit)]) test_results.append(result) return [oks, timeouts, fails, test_results] From 7d81ecb1835e7818020ad795e63d20372b6bf9ce Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 20 Oct 2024 02:16:50 +0200 Subject: [PATCH 224/680] Parallel compression --- src/Common/CurrentMetrics.cpp | 4 + src/Compression/CompressedWriteBuffer.cpp | 1 - src/Compression/ICompressionCodec.cpp | 11 ++ .../ParallelCompressedWriteBuffer.cpp | 118 ++++++++++++++++++ .../ParallelCompressedWriteBuffer.h | 87 +++++++++++++ 5 files changed, 220 insertions(+), 1 deletion(-) create mode 100644 src/Compression/ParallelCompressedWriteBuffer.cpp create mode 100644 src/Compression/ParallelCompressedWriteBuffer.h diff --git a/src/Common/CurrentMetrics.cpp b/src/Common/CurrentMetrics.cpp index bd62e7e8aae..da3b5557dbf 100644 --- a/src/Common/CurrentMetrics.cpp +++ b/src/Common/CurrentMetrics.cpp @@ -41,6 +41,10 @@ M(PostgreSQLConnection, "Number of client connections using PostgreSQL protocol") \ M(OpenFileForRead, "Number of files open for reading") \ M(OpenFileForWrite, "Number of files open for writing") \ + M(Compressing, "Number of compress operations using internal compression codecs") \ + M(Decompressing, "Number of decompress operations using internal compression codecs") \ + M(ParallelCompressedWriteBufferThreads, "Number of threads in all instances of ParallelCompressedWriteBuffer - these threads are doing parallel compression and writing") \ + M(ParallelCompressedWriteBufferWait, "Number of threads in all instances of ParallelCompressedWriteBuffer that are currently waiting for buffer to become available for writing") \ M(TotalTemporaryFiles, "Number of temporary files created") \ M(TemporaryFilesForSort, "Number of temporary files created for external sorting") \ M(TemporaryFilesForAggregation, "Number of temporary files created for external aggregation") \ diff --git a/src/Compression/CompressedWriteBuffer.cpp b/src/Compression/CompressedWriteBuffer.cpp index c3acfcb7da6..b6dab2a190e 100644 --- a/src/Compression/CompressedWriteBuffer.cpp +++ b/src/Compression/CompressedWriteBuffer.cpp @@ -2,7 +2,6 @@ #include #include -#include #include #include diff --git a/src/Compression/ICompressionCodec.cpp b/src/Compression/ICompressionCodec.cpp index 418667a3a8f..a31d0485982 100644 --- a/src/Compression/ICompressionCodec.cpp +++ b/src/Compression/ICompressionCodec.cpp @@ -5,11 +5,18 @@ #include #include #include +#include #include #include #include +namespace CurrentMetrics +{ + extern const Metric Compressing; + extern const Metric Decompressing; +} + namespace DB { @@ -80,6 +87,8 @@ UInt32 ICompressionCodec::compress(const char * source, UInt32 source_size, char { assert(source != nullptr && dest != nullptr); + CurrentMetrics::Increment metric_increment(CurrentMetrics::Compressing); + dest[0] = getMethodByte(); UInt8 header_size = getHeaderSize(); /// Write data from header_size @@ -93,6 +102,8 @@ UInt32 ICompressionCodec::decompress(const char * source, UInt32 source_size, ch { assert(source != nullptr && dest != nullptr); + CurrentMetrics::Increment metric_increment(CurrentMetrics::Decompressing); + UInt8 header_size = getHeaderSize(); if (source_size < header_size) throw Exception(decompression_error_code, diff --git a/src/Compression/ParallelCompressedWriteBuffer.cpp b/src/Compression/ParallelCompressedWriteBuffer.cpp new file mode 100644 index 00000000000..270c331e4df --- /dev/null +++ b/src/Compression/ParallelCompressedWriteBuffer.cpp @@ -0,0 +1,118 @@ +#include +#include + +#include +#include + +#include +#include +#include +#include +#include + +#include + + +namespace CurrentMetrics +{ + extern const Metric ParallelCompressedWriteBufferThreads; + extern const Metric ParallelCompressedWriteBufferWait; +} + +namespace DB +{ + +ParallelCompressedWriteBuffer::ParallelCompressedWriteBuffer( + WriteBuffer & out_, + CompressionCodecPtr codec_, + size_t buf_size_, + size_t num_threads_, + ThreadPool & pool_) + : WriteBuffer(nullptr, 0), out(out_), codec(codec_), buf_size(buf_size_), num_threads(num_threads_), pool(pool_) +{ + buffers.emplace_back(buf_size); + current_buffer = buffers.begin(); + BufferBase::set(current_buffer->uncompressed.data(), buf_size, 0); +} + +void ParallelCompressedWriteBuffer::nextImpl() +{ + if (!offset()) + return; + + std::unique_lock lock(mutex); + + /// The buffer will be compressed and processed in the thread. + current_buffer->busy = true; + pool.trySchedule([this, my_current_buffer = current_buffer, thread_group = CurrentThread::getGroup()] + { + SCOPE_EXIT_SAFE( + if (thread_group) + CurrentThread::detachFromGroupIfNotDetached(); + ); + + if (thread_group) + CurrentThread::attachToGroupIfDetached(thread_group); + setThreadName("ParallelCompres"); + + compress(my_current_buffer); + }); + + const BufferPair * previous_buffer = &*current_buffer; + ++current_buffer; + if (current_buffer == buffers.end()) + { + if (buffers.size() < num_threads) + { + /// If we didn't use all num_threads buffers yet, create a new one. + current_buffer = buffers.emplace(current_buffer, buf_size); + } + else + { + /// Otherwise, wrap around to the first buffer in the list. + current_buffer = buffers.begin(); + } + } + + /// Wait while the buffer becomes not busy + { + CurrentMetrics::Increment metric_increment(CurrentMetrics::ParallelCompressedWriteBufferWait); + cond.wait(lock, [&]{ return !current_buffer->busy; }); + } + + /// Now this buffer can be used. + current_buffer->previous = previous_buffer; + BufferBase::set(current_buffer->uncompressed.data(), buf_size, 0); +} + +void ParallelCompressedWriteBuffer::compress(Iterator buffer) +{ + CurrentMetrics::Increment metric_increment(CurrentMetrics::ParallelCompressedWriteBufferThreads); + + chassert(offset() <= INT_MAX); + UInt32 decompressed_size = static_cast(offset()); + UInt32 compressed_reserve_size = codec->getCompressedReserveSize(decompressed_size); + + buffer->compressed.resize(compressed_reserve_size); + UInt32 compressed_size = codec->compress(working_buffer.begin(), decompressed_size, buffer->compressed.data()); + + CityHash_v1_0_2::uint128 checksum = CityHash_v1_0_2::CityHash128(buffer->compressed.data(), compressed_size); + + /// Wait while all previous buffers have been written. + { + CurrentMetrics::Increment metric_wait_increment(CurrentMetrics::ParallelCompressedWriteBufferWait); + std::unique_lock lock(mutex); + cond.wait(lock, [&]{ return !buffer->previous || !buffer->previous->busy; }); + } + + writeBinaryLittleEndian(checksum.low64, out); + writeBinaryLittleEndian(checksum.high64, out); + + out.write(buffer->compressed.data(), compressed_size); + + std::unique_lock lock(mutex); + buffer->busy = false; + cond.notify_all(); +} + +} diff --git a/src/Compression/ParallelCompressedWriteBuffer.h b/src/Compression/ParallelCompressedWriteBuffer.h new file mode 100644 index 00000000000..e824dcacb46 --- /dev/null +++ b/src/Compression/ParallelCompressedWriteBuffer.h @@ -0,0 +1,87 @@ +#pragma once + +#include +#include + +#include + +#include +#include +#include +#include +#include + + +namespace DB +{ + +/** Uses multi-buffering for parallel compression. + * When the buffer is filled, it will be compressed in the background, + * and a new buffer is created for the next input data. + */ +class ParallelCompressedWriteBuffer final : public WriteBuffer +{ +public: + explicit ParallelCompressedWriteBuffer( + WriteBuffer & out_, + CompressionCodecPtr codec_, + size_t buf_size_, + size_t num_threads_, + ThreadPool & pool_); + + ~ParallelCompressedWriteBuffer() override; + + /// The amount of compressed data + size_t getCompressedBytes() + { + nextIfAtEnd(); + return out.count(); + } + + /// How many uncompressed bytes were written to the buffer + size_t getUncompressedBytes() + { + return count(); + } + + /// How many bytes are in the buffer (not yet compressed) + size_t getRemainingBytes() + { + nextIfAtEnd(); + return offset(); + } + +private: + void nextImpl() override; + void finalizeImpl() override; + + WriteBuffer & out; + CompressionCodecPtr codec; + size_t buf_size; + size_t num_threads; + ThreadPool & pool; + + struct BufferPair + { + BufferPair(size_t input_size) + : uncompressed(input_size) + { + } + + Memory<> uncompressed; + PODArray compressed; + const BufferPair * previous = nullptr; + bool busy = false; + }; + + std::mutex mutex; + std::condition_variable cond; + std::list buffers; + + using Iterator = std::list::iterator; + Iterator current_buffer; + + void compress(Iterator buffer); +}; + +} From 5e433ea537d42aca8fa1076f7c054b1b3dc83854 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 20 Oct 2024 03:11:16 +0200 Subject: [PATCH 225/680] Parallel compression: development --- programs/compressor/Compressor.cpp | 40 ++++++++++++++++--- .../ParallelCompressedWriteBuffer.cpp | 38 +++++++++++++++--- .../ParallelCompressedWriteBuffer.h | 3 ++ 3 files changed, 69 insertions(+), 12 deletions(-) diff --git a/programs/compressor/Compressor.cpp b/programs/compressor/Compressor.cpp index 050bb495024..aac56fba94a 100644 --- a/programs/compressor/Compressor.cpp +++ b/programs/compressor/Compressor.cpp @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -17,6 +18,8 @@ #include #include #include +#include +#include #include @@ -29,6 +32,13 @@ namespace DB } } +namespace CurrentMetrics +{ + extern const Metric LocalThread; + extern const Metric LocalThreadActive; + extern const Metric LocalThreadScheduled; +} + namespace { @@ -77,12 +87,13 @@ int mainEntryClickHouseCompressor(int argc, char ** argv) ("decompress,d", "decompress") ("offset-in-compressed-file", po::value()->default_value(0ULL), "offset to the compressed block (i.e. physical file offset)") ("offset-in-decompressed-block", po::value()->default_value(0ULL), "offset to the decompressed block (i.e. virtual offset)") - ("block-size,b", po::value()->default_value(DBMS_DEFAULT_BUFFER_SIZE), "compress in blocks of specified size") + ("block-size,b", po::value()->default_value(DBMS_DEFAULT_BUFFER_SIZE), "compress in blocks of specified size") ("hc", "use LZ4HC instead of LZ4") ("zstd", "use ZSTD instead of LZ4") ("deflate_qpl", "use deflate_qpl instead of LZ4") ("codec", po::value>()->multitoken(), "use codecs combination instead of LZ4") ("level", po::value(), "compression level for codecs specified via flags") + ("threads", po::value()->default_value(1), "number of threads for parallel compression") ("none", "use no compression instead of LZ4") ("stat", "print block statistics of compressed data") ("stacktrace", "print stacktrace of exception") @@ -111,7 +122,8 @@ int mainEntryClickHouseCompressor(int argc, char ** argv) bool stat_mode = options.count("stat"); bool use_none = options.count("none"); print_stacktrace = options.count("stacktrace"); - unsigned block_size = options["block-size"].as(); + size_t block_size = options["block-size"].as(); + size_t num_threads = options["threads"].as(); std::vector codecs; if (options.count("codec")) codecs = options["codec"].as>(); @@ -119,6 +131,12 @@ int mainEntryClickHouseCompressor(int argc, char ** argv) if ((use_lz4hc || use_zstd || use_deflate_qpl || use_none) && !codecs.empty()) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Wrong options, codec flags like --zstd and --codec options are mutually exclusive"); + if (num_threads < 1) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Invalid value of `threads` parameter"); + + if (num_threads > 1 && decompress) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Parallel mode is only implemented for compression (not for decompression)"); + if (!codecs.empty() && options.count("level")) throw Exception(ErrorCodes::BAD_ARGUMENTS, "Wrong options, --level is not compatible with --codec list"); @@ -149,7 +167,6 @@ int mainEntryClickHouseCompressor(int argc, char ** argv) else codec = CompressionCodecFactory::instance().get(method_family, level); - std::unique_ptr rb; std::unique_ptr wb; @@ -190,9 +207,20 @@ int mainEntryClickHouseCompressor(int argc, char ** argv) else { /// Compression - CompressedWriteBuffer to(*wb, codec, block_size); - copyData(*rb, to); - to.finalize(); + + if (num_threads == 1) + { + CompressedWriteBuffer to(*wb, codec, block_size); + copyData(*rb, to); + to.finalize(); + } + else + { + ThreadPool pool(CurrentMetrics::LocalThread, CurrentMetrics::LocalThreadActive, CurrentMetrics::LocalThreadScheduled, num_threads); + ParallelCompressedWriteBuffer to(*wb, codec, block_size, num_threads, pool); + copyData(*rb, to); + to.finalize(); + } } } catch (...) diff --git a/src/Compression/ParallelCompressedWriteBuffer.cpp b/src/Compression/ParallelCompressedWriteBuffer.cpp index 270c331e4df..4ffb6056d18 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.cpp +++ b/src/Compression/ParallelCompressedWriteBuffer.cpp @@ -1,8 +1,9 @@ #include -#include +#include #include #include +#include #include #include @@ -30,6 +31,7 @@ ParallelCompressedWriteBuffer::ParallelCompressedWriteBuffer( ThreadPool & pool_) : WriteBuffer(nullptr, 0), out(out_), codec(codec_), buf_size(buf_size_), num_threads(num_threads_), pool(pool_) { + std::cerr << getThreadId() << " Create a new buffer 1\n"; buffers.emplace_back(buf_size); current_buffer = buffers.begin(); BufferBase::set(current_buffer->uncompressed.data(), buf_size, 0); @@ -44,6 +46,9 @@ void ParallelCompressedWriteBuffer::nextImpl() /// The buffer will be compressed and processed in the thread. current_buffer->busy = true; + current_buffer->sequence_num = current_sequence_num; + ++current_sequence_num; + current_buffer->uncompressed_size = offset(); pool.trySchedule([this, my_current_buffer = current_buffer, thread_group = CurrentThread::getGroup()] { SCOPE_EXIT_SAFE( @@ -65,15 +70,19 @@ void ParallelCompressedWriteBuffer::nextImpl() if (buffers.size() < num_threads) { /// If we didn't use all num_threads buffers yet, create a new one. + std::cerr << getThreadId() << " Create a new buffer " << (buffers.size() + 1) << "\n"; current_buffer = buffers.emplace(current_buffer, buf_size); } else { /// Otherwise, wrap around to the first buffer in the list. + std::cerr << getThreadId() << " Wrap around\n"; current_buffer = buffers.begin(); } } + if (current_buffer->busy) + std::cerr << getThreadId() << " Wait while the buffer " << current_buffer->sequence_num << " becomes not busy\n"; /// Wait while the buffer becomes not busy { CurrentMetrics::Increment metric_increment(CurrentMetrics::ParallelCompressedWriteBufferWait); @@ -85,26 +94,37 @@ void ParallelCompressedWriteBuffer::nextImpl() BufferBase::set(current_buffer->uncompressed.data(), buf_size, 0); } +void ParallelCompressedWriteBuffer::finalizeImpl() +{ + next(); + pool.wait(); +} + void ParallelCompressedWriteBuffer::compress(Iterator buffer) { + std::cerr << getThreadId() << " Compressing " << buffer->sequence_num << "...\n"; CurrentMetrics::Increment metric_increment(CurrentMetrics::ParallelCompressedWriteBufferThreads); - chassert(offset() <= INT_MAX); - UInt32 decompressed_size = static_cast(offset()); - UInt32 compressed_reserve_size = codec->getCompressedReserveSize(decompressed_size); + chassert(buffer->uncompressed_size <= INT_MAX); + UInt32 uncompressed_size = static_cast(buffer->uncompressed_size); + UInt32 compressed_reserve_size = codec->getCompressedReserveSize(uncompressed_size); buffer->compressed.resize(compressed_reserve_size); - UInt32 compressed_size = codec->compress(working_buffer.begin(), decompressed_size, buffer->compressed.data()); + UInt32 compressed_size = codec->compress(buffer->uncompressed.data(), uncompressed_size, buffer->compressed.data()); CityHash_v1_0_2::uint128 checksum = CityHash_v1_0_2::CityHash128(buffer->compressed.data(), compressed_size); + if (buffer->previous && buffer->previous->busy) + std::cerr << getThreadId() << " Compressed " << buffer->sequence_num << ", waiting for prev buffer to be written...\n"; /// Wait while all previous buffers have been written. { CurrentMetrics::Increment metric_wait_increment(CurrentMetrics::ParallelCompressedWriteBufferWait); std::unique_lock lock(mutex); - cond.wait(lock, [&]{ return !buffer->previous || !buffer->previous->busy; }); + cond.wait(lock, [&]{ return !buffer->previous || !buffer->previous->busy || buffer->previous->sequence_num > buffer->sequence_num; }); } + std::cerr << getThreadId() << " Writing " << buffer->sequence_num << "...\n"; + writeBinaryLittleEndian(checksum.low64, out); writeBinaryLittleEndian(checksum.high64, out); @@ -115,4 +135,10 @@ void ParallelCompressedWriteBuffer::compress(Iterator buffer) cond.notify_all(); } +ParallelCompressedWriteBuffer::~ParallelCompressedWriteBuffer() +{ + if (!canceled) + finalize(); +} + } diff --git a/src/Compression/ParallelCompressedWriteBuffer.h b/src/Compression/ParallelCompressedWriteBuffer.h index e824dcacb46..ade49837f6b 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.h +++ b/src/Compression/ParallelCompressedWriteBuffer.h @@ -69,8 +69,10 @@ private: } Memory<> uncompressed; + size_t uncompressed_size = 0; PODArray compressed; const BufferPair * previous = nullptr; + size_t sequence_num = 0; bool busy = false; }; @@ -80,6 +82,7 @@ private: using Iterator = std::list::iterator; Iterator current_buffer; + size_t current_sequence_num = 0; void compress(Iterator buffer); }; From 7229ffd507db02a2b4c2a468ce1ac3dfcff26901 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 20 Oct 2024 03:15:10 +0200 Subject: [PATCH 226/680] Parallel compression: development --- src/Compression/ParallelCompressedWriteBuffer.cpp | 12 ------------ 1 file changed, 12 deletions(-) diff --git a/src/Compression/ParallelCompressedWriteBuffer.cpp b/src/Compression/ParallelCompressedWriteBuffer.cpp index 4ffb6056d18..30eaba33570 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.cpp +++ b/src/Compression/ParallelCompressedWriteBuffer.cpp @@ -1,9 +1,7 @@ #include -#include #include #include -#include #include #include @@ -31,7 +29,6 @@ ParallelCompressedWriteBuffer::ParallelCompressedWriteBuffer( ThreadPool & pool_) : WriteBuffer(nullptr, 0), out(out_), codec(codec_), buf_size(buf_size_), num_threads(num_threads_), pool(pool_) { - std::cerr << getThreadId() << " Create a new buffer 1\n"; buffers.emplace_back(buf_size); current_buffer = buffers.begin(); BufferBase::set(current_buffer->uncompressed.data(), buf_size, 0); @@ -70,19 +67,15 @@ void ParallelCompressedWriteBuffer::nextImpl() if (buffers.size() < num_threads) { /// If we didn't use all num_threads buffers yet, create a new one. - std::cerr << getThreadId() << " Create a new buffer " << (buffers.size() + 1) << "\n"; current_buffer = buffers.emplace(current_buffer, buf_size); } else { /// Otherwise, wrap around to the first buffer in the list. - std::cerr << getThreadId() << " Wrap around\n"; current_buffer = buffers.begin(); } } - if (current_buffer->busy) - std::cerr << getThreadId() << " Wait while the buffer " << current_buffer->sequence_num << " becomes not busy\n"; /// Wait while the buffer becomes not busy { CurrentMetrics::Increment metric_increment(CurrentMetrics::ParallelCompressedWriteBufferWait); @@ -102,7 +95,6 @@ void ParallelCompressedWriteBuffer::finalizeImpl() void ParallelCompressedWriteBuffer::compress(Iterator buffer) { - std::cerr << getThreadId() << " Compressing " << buffer->sequence_num << "...\n"; CurrentMetrics::Increment metric_increment(CurrentMetrics::ParallelCompressedWriteBufferThreads); chassert(buffer->uncompressed_size <= INT_MAX); @@ -114,8 +106,6 @@ void ParallelCompressedWriteBuffer::compress(Iterator buffer) CityHash_v1_0_2::uint128 checksum = CityHash_v1_0_2::CityHash128(buffer->compressed.data(), compressed_size); - if (buffer->previous && buffer->previous->busy) - std::cerr << getThreadId() << " Compressed " << buffer->sequence_num << ", waiting for prev buffer to be written...\n"; /// Wait while all previous buffers have been written. { CurrentMetrics::Increment metric_wait_increment(CurrentMetrics::ParallelCompressedWriteBufferWait); @@ -123,8 +113,6 @@ void ParallelCompressedWriteBuffer::compress(Iterator buffer) cond.wait(lock, [&]{ return !buffer->previous || !buffer->previous->busy || buffer->previous->sequence_num > buffer->sequence_num; }); } - std::cerr << getThreadId() << " Writing " << buffer->sequence_num << "...\n"; - writeBinaryLittleEndian(checksum.low64, out); writeBinaryLittleEndian(checksum.high64, out); From 66024821cf591d790ba1017b25dc4ebd75e8ea41 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 20 Oct 2024 03:23:07 +0200 Subject: [PATCH 227/680] Parallel compression: development --- src/Compression/ParallelCompressedWriteBuffer.cpp | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/Compression/ParallelCompressedWriteBuffer.cpp b/src/Compression/ParallelCompressedWriteBuffer.cpp index 30eaba33570..954fae242e4 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.cpp +++ b/src/Compression/ParallelCompressedWriteBuffer.cpp @@ -77,6 +77,7 @@ void ParallelCompressedWriteBuffer::nextImpl() } /// Wait while the buffer becomes not busy + if (current_buffer->busy) { CurrentMetrics::Increment metric_increment(CurrentMetrics::ParallelCompressedWriteBufferWait); cond.wait(lock, [&]{ return !current_buffer->busy; }); @@ -107,10 +108,11 @@ void ParallelCompressedWriteBuffer::compress(Iterator buffer) CityHash_v1_0_2::uint128 checksum = CityHash_v1_0_2::CityHash128(buffer->compressed.data(), compressed_size); /// Wait while all previous buffers have been written. + if (buffer->previous) { CurrentMetrics::Increment metric_wait_increment(CurrentMetrics::ParallelCompressedWriteBufferWait); std::unique_lock lock(mutex); - cond.wait(lock, [&]{ return !buffer->previous || !buffer->previous->busy || buffer->previous->sequence_num > buffer->sequence_num; }); + cond.wait(lock, [&]{ return !buffer->previous->busy || buffer->previous->sequence_num > buffer->sequence_num; }); } writeBinaryLittleEndian(checksum.low64, out); From d6e0da177744a9890ff269aad9168725f6a717c3 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 20 Oct 2024 03:33:42 +0200 Subject: [PATCH 228/680] Less memcpy --- .../ParallelCompressedWriteBuffer.cpp | 54 +++++++++++++++---- 1 file changed, 43 insertions(+), 11 deletions(-) diff --git a/src/Compression/ParallelCompressedWriteBuffer.cpp b/src/Compression/ParallelCompressedWriteBuffer.cpp index 954fae242e4..1041a14979e 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.cpp +++ b/src/Compression/ParallelCompressedWriteBuffer.cpp @@ -102,23 +102,55 @@ void ParallelCompressedWriteBuffer::compress(Iterator buffer) UInt32 uncompressed_size = static_cast(buffer->uncompressed_size); UInt32 compressed_reserve_size = codec->getCompressedReserveSize(uncompressed_size); - buffer->compressed.resize(compressed_reserve_size); - UInt32 compressed_size = codec->compress(buffer->uncompressed.data(), uncompressed_size, buffer->compressed.data()); + /// If all previous buffers have been written, + /// and if the output buffer has the required capacity, + /// we can compress data directly into the output buffer. + size_t required_out_capacity = compressed_reserve_size + sizeof(CityHash_v1_0_2::uint128); + bool can_write_directly = false; - CityHash_v1_0_2::uint128 checksum = CityHash_v1_0_2::CityHash128(buffer->compressed.data(), compressed_size); - - /// Wait while all previous buffers have been written. - if (buffer->previous) + if (!buffer->previous) + { + can_write_directly = out.available() >= required_out_capacity; + } + else { - CurrentMetrics::Increment metric_wait_increment(CurrentMetrics::ParallelCompressedWriteBufferWait); std::unique_lock lock(mutex); - cond.wait(lock, [&]{ return !buffer->previous->busy || buffer->previous->sequence_num > buffer->sequence_num; }); + can_write_directly = (!buffer->previous->busy || buffer->previous->sequence_num > buffer->sequence_num) + && out.available() >= required_out_capacity; } - writeBinaryLittleEndian(checksum.low64, out); - writeBinaryLittleEndian(checksum.high64, out); + if (can_write_directly) + { + char * out_compressed_ptr = out.position() + sizeof(CityHash_v1_0_2::uint128); + UInt32 compressed_size = codec->compress(working_buffer.begin(), uncompressed_size, out_compressed_ptr); - out.write(buffer->compressed.data(), compressed_size); + CityHash_v1_0_2::uint128 checksum = CityHash_v1_0_2::CityHash128(out_compressed_ptr, compressed_size); + + writeBinaryLittleEndian(checksum.low64, out); + writeBinaryLittleEndian(checksum.high64, out); + + out.position() += compressed_size; + } + else + { + buffer->compressed.resize(compressed_reserve_size); + UInt32 compressed_size = codec->compress(buffer->uncompressed.data(), uncompressed_size, buffer->compressed.data()); + + CityHash_v1_0_2::uint128 checksum = CityHash_v1_0_2::CityHash128(buffer->compressed.data(), compressed_size); + + /// Wait while all previous buffers have been written. + if (buffer->previous) + { + CurrentMetrics::Increment metric_wait_increment(CurrentMetrics::ParallelCompressedWriteBufferWait); + std::unique_lock lock(mutex); + cond.wait(lock, [&]{ return !buffer->previous->busy || buffer->previous->sequence_num > buffer->sequence_num; }); + } + + writeBinaryLittleEndian(checksum.low64, out); + writeBinaryLittleEndian(checksum.high64, out); + + out.write(buffer->compressed.data(), compressed_size); + } std::unique_lock lock(mutex); buffer->busy = false; From 157f7c0f471839942d9a34aa9fc3e5ea18939bed Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sun, 20 Oct 2024 11:39:27 +0000 Subject: [PATCH 229/680] fix --- tests/ci/libfuzzer_test_check.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 703ff861eb7..cba3b3410db 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -191,13 +191,13 @@ def process_results(result_path: Path): elif status[0] == "Timeout": timeouts += 1 if file_path_out.exists(): - result.set_log_files([str(file_path_out)]) + result.set_log_files(f"[{file_path_unit}]") else: fails += 1 if file_path_out.exists(): result.set_raw_logs("\n".join(process_error(file_path_out))) if file_path_unit.exists: - result.set_log_files([str(file_path_unit)]) + result.set_log_files(f"[{file_path_unit}]") test_results.append(result) return [oks, timeouts, fails, test_results] From 59c8fe9a240f9aee3344d09d5e16b59cbb58b581 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sun, 20 Oct 2024 12:38:28 +0000 Subject: [PATCH 230/680] fix --- tests/ci/libfuzzer_test_check.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index cba3b3410db..dbc2a2cc61b 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -191,13 +191,13 @@ def process_results(result_path: Path): elif status[0] == "Timeout": timeouts += 1 if file_path_out.exists(): - result.set_log_files(f"[{file_path_unit}]") + result.set_log_files(f"['{file_path_unit}']") else: fails += 1 if file_path_out.exists(): result.set_raw_logs("\n".join(process_error(file_path_out))) if file_path_unit.exists: - result.set_log_files(f"[{file_path_unit}]") + result.set_log_files(f"['{file_path_unit}']") test_results.append(result) return [oks, timeouts, fails, test_results] From 4b09224876c576af908ce95aa0ef295ce4820731 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sun, 20 Oct 2024 14:05:18 +0000 Subject: [PATCH 231/680] fix --- tests/ci/libfuzzer_test_check.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index dbc2a2cc61b..7012bd08418 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -191,7 +191,7 @@ def process_results(result_path: Path): elif status[0] == "Timeout": timeouts += 1 if file_path_out.exists(): - result.set_log_files(f"['{file_path_unit}']") + result.set_log_files(f"['{file_path_out}']") else: fails += 1 if file_path_out.exists(): From 567d113697a29e06efefea9e0e3089fd1114622d Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sun, 20 Oct 2024 15:19:22 +0000 Subject: [PATCH 232/680] fix --- tests/ci/libfuzzer_test_check.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 7012bd08418..6899083e837 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -196,7 +196,7 @@ def process_results(result_path: Path): fails += 1 if file_path_out.exists(): result.set_raw_logs("\n".join(process_error(file_path_out))) - if file_path_unit.exists: + if file_path_unit.exists(): result.set_log_files(f"['{file_path_unit}']") test_results.append(result) From b03d055aab13526bd25b3ea63017f02314cd8f67 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 20 Oct 2024 18:19:16 +0200 Subject: [PATCH 233/680] Fix clang-tidy --- src/Compression/ParallelCompressedWriteBuffer.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Compression/ParallelCompressedWriteBuffer.h b/src/Compression/ParallelCompressedWriteBuffer.h index ade49837f6b..4d1dfc79797 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.h +++ b/src/Compression/ParallelCompressedWriteBuffer.h @@ -63,7 +63,7 @@ private: struct BufferPair { - BufferPair(size_t input_size) + explicit BufferPair(size_t input_size) : uncompressed(input_size) { } From 5c3e9efdafa0d99328154715b6e2755dbcbcc0a5 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sun, 20 Oct 2024 18:23:34 +0000 Subject: [PATCH 234/680] fix, cleanup --- tests/ci/libfuzzer_test_check.py | 2 ++ tests/fuzz/runner.py | 58 -------------------------------- 2 files changed, 2 insertions(+), 58 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 6899083e837..d7e79cc26fe 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -198,6 +198,8 @@ def process_results(result_path: Path): result.set_raw_logs("\n".join(process_error(file_path_out))) if file_path_unit.exists(): result.set_log_files(f"['{file_path_unit}']") + elif file_path_out.exists(): + result.set_log_files(f"['{file_path_out}']") test_results.append(result) return [oks, timeouts, fails, test_results] diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index c23f4cbc31c..d3129a05b7c 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -32,46 +32,6 @@ class Stopwatch: self.start_time_str_value = self.start_time.strftime("%Y-%m-%d %H:%M:%S") -def report(source: str, reason: str, call_stack: list, test_unit: str): - logging.info("########### REPORT: %s %s %s", source, reason, test_unit) - logging.info("".join(call_stack)) - logging.info("########### END OF REPORT ###########") - - -# pylint: disable=unused-argument -def process_fuzzer_output(output: str): - pass - - -def process_error(error: str) -> list: - ERROR = r"^==\d+==\s?ERROR: (\S+): (.*)" - error_source = "" - error_reason = "" - test_unit = "" - TEST_UNIT_LINE = r"artifact_prefix='.*\/'; Test unit written to (.*)" - error_info = [] - is_error = False - - # pylint: disable=unused-variable - for line_num, line in enumerate(error.splitlines(), 1): - if is_error: - error_info.append(line) - match = re.search(TEST_UNIT_LINE, line) - if match: - test_unit = match.group(1) - continue - - match = re.search(ERROR, line) - if match: - error_info.append(line) - error_source = match.group(1) - error_reason = match.group(2) - is_error = True - - report(error_source, error_reason, error_info, test_unit) - return error_info - - def kill_fuzzer(fuzzer: str): with subprocess.Popen(["ps", "-A", "u"], stdout=subprocess.PIPE) as p: out, _ = p.communicate() @@ -91,10 +51,6 @@ def run_fuzzer(fuzzer: str, timeout: int): seed_corpus_dir = "" active_corpus_dir = f"corpus/{fuzzer}" - # new_corpus_dir = f"{OUTPUT}/corpus/{fuzzer}" - # if not os.path.exists(new_corpus_dir): - # os.makedirs(new_corpus_dir) - options_file = f"{fuzzer}.options" custom_libfuzzer_options = "" fuzzer_arguments = "" @@ -139,7 +95,6 @@ def run_fuzzer(fuzzer: str, timeout: int): cmd_line = ( f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {active_corpus_dir} {seed_corpus_dir}" ) - # cmd_line = f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {new_corpus_dir} {active_corpus_dir} {seed_corpus_dir}" cmd_line += f" -exact_artifact_path={exact_artifact_path}" @@ -169,34 +124,24 @@ def run_fuzzer(fuzzer: str, timeout: int): timeout=timeout, ) except subprocess.CalledProcessError as e: - # print("Command failed with error:", e) - logging.info("Stderr output: %s", e.stderr) with open(status_path, "w", encoding="utf-8") as status: status.write( f"FAIL\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) except subprocess.TimeoutExpired as e: - logging.info("Timeout for %s", cmd_line) kill_fuzzer(fuzzer) sleep(10) - process_fuzzer_output(e.stderr) with open(status_path, "w", encoding="utf-8") as status: status.write( f"Timeout\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) - os.remove(out_path) else: - process_fuzzer_output(result.stderr) with open(status_path, "w", encoding="utf-8") as status: status.write( f"OK\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) os.remove(out_path) - # s3.upload_build_directory_to_s3( - # Path(new_corpus_dir), f"fuzzer/corpus/{fuzzer}", False - # ) - def main(): logging.basicConfig(level=logging.INFO) @@ -216,9 +161,6 @@ def main(): subprocess.check_call(f"ls -al {OUTPUT}", shell=True) - # ch_helper = ClickHouseHelper() - # ch_helper.insert_events_into(db="default", table="checks", events=prepared_results) - if __name__ == "__main__": main() From f2b741202d432dce239302363bf81a607d6b5344 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sun, 20 Oct 2024 18:38:35 +0000 Subject: [PATCH 235/680] rename to clickhouse_fuzzer, fix --- tests/fuzz/build.sh | 3 +++ tests/fuzz/runner.py | 6 +++--- 2 files changed, 6 insertions(+), 3 deletions(-) diff --git a/tests/fuzz/build.sh b/tests/fuzz/build.sh index 12f41f6e079..f60336e6b53 100755 --- a/tests/fuzz/build.sh +++ b/tests/fuzz/build.sh @@ -1,5 +1,8 @@ #!/bin/bash -eu +# rename clickhouse +mv $OUT/clickhouse $OUT/clickhouse_fuzzer + # copy fuzzer options and dictionaries cp $SRC/tests/fuzz/*.dict $OUT/ cp $SRC/tests/fuzz/*.options $OUT/ diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index d3129a05b7c..c84e34ffdbd 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -113,7 +113,7 @@ def run_fuzzer(fuzzer: str, timeout: int): stopwatch = Stopwatch() try: with open(out_path, "wb") as out: - result = subprocess.run( + subprocess.run( cmd_line, stderr=out, stdout=subprocess.DEVNULL, @@ -123,12 +123,12 @@ def run_fuzzer(fuzzer: str, timeout: int): errors="replace", timeout=timeout, ) - except subprocess.CalledProcessError as e: + except subprocess.CalledProcessError: with open(status_path, "w", encoding="utf-8") as status: status.write( f"FAIL\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) - except subprocess.TimeoutExpired as e: + except subprocess.TimeoutExpired: kill_fuzzer(fuzzer) sleep(10) with open(status_path, "w", encoding="utf-8") as status: From a8c59df8d7da00440b4eb4e30c728c3744c43888 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Sun, 20 Oct 2024 15:38:28 -0400 Subject: [PATCH 236/680] trigger build --- src/DataTypes/fuzzers/CMakeLists.txt | 1 - 1 file changed, 1 deletion(-) diff --git a/src/DataTypes/fuzzers/CMakeLists.txt b/src/DataTypes/fuzzers/CMakeLists.txt index 8940586fc70..8dedd3470e2 100644 --- a/src/DataTypes/fuzzers/CMakeLists.txt +++ b/src/DataTypes/fuzzers/CMakeLists.txt @@ -1,3 +1,2 @@ clickhouse_add_executable(data_type_deserialization_fuzzer data_type_deserialization_fuzzer.cpp ${SRCS}) - target_link_libraries(data_type_deserialization_fuzzer PRIVATE clickhouse_aggregate_functions dbms) From 8f038e2e1cd3c6635f328c7bbece1573c225dfb3 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 20 Oct 2024 23:08:22 +0200 Subject: [PATCH 237/680] Preparation --- .../MergeTree/MergeTreeDataPartWriterOnDisk.cpp | 10 +++++----- src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h | 5 +++-- 2 files changed, 8 insertions(+), 7 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp index 58a67fc4ba2..a006e2da368 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp @@ -34,7 +34,7 @@ void MergeTreeDataPartWriterOnDisk::Stream::preFinalize() /// Also the order is important compressed_hashing.finalize(); - compressor.finalize(); + compressor->finalize(); plain_hashing.finalize(); if constexpr (!only_plain_file) @@ -92,8 +92,8 @@ MergeTreeDataPartWriterOnDisk::Stream::Stream( marks_file_extension{marks_file_extension_}, plain_file(data_part_storage->writeFile(data_path_ + data_file_extension, max_compress_block_size_, query_write_settings)), plain_hashing(*plain_file), - compressor(plain_hashing, compression_codec_, max_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size), - compressed_hashing(compressor), + compressor(std::make_unique(plain_hashing, compression_codec_, max_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size)), + compressed_hashing(*compressor), marks_file(data_part_storage->writeFile(marks_path_ + marks_file_extension, 4096, query_write_settings)), marks_hashing(*marks_file), marks_compressor(marks_hashing, marks_compression_codec_, marks_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size), @@ -115,8 +115,8 @@ MergeTreeDataPartWriterOnDisk::Stream::Stream( data_file_extension{data_file_extension_}, plain_file(data_part_storage->writeFile(data_path_ + data_file_extension, max_compress_block_size_, query_write_settings)), plain_hashing(*plain_file), - compressor(plain_hashing, compression_codec_, max_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size), - compressed_hashing(compressor), + compressor(std::make_unique(plain_hashing, compression_codec_, max_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size)), + compressed_hashing(*compressor), compress_marks(false) { } diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h index 8d84442981e..3b6687dff99 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h @@ -44,7 +44,7 @@ public: /// Helper class, which holds chain of buffers to write data file with marks. /// It is used to write: one column, skip index or all columns (in compact format). - template + template struct Stream { Stream( @@ -76,7 +76,8 @@ public: /// compressed_hashing -> compressor -> plain_hashing -> plain_file std::unique_ptr plain_file; HashingWriteBuffer plain_hashing; - CompressedWriteBuffer compressor; + /// This could be either CompressedWriteBuffer or ParallelCompressedWriteBuffer + std::unique_ptr compressor; HashingWriteBuffer compressed_hashing; /// marks_compressed_hashing -> marks_compressor -> marks_hashing -> marks_file From 2995cf9d10da2814e9bf215fd1a8bca9f1ab5438 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sun, 20 Oct 2024 21:24:10 +0000 Subject: [PATCH 238/680] fix --- tests/fuzz/runner.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index c84e34ffdbd..9eac0755d78 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -51,6 +51,8 @@ def run_fuzzer(fuzzer: str, timeout: int): seed_corpus_dir = "" active_corpus_dir = f"corpus/{fuzzer}" + if not os.path.exists(active_corpus_dir): + os.makedirs(active_corpus_dir) options_file = f"{fuzzer}.options" custom_libfuzzer_options = "" fuzzer_arguments = "" From 1236422559c5fd957f99e65f08e25d7a9806190e Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 20 Oct 2024 23:28:23 +0200 Subject: [PATCH 239/680] Templates are shit --- .../MergeTreeDataPartWriterOnDisk.cpp | 83 +++++++++---------- .../MergeTree/MergeTreeDataPartWriterOnDisk.h | 12 ++- .../MergeTree/MergeTreeDataPartWriterWide.cpp | 4 +- 3 files changed, 44 insertions(+), 55 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp index a006e2da368..c250726aba1 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp @@ -25,8 +25,7 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; } -template -void MergeTreeDataPartWriterOnDisk::Stream::preFinalize() +void MergeTreeDataPartWriterOnDisk::Stream::preFinalize() { /// Here the main goal is to do preFinalize calls for plain_file and marks_file /// Before that all hashing and compression buffers have to be finalized @@ -37,45 +36,42 @@ void MergeTreeDataPartWriterOnDisk::Stream::preFinalize() compressor->finalize(); plain_hashing.finalize(); - if constexpr (!only_plain_file) + if (marks_hashing) { if (compress_marks) { - marks_compressed_hashing.finalize(); - marks_compressor.finalize(); + marks_compressed_hashing->finalize(); + marks_compressor->finalize(); } - marks_hashing.finalize(); + marks_hashing->finalize(); } plain_file->preFinalize(); - if constexpr (!only_plain_file) + if (marks_file) marks_file->preFinalize(); is_prefinalized = true; } -template -void MergeTreeDataPartWriterOnDisk::Stream::finalize() +void MergeTreeDataPartWriterOnDisk::Stream::finalize() { if (!is_prefinalized) preFinalize(); plain_file->finalize(); - if constexpr (!only_plain_file) + if (marks_file) marks_file->finalize(); } -template -void MergeTreeDataPartWriterOnDisk::Stream::sync() const +void MergeTreeDataPartWriterOnDisk::Stream::sync() const { plain_file->sync(); - if constexpr (!only_plain_file) + if (marks_file) marks_file->sync(); } -template<> -MergeTreeDataPartWriterOnDisk::Stream::Stream( +MergeTreeDataPartWriterOnDisk::Stream::Stream( const String & escaped_column_name_, const MutableDataPartStoragePtr & data_part_storage, const String & data_path_, @@ -94,16 +90,15 @@ MergeTreeDataPartWriterOnDisk::Stream::Stream( plain_hashing(*plain_file), compressor(std::make_unique(plain_hashing, compression_codec_, max_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size)), compressed_hashing(*compressor), - marks_file(data_part_storage->writeFile(marks_path_ + marks_file_extension, 4096, query_write_settings)), - marks_hashing(*marks_file), - marks_compressor(marks_hashing, marks_compression_codec_, marks_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size), - marks_compressed_hashing(marks_compressor), compress_marks(MarkType(marks_file_extension).compressed) { + marks_file = data_part_storage->writeFile(marks_path_ + marks_file_extension, 4096, query_write_settings); + marks_hashing.emplace(*marks_file); + marks_compressor.emplace(*marks_hashing, marks_compression_codec_, marks_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size); + marks_compressed_hashing.emplace(*marks_compressor); } -template<> -MergeTreeDataPartWriterOnDisk::Stream::Stream( +MergeTreeDataPartWriterOnDisk::Stream::Stream( const String & escaped_column_name_, const MutableDataPartStoragePtr & data_part_storage, const String & data_path_, @@ -121,8 +116,7 @@ MergeTreeDataPartWriterOnDisk::Stream::Stream( { } -template -void MergeTreeDataPartWriterOnDisk::Stream::addToChecksums(MergeTreeData::DataPart::Checksums & checksums) +void MergeTreeDataPartWriterOnDisk::Stream::addToChecksums(MergeTreeData::DataPart::Checksums & checksums) { String name = escaped_column_name; @@ -132,17 +126,17 @@ void MergeTreeDataPartWriterOnDisk::Stream::addToChecksums(Merg checksums.files[name + data_file_extension].file_size = plain_hashing.count(); checksums.files[name + data_file_extension].file_hash = plain_hashing.getHash(); - if constexpr (!only_plain_file) + if (marks_hashing) { if (compress_marks) { checksums.files[name + marks_file_extension].is_compressed = true; - checksums.files[name + marks_file_extension].uncompressed_size = marks_compressed_hashing.count(); - checksums.files[name + marks_file_extension].uncompressed_hash = marks_compressed_hashing.getHash(); + checksums.files[name + marks_file_extension].uncompressed_size = marks_compressed_hashing->count(); + checksums.files[name + marks_file_extension].uncompressed_hash = marks_compressed_hashing->getHash(); } - checksums.files[name + marks_file_extension].file_size = marks_hashing.count(); - checksums.files[name + marks_file_extension].file_hash = marks_hashing.getHash(); + checksums.files[name + marks_file_extension].file_size = marks_hashing->count(); + checksums.files[name + marks_file_extension].file_hash = marks_hashing->getHash(); } } @@ -276,12 +270,12 @@ void MergeTreeDataPartWriterOnDisk::initStatistics() for (const auto & stat_ptr : stats) { String stats_name = stat_ptr->getFileName(); - stats_streams.emplace_back(std::make_unique>( - stats_name, - data_part_storage, - stats_name, STATS_FILE_SUFFIX, - default_codec, settings.max_compress_block_size, - settings.query_write_settings)); + stats_streams.emplace_back(std::make_unique( + stats_name, + data_part_storage, + stats_name, STATS_FILE_SUFFIX, + default_codec, settings.max_compress_block_size, + settings.query_write_settings)); } } @@ -298,14 +292,14 @@ void MergeTreeDataPartWriterOnDisk::initSkipIndices() { String stream_name = skip_index->getFileName(); skip_indices_streams.emplace_back( - std::make_unique>( - stream_name, - data_part_storage, - stream_name, skip_index->getSerializedFileExtension(), - stream_name, marks_file_extension, - default_codec, settings.max_compress_block_size, - marks_compression_codec, settings.marks_compress_block_size, - settings.query_write_settings)); + std::make_unique( + stream_name, + data_part_storage, + stream_name, skip_index->getSerializedFileExtension(), + stream_name, marks_file_extension, + default_codec, settings.max_compress_block_size, + marks_compression_codec, settings.marks_compress_block_size, + settings.query_write_settings)); GinIndexStorePtr store = nullptr; if (typeid_cast(&*skip_index) != nullptr) @@ -381,7 +375,7 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializeSkipIndices(const Block { const auto index_helper = skip_indices[i]; auto & stream = *skip_indices_streams[i]; - WriteBuffer & marks_out = stream.compress_marks ? stream.marks_compressed_hashing : stream.marks_hashing; + WriteBuffer & marks_out = stream.compress_marks ? *stream.marks_compressed_hashing : *stream.marks_hashing; GinIndexStorePtr store; if (typeid_cast(&*index_helper) != nullptr) @@ -564,7 +558,4 @@ Names MergeTreeDataPartWriterOnDisk::getSkipIndicesColumns() const return Names(skip_indexes_column_names_set.begin(), skip_indexes_column_names_set.end()); } -template struct MergeTreeDataPartWriterOnDisk::Stream; -template struct MergeTreeDataPartWriterOnDisk::Stream; - } diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h index 3b6687dff99..0d80333368d 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h @@ -44,7 +44,6 @@ public: /// Helper class, which holds chain of buffers to write data file with marks. /// It is used to write: one column, skip index or all columns (in compact format). - template struct Stream { Stream( @@ -82,9 +81,9 @@ public: /// marks_compressed_hashing -> marks_compressor -> marks_hashing -> marks_file std::unique_ptr marks_file; - std::conditional_t marks_hashing; - std::conditional_t marks_compressor; - std::conditional_t marks_compressed_hashing; + std::optional marks_hashing; + std::optional marks_compressor; + std::optional marks_compressed_hashing; bool compress_marks; bool is_prefinalized = false; @@ -98,8 +97,7 @@ public: void addToChecksums(MergeTreeDataPartChecksums & checksums); }; - using StreamPtr = std::unique_ptr>; - using StatisticStreamPtr = std::unique_ptr>; + using StreamPtr = std::unique_ptr; MergeTreeDataPartWriterOnDisk( const String & data_part_name_, @@ -157,7 +155,7 @@ protected: const MergeTreeIndices skip_indices; const ColumnsStatistics stats; - std::vector stats_streams; + std::vector stats_streams; const String marks_file_extension; const CompressionCodecPtr default_codec; diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp index 459ddc1ca79..d1d4aa4f5b0 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp @@ -187,7 +187,7 @@ void MergeTreeDataPartWriterWide::addStreams( query_write_settings.use_adaptive_write_buffer = settings.use_adaptive_write_buffer_for_dynamic_subcolumns && ISerialization::isDynamicSubcolumn(substream_path, substream_path.size()); query_write_settings.adaptive_write_buffer_initial_size = settings.adaptive_write_buffer_initial_size; - column_streams[stream_name] = std::make_unique>( + column_streams[stream_name] = std::make_unique( stream_name, data_part_storage, stream_name, DATA_FILE_EXTENSION, @@ -362,7 +362,7 @@ void MergeTreeDataPartWriterWide::writeSingleMark( void MergeTreeDataPartWriterWide::flushMarkToFile(const StreamNameAndMark & stream_with_mark, size_t rows_in_mark) { auto & stream = *column_streams[stream_with_mark.stream_name]; - WriteBuffer & marks_out = stream.compress_marks ? stream.marks_compressed_hashing : stream.marks_hashing; + WriteBuffer & marks_out = stream.compress_marks ? *stream.marks_compressed_hashing : *stream.marks_hashing; writeBinaryLittleEndian(stream_with_mark.mark.offset_in_compressed_file, marks_out); writeBinaryLittleEndian(stream_with_mark.mark.offset_in_decompressed_block, marks_out); From ab10830317d1a0d155aceec3f5285f299f661f02 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 21 Oct 2024 00:12:06 +0200 Subject: [PATCH 240/680] Preparation --- src/Common/CurrentMetrics.cpp | 3 + src/IO/WriteSettings.h | 2 + .../MergeTreeDataPartWriterOnDisk.cpp | 74 ++++++++++++++----- .../MergeTree/MergeTreeDataPartWriterOnDisk.h | 11 +-- .../MergeTree/MergeTreeDataPartWriterWide.cpp | 13 ++-- 5 files changed, 72 insertions(+), 31 deletions(-) diff --git a/src/Common/CurrentMetrics.cpp b/src/Common/CurrentMetrics.cpp index da3b5557dbf..c9737e2e846 100644 --- a/src/Common/CurrentMetrics.cpp +++ b/src/Common/CurrentMetrics.cpp @@ -103,6 +103,9 @@ M(IOThreads, "Number of threads in the IO thread pool.") \ M(IOThreadsActive, "Number of threads in the IO thread pool running a task.") \ M(IOThreadsScheduled, "Number of queued or active jobs in the IO thread pool.") \ + M(CompressionThread, "Number of threads in compression thread pools.") \ + M(CompressionThreadActive, "Number of threads in compression thread pools running a task.") \ + M(CompressionThreadScheduled, "Number of queued or active jobs in compression thread pools.") \ M(ThreadPoolRemoteFSReaderThreads, "Number of threads in the thread pool for remote_filesystem_read_method=threadpool.") \ M(ThreadPoolRemoteFSReaderThreadsActive, "Number of threads in the thread pool for remote_filesystem_read_method=threadpool running a task.") \ M(ThreadPoolRemoteFSReaderThreadsScheduled, "Number of queued or active jobs in the thread pool for remote_filesystem_read_method=threadpool.") \ diff --git a/src/IO/WriteSettings.h b/src/IO/WriteSettings.h index 94410f787f0..8016dede6ea 100644 --- a/src/IO/WriteSettings.h +++ b/src/IO/WriteSettings.h @@ -28,6 +28,8 @@ struct WriteSettings bool use_adaptive_write_buffer = false; size_t adaptive_write_buffer_initial_size = 16 * 1024; + size_t max_compression_threads = 1; + bool write_through_distributed_cache = false; DistributedCacheSettings distributed_cache_settings; diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp index c250726aba1..8047d1cf2e9 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp @@ -3,16 +3,26 @@ #include #include #include +#include #include + namespace ProfileEvents { -extern const Event MergeTreeDataWriterSkipIndicesCalculationMicroseconds; -extern const Event MergeTreeDataWriterStatisticsCalculationMicroseconds; + extern const Event MergeTreeDataWriterSkipIndicesCalculationMicroseconds; + extern const Event MergeTreeDataWriterStatisticsCalculationMicroseconds; +} + +namespace CurrentMetrics +{ + extern const Metric CompressionThread; + extern const Metric CompressionThreadActive; + extern const Metric CompressionThreadScheduled; } namespace DB { + namespace MergeTreeSetting { extern const MergeTreeSettingsUInt64 index_granularity; @@ -32,9 +42,9 @@ void MergeTreeDataPartWriterOnDisk::Stream::preFinalize() /// Otherwise some data might stuck in the buffers above plain_file and marks_file /// Also the order is important - compressed_hashing.finalize(); + compressed_hashing->finalize(); compressor->finalize(); - plain_hashing.finalize(); + plain_hashing->finalize(); if (marks_hashing) { @@ -86,12 +96,36 @@ MergeTreeDataPartWriterOnDisk::Stream::Stream( escaped_column_name(escaped_column_name_), data_file_extension{data_file_extension_}, marks_file_extension{marks_file_extension_}, - plain_file(data_part_storage->writeFile(data_path_ + data_file_extension, max_compress_block_size_, query_write_settings)), - plain_hashing(*plain_file), - compressor(std::make_unique(plain_hashing, compression_codec_, max_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size)), - compressed_hashing(*compressor), compress_marks(MarkType(marks_file_extension).compressed) { + plain_file = data_part_storage->writeFile(data_path_ + data_file_extension, max_compress_block_size_, query_write_settings); + plain_hashing.emplace(*plain_file); + + if (query_write_settings.max_compression_threads > 1) + { + compression_thread_pool.emplace( + CurrentMetrics::CompressionThread, CurrentMetrics::CompressionThreadActive, CurrentMetrics::CompressionThreadScheduled, + query_write_settings.max_compression_threads); + + compressor = std::make_unique( + *plain_hashing, + compression_codec_, + max_compress_block_size_, + query_write_settings.max_compression_threads, + *compression_thread_pool); + } + else + { + compressor = std::make_unique( + *plain_hashing, + compression_codec_, + max_compress_block_size_, + query_write_settings.use_adaptive_write_buffer, + query_write_settings.adaptive_write_buffer_initial_size); + } + + compressed_hashing.emplace(*compressor); + marks_file = data_part_storage->writeFile(marks_path_ + marks_file_extension, 4096, query_write_settings); marks_hashing.emplace(*marks_file); marks_compressor.emplace(*marks_hashing, marks_compression_codec_, marks_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size); @@ -110,7 +144,7 @@ MergeTreeDataPartWriterOnDisk::Stream::Stream( data_file_extension{data_file_extension_}, plain_file(data_part_storage->writeFile(data_path_ + data_file_extension, max_compress_block_size_, query_write_settings)), plain_hashing(*plain_file), - compressor(std::make_unique(plain_hashing, compression_codec_, max_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size)), + compressor(std::make_unique(*plain_hashing, compression_codec_, max_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size)), compressed_hashing(*compressor), compress_marks(false) { @@ -121,10 +155,10 @@ void MergeTreeDataPartWriterOnDisk::Stream::addToChecksums(MergeTreeData::DataPa String name = escaped_column_name; checksums.files[name + data_file_extension].is_compressed = true; - checksums.files[name + data_file_extension].uncompressed_size = compressed_hashing.count(); - checksums.files[name + data_file_extension].uncompressed_hash = compressed_hashing.getHash(); - checksums.files[name + data_file_extension].file_size = plain_hashing.count(); - checksums.files[name + data_file_extension].file_hash = plain_hashing.getHash(); + checksums.files[name + data_file_extension].uncompressed_size = compressed_hashing->count(); + checksums.files[name + data_file_extension].uncompressed_hash = compressed_hashing->getHash(); + checksums.files[name + data_file_extension].file_size = plain_hashing->count(); + checksums.files[name + data_file_extension].file_hash = plain_hashing->getHash(); if (marks_hashing) { @@ -391,7 +425,7 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializeSkipIndices(const Block { if (skip_index_accumulated_marks[i] == index_helper->index.granularity) { - skip_indices_aggregators[i]->getGranuleAndReset()->serializeBinary(stream.compressed_hashing); + skip_indices_aggregators[i]->getGranuleAndReset()->serializeBinary(*stream.compressed_hashing); skip_index_accumulated_marks[i] = 0; } @@ -399,11 +433,11 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializeSkipIndices(const Block { skip_indices_aggregators[i] = index_helper->createIndexAggregatorForPart(store, settings); - if (stream.compressed_hashing.offset() >= settings.min_compress_block_size) - stream.compressed_hashing.next(); + if (stream.compressed_hashing->offset() >= settings.min_compress_block_size) + stream.compressed_hashing->next(); - writeBinaryLittleEndian(stream.plain_hashing.count(), marks_out); - writeBinaryLittleEndian(stream.compressed_hashing.offset(), marks_out); + writeBinaryLittleEndian(stream.plain_hashing->count(), marks_out); + writeBinaryLittleEndian(stream.compressed_hashing->offset(), marks_out); /// Actually this numbers is redundant, but we have to store them /// to be compatible with the normal .mrk2 file format @@ -483,7 +517,7 @@ void MergeTreeDataPartWriterOnDisk::fillSkipIndicesChecksums(MergeTreeData::Data { auto & stream = *skip_indices_streams[i]; if (!skip_indices_aggregators[i]->empty()) - skip_indices_aggregators[i]->getGranuleAndReset()->serializeBinary(stream.compressed_hashing); + skip_indices_aggregators[i]->getGranuleAndReset()->serializeBinary(*stream.compressed_hashing); /// Register additional files written only by the full-text index. Required because otherwise DROP TABLE complains about unknown /// files. Note that the provided actual checksums are bogus. The problem is that at this point the file writes happened already and @@ -523,7 +557,7 @@ void MergeTreeDataPartWriterOnDisk::fillStatisticsChecksums(MergeTreeData::DataP for (size_t i = 0; i < stats.size(); i++) { auto & stream = *stats_streams[i]; - stats[i]->serialize(stream.compressed_hashing); + stats[i]->serialize(*stream.compressed_hashing); stream.preFinalize(); stream.addToChecksums(checksums); } diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h index 0d80333368d..046571cb83f 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h @@ -27,7 +27,7 @@ struct Granule /// this granule can be continuation of the previous one. bool mark_on_start; /// if true: When this granule will be written to disk all rows for corresponding mark will - /// be wrtten. It doesn't mean that rows_to_write == index_granularity.getMarkRows(mark_number), + /// be written. It doesn't mean that rows_to_write == index_granularity.getMarkRows(mark_number), /// We may have a lot of small blocks between two marks and this may be the last one. bool is_complete; }; @@ -74,10 +74,10 @@ public: /// compressed_hashing -> compressor -> plain_hashing -> plain_file std::unique_ptr plain_file; - HashingWriteBuffer plain_hashing; + std::optional plain_hashing; /// This could be either CompressedWriteBuffer or ParallelCompressedWriteBuffer std::unique_ptr compressor; - HashingWriteBuffer compressed_hashing; + std::optional compressed_hashing; /// marks_compressed_hashing -> marks_compressor -> marks_hashing -> marks_file std::unique_ptr marks_file; @@ -88,10 +88,11 @@ public: bool is_prefinalized = false; + /// Thread pool for parallel compression. + std::optional compression_thread_pool; + void preFinalize(); - void finalize(); - void sync() const; void addToChecksums(MergeTreeDataPartChecksums & checksums); diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp index d1d4aa4f5b0..523e4c4a31d 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp @@ -9,6 +9,7 @@ #include #include + namespace DB { @@ -230,7 +231,7 @@ ISerialization::OutputStreamGetter MergeTreeDataPartWriterWide::createStreamGett if (is_offsets && offset_columns.contains(stream_name)) return nullptr; - return &column_streams.at(stream_name)->compressed_hashing; + return &column_streams.at(stream_name)->compressed_hashing.value(); }; } @@ -399,13 +400,13 @@ StreamsWithMarks MergeTreeDataPartWriterWide::getCurrentMarksForColumn( auto & stream = *column_streams[stream_name]; /// There could already be enough data to compress into the new block. - if (stream.compressed_hashing.offset() >= min_compress_block_size) - stream.compressed_hashing.next(); + if (stream.compressed_hashing->offset() >= min_compress_block_size) + stream.compressed_hashing->next(); StreamNameAndMark stream_with_mark; stream_with_mark.stream_name = stream_name; - stream_with_mark.mark.offset_in_compressed_file = stream.plain_hashing.count(); - stream_with_mark.mark.offset_in_decompressed_block = stream.compressed_hashing.offset(); + stream_with_mark.mark.offset_in_compressed_file = stream.plain_hashing->count(); + stream_with_mark.mark.offset_in_decompressed_block = stream.compressed_hashing->offset(); result.push_back(stream_with_mark); }, name_and_type.type, column_sample); @@ -438,7 +439,7 @@ void MergeTreeDataPartWriterWide::writeSingleGranule( if (is_offsets && offset_columns.contains(stream_name)) return; - column_streams.at(stream_name)->compressed_hashing.nextIfAtEnd(); + column_streams.at(stream_name)->compressed_hashing->nextIfAtEnd(); }, name_and_type.type, column.getPtr()); } From bb3bfa536ac54185fea01d23fa22d970a2cccf84 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 21 Oct 2024 00:35:01 +0200 Subject: [PATCH 241/680] Make it configurable --- src/IO/WriteSettings.h | 1 + .../MergeTree/IMergeTreeDataPartWriter.cpp | 22 ++++++++++++++++--- .../MergeTree/IMergedBlockOutputStream.cpp | 1 + .../MergeTree/IMergedBlockOutputStream.h | 1 + src/Storages/MergeTree/MergeTask.cpp | 9 +------- .../MergeTree/MergeTreeIOSettings.cpp | 2 ++ src/Storages/MergeTree/MergeTreeSettings.cpp | 3 +++ 7 files changed, 28 insertions(+), 11 deletions(-) diff --git a/src/IO/WriteSettings.h b/src/IO/WriteSettings.h index 8016dede6ea..4eeb01b5acc 100644 --- a/src/IO/WriteSettings.h +++ b/src/IO/WriteSettings.h @@ -4,6 +4,7 @@ #include #include + namespace DB { diff --git a/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp b/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp index 3d6366f9217..a9f188338e1 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp @@ -178,9 +178,24 @@ MergeTreeDataPartWriterPtr createMergeTreeDataPartWriter( const MergeTreeIndexGranularity & computed_index_granularity) { if (part_type == MergeTreeDataPartType::Compact) - return createMergeTreeDataPartCompactWriter(data_part_name_, logger_name_, serializations_, data_part_storage_, - index_granularity_info_, storage_settings_, columns_list, column_positions, metadata_snapshot, virtual_columns, indices_to_recalc, stats_to_recalc_, - marks_file_extension_, default_codec_, writer_settings, computed_index_granularity); + return createMergeTreeDataPartCompactWriter( + data_part_name_, + logger_name_, + serializations_, + data_part_storage_, + index_granularity_info_, + storage_settings_, + columns_list, + column_positions, + metadata_snapshot, + virtual_columns, + indices_to_recalc, + stats_to_recalc_, + marks_file_extension_, + default_codec_, + writer_settings, + computed_index_granularity); + if (part_type == MergeTreeDataPartType::Wide) return createMergeTreeDataPartWideWriter( data_part_name_, @@ -198,6 +213,7 @@ MergeTreeDataPartWriterPtr createMergeTreeDataPartWriter( default_codec_, writer_settings, computed_index_granularity); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown part type: {}", part_type.toString()); } diff --git a/src/Storages/MergeTree/IMergedBlockOutputStream.cpp b/src/Storages/MergeTree/IMergedBlockOutputStream.cpp index eb904a8e2ef..209b274ee6a 100644 --- a/src/Storages/MergeTree/IMergedBlockOutputStream.cpp +++ b/src/Storages/MergeTree/IMergedBlockOutputStream.cpp @@ -4,6 +4,7 @@ #include #include + namespace DB { diff --git a/src/Storages/MergeTree/IMergedBlockOutputStream.h b/src/Storages/MergeTree/IMergedBlockOutputStream.h index cfcfb177e05..f67cf66ee50 100644 --- a/src/Storages/MergeTree/IMergedBlockOutputStream.h +++ b/src/Storages/MergeTree/IMergedBlockOutputStream.h @@ -7,6 +7,7 @@ #include #include + namespace DB { diff --git a/src/Storages/MergeTree/MergeTask.cpp b/src/Storages/MergeTree/MergeTask.cpp index 74d6d60ba1b..b03fb1b12cf 100644 --- a/src/Storages/MergeTree/MergeTask.cpp +++ b/src/Storages/MergeTree/MergeTask.cpp @@ -6,11 +6,8 @@ #include #include -#include #include #include -#include -#include #include #include #include @@ -20,10 +17,8 @@ #include #include #include -#include #include #include -#include #include #include #include @@ -34,9 +29,6 @@ #include #include #include -#include -#include -#include #include #include #include @@ -48,6 +40,7 @@ #include #include + namespace ProfileEvents { extern const Event Merge; diff --git a/src/Storages/MergeTree/MergeTreeIOSettings.cpp b/src/Storages/MergeTree/MergeTreeIOSettings.cpp index 8b87c35b4e6..6705d75af41 100644 --- a/src/Storages/MergeTree/MergeTreeIOSettings.cpp +++ b/src/Storages/MergeTree/MergeTreeIOSettings.cpp @@ -26,6 +26,7 @@ namespace MergeTreeSetting extern const MergeTreeSettingsString primary_key_compression_codec; extern const MergeTreeSettingsBool use_adaptive_write_buffer_for_dynamic_subcolumns; extern const MergeTreeSettingsBool use_compact_variant_discriminators_serialization; + extern const MergeTreeSettingsUInt64 max_compression_threads; } MergeTreeWriterSettings::MergeTreeWriterSettings( @@ -54,6 +55,7 @@ MergeTreeWriterSettings::MergeTreeWriterSettings( , use_adaptive_write_buffer_for_dynamic_subcolumns((*storage_settings)[MergeTreeSetting::use_adaptive_write_buffer_for_dynamic_subcolumns]) , adaptive_write_buffer_initial_size((*storage_settings)[MergeTreeSetting::adaptive_write_buffer_initial_size]) { + query_write_settings.max_compression_threads = (*storage_settings)[MergeTreeSetting::max_compression_threads]; } } diff --git a/src/Storages/MergeTree/MergeTreeSettings.cpp b/src/Storages/MergeTree/MergeTreeSettings.cpp index 86d95aee242..4e7d0c0a721 100644 --- a/src/Storages/MergeTree/MergeTreeSettings.cpp +++ b/src/Storages/MergeTree/MergeTreeSettings.cpp @@ -53,6 +53,9 @@ namespace ErrorCodes M(Bool, load_existing_rows_count_for_old_parts, false, "Whether to load existing_rows_count for existing parts. If false, existing_rows_count will be equal to rows_count for existing parts.", 0) \ M(Bool, use_compact_variant_discriminators_serialization, true, "Use compact version of Variant discriminators serialization.", 0) \ \ + /** Merge and insert settings */ \ + M(UInt64, max_compression_threads, 1, "Maximum number of threads for writing compressed data. This is an expert-level setting, do not change it.", 0) \ + \ /** Merge selector settings. */ \ M(UInt64, merge_selector_blurry_base_scale_factor, 0, "Controls when the logic kicks in relatively to the number of parts in partition. The bigger the factor the more belated reaction will be.", 0) \ M(UInt64, merge_selector_window_size, 1000, "How many parts to look at once.", 0) \ From f4bd651b9474e5862c78ba84db15826a87a4b6e0 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Sun, 20 Oct 2024 22:37:48 +0000 Subject: [PATCH 242/680] cleanup --- tests/ci/ci.py | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/ci/ci.py b/tests/ci/ci.py index e820f445e7a..10431ce038f 100644 --- a/tests/ci/ci.py +++ b/tests/ci/ci.py @@ -1284,7 +1284,6 @@ def main() -> int: dump_to_file=True, ) print(f"Job report url: [{check_url}]") - print(job_report) prepared_events = prepare_tests_results_for_clickhouse( pr_info, job_report.test_results, From 0f3f15338d5c85fa159a85e929c689d69610d4ef Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 21 Oct 2024 02:09:15 +0200 Subject: [PATCH 243/680] Something --- src/Common/ThreadPool.h | 4 +-- .../ParallelCompressedWriteBuffer.cpp | 2 +- src/Dictionaries/HashedDictionary.h | 28 +++++++++++-------- .../Transforms/AggregatingTransform.h | 2 +- 4 files changed, 20 insertions(+), 16 deletions(-) diff --git a/src/Common/ThreadPool.h b/src/Common/ThreadPool.h index 7e497245acc..b52e4a60571 100644 --- a/src/Common/ThreadPool.h +++ b/src/Common/ThreadPool.h @@ -122,7 +122,7 @@ public: void scheduleOrThrowOnError(Job job, Priority priority = {}); /// Similar to scheduleOrThrowOnError(...). Wait for specified amount of time and schedule a job or return false. - bool trySchedule(Job job, Priority priority = {}, uint64_t wait_microseconds = 0) noexcept; + [[nodiscard]] bool trySchedule(Job job, Priority priority = {}, uint64_t wait_microseconds = 0) noexcept; /// Similar to scheduleOrThrowOnError(...). Wait for specified amount of time and schedule a job or throw an exception. void scheduleOrThrow(Job job, Priority priority = {}, uint64_t wait_microseconds = 0, bool propagate_opentelemetry_tracing_context = true); @@ -142,7 +142,7 @@ public: /// Returns true if the pool already terminated /// (and any further scheduling will produce CANNOT_SCHEDULE_TASK exception) - bool finished() const; + [[nodiscard]] bool finished() const; void setMaxThreads(size_t value); void setMaxFreeThreads(size_t value); diff --git a/src/Compression/ParallelCompressedWriteBuffer.cpp b/src/Compression/ParallelCompressedWriteBuffer.cpp index 1041a14979e..bd8d6371501 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.cpp +++ b/src/Compression/ParallelCompressedWriteBuffer.cpp @@ -46,7 +46,7 @@ void ParallelCompressedWriteBuffer::nextImpl() current_buffer->sequence_num = current_sequence_num; ++current_sequence_num; current_buffer->uncompressed_size = offset(); - pool.trySchedule([this, my_current_buffer = current_buffer, thread_group = CurrentThread::getGroup()] + pool.scheduleOrThrowOnError([this, my_current_buffer = current_buffer, thread_group = CurrentThread::getGroup()] { SCOPE_EXIT_SAFE( if (thread_group) diff --git a/src/Dictionaries/HashedDictionary.h b/src/Dictionaries/HashedDictionary.h index 7e935fe4855..ec5bc0a8a35 100644 --- a/src/Dictionaries/HashedDictionary.h +++ b/src/Dictionaries/HashedDictionary.h @@ -334,22 +334,26 @@ HashedDictionary::~HashedDictionary() if (container.empty()) return; - pool.trySchedule([&container, thread_group = CurrentThread::getGroup()] - { - SCOPE_EXIT_SAFE( + if (!pool.trySchedule([&container, thread_group = CurrentThread::getGroup()] + { + SCOPE_EXIT_SAFE( + if (thread_group) + CurrentThread::detachFromGroupIfNotDetached(); + ); + + /// Do not account memory that was occupied by the dictionaries for the query/user context. + MemoryTrackerBlockerInThread memory_blocker; + if (thread_group) - CurrentThread::detachFromGroupIfNotDetached(); - ); + CurrentThread::attachToGroupIfDetached(thread_group); + setThreadName("HashedDictDtor"); - /// Do not account memory that was occupied by the dictionaries for the query/user context. + clearContainer(container); + })) + { MemoryTrackerBlockerInThread memory_blocker; - - if (thread_group) - CurrentThread::attachToGroupIfDetached(thread_group); - setThreadName("HashedDictDtor"); - clearContainer(container); - }); + } ++hash_tables_count; }; diff --git a/src/Processors/Transforms/AggregatingTransform.h b/src/Processors/Transforms/AggregatingTransform.h index b9212375c91..398d7efa97e 100644 --- a/src/Processors/Transforms/AggregatingTransform.h +++ b/src/Processors/Transforms/AggregatingTransform.h @@ -107,7 +107,7 @@ struct ManyAggregatedData if (variant->aggregator) { // variant is moved here and will be destroyed in the destructor of the lambda function. - pool->trySchedule( + pool->scheduleOrThrowOnError( [my_variant = std::move(variant), thread_group = CurrentThread::getGroup()]() { SCOPE_EXIT_SAFE( From d552f51dfed0e6e6869a3b3b8a4e026d5fd2ca62 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Mon, 21 Oct 2024 00:12:55 +0000 Subject: [PATCH 244/680] cleanup --- CMakeLists.txt | 1 + utils/CMakeLists.txt | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/CMakeLists.txt b/CMakeLists.txt index f0965530739..a165be799c0 100644 --- a/CMakeLists.txt +++ b/CMakeLists.txt @@ -88,6 +88,7 @@ string (TOUPPER ${CMAKE_BUILD_TYPE} CMAKE_BUILD_TYPE_UC) list(REVERSE CMAKE_FIND_LIBRARY_SUFFIXES) option (ENABLE_FUZZING "Fuzzy testing using libfuzzer" OFF) +option (ENABLE_FUZZER_TEST "Build testing fuzzers in order to test libFuzzer functionality" OFF) if (ENABLE_FUZZING) # Also set WITH_COVERAGE=1 for better fuzzing process diff --git a/utils/CMakeLists.txt b/utils/CMakeLists.txt index 8c706ee6b67..2373a98239a 100644 --- a/utils/CMakeLists.txt +++ b/utils/CMakeLists.txt @@ -24,6 +24,6 @@ if (ENABLE_UTILS) add_subdirectory (memcpy-bench) endif () -if (ENABLE_FUZZING) +if (ENABLE_FUZZING AND ENABLE_FUZZER_TEST) add_subdirectory (libfuzzer-test) endif () From dffaf9b9a5b69ec4e342192ecbf00ad2c90208d8 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 21 Oct 2024 03:32:19 +0200 Subject: [PATCH 245/680] Fix error --- src/Compression/ParallelCompressedWriteBuffer.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Compression/ParallelCompressedWriteBuffer.cpp b/src/Compression/ParallelCompressedWriteBuffer.cpp index bd8d6371501..08c2a78c80b 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.cpp +++ b/src/Compression/ParallelCompressedWriteBuffer.cpp @@ -122,7 +122,7 @@ void ParallelCompressedWriteBuffer::compress(Iterator buffer) if (can_write_directly) { char * out_compressed_ptr = out.position() + sizeof(CityHash_v1_0_2::uint128); - UInt32 compressed_size = codec->compress(working_buffer.begin(), uncompressed_size, out_compressed_ptr); + UInt32 compressed_size = codec->compress(buffer->uncompressed.data(), uncompressed_size, out_compressed_ptr); CityHash_v1_0_2::uint128 checksum = CityHash_v1_0_2::CityHash128(out_compressed_ptr, compressed_size); From dba7c9cf4a990c2b29f8351b51f16a3614802499 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 21 Oct 2024 05:13:34 +0200 Subject: [PATCH 246/680] Add a test --- .../0_stateless/03254_parallel_compression.sql | 11 +++++++++++ 1 file changed, 11 insertions(+) create mode 100644 tests/queries/0_stateless/03254_parallel_compression.sql diff --git a/tests/queries/0_stateless/03254_parallel_compression.sql b/tests/queries/0_stateless/03254_parallel_compression.sql new file mode 100644 index 00000000000..a17deed7d8c --- /dev/null +++ b/tests/queries/0_stateless/03254_parallel_compression.sql @@ -0,0 +1,11 @@ +DROP TABLE IF EXISTS test2; + +CREATE TABLE test2 +( + k UInt64 +) ENGINE = MergeTree ORDER BY k SETTINGS min_compress_block_size = 10240, min_bytes_for_wide_part = 1, max_compression_threads = 64; + +INSERT INTO test2 SELECT number FROM numbers(20000); +SELECT sum(k) = (9999 * 10000 / 2 + 10000 * 9999) FROM test2 WHERE k > 10000; + +DROP TABLE test2; From d40a45399a5cd637ba6605d3306b8bdf91a00409 Mon Sep 17 00:00:00 2001 From: vdimir Date: Mon, 21 Oct 2024 07:59:30 +0000 Subject: [PATCH 247/680] fix build --- src/Processors/QueryPlan/JoinStep.cpp | 2 +- src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp | 8 ++++---- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index 848999e339c..6925d591968 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -186,7 +186,7 @@ void JoinStep::updateOutputHeader() return; } - auto column_permutation = getPermutationForBlock(result_header, input_streams[0].header, input_streams[1].header, required_output); + auto column_permutation = getPermutationForBlock(result_header, input_headers[0], input_headers[1], required_output); if (!column_permutation.empty()) result_header = ColumnPermuteTransform::permute(result_header, column_permutation); diff --git a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp index ced3b987b64..c0b31864eac 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp @@ -86,12 +86,12 @@ void optimizeJoin(QueryPlan::Node & node, QueryPlan::Nodes &) if (!need_swap) return; - const auto & streams = join_step->getInputStreams(); - if (streams.size() != 2) + const auto & headers = join_step->getInputHeaders(); + if (headers.size() != 2) return; - const auto & left_stream_input_header = streams.front().header; - const auto & right_stream_input_header = streams.back().header; + const auto & left_stream_input_header = headers.front(); + const auto & right_stream_input_header = headers.back(); auto updated_table_join = std::make_shared(table_join); updated_table_join->swapSides(); From a3405a0c042908753dc4c6e42d236c9601505a91 Mon Sep 17 00:00:00 2001 From: vdimir Date: Mon, 21 Oct 2024 13:08:14 +0000 Subject: [PATCH 248/680] upd test --- tests/integration/test_peak_memory_usage/test.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/integration/test_peak_memory_usage/test.py b/tests/integration/test_peak_memory_usage/test.py index 877cf97bb18..c31a2d8ae05 100644 --- a/tests/integration/test_peak_memory_usage/test.py +++ b/tests/integration/test_peak_memory_usage/test.py @@ -90,7 +90,7 @@ def test_clickhouse_client_max_peak_memory_usage_distributed(started_cluster): with client(name="client1>", log=client_output, command=command_text) as client1: client1.expect(prompt) client1.send( - "SELECT COUNT(*) FROM distributed_fixed_numbers JOIN fixed_numbers_2 ON distributed_fixed_numbers.number=fixed_numbers_2.number", + "SELECT COUNT(*) FROM distributed_fixed_numbers JOIN fixed_numbers_2 ON distributed_fixed_numbers.number=fixed_numbers_2.number SETTINGS query_plan_join_inner_table_selection = 'right'", ) client1.expect("Peak memory usage", timeout=60) client1.expect(prompt) From 40029beaf9891bc250d1203e2cd96788b6650a7b Mon Sep 17 00:00:00 2001 From: Igor Nikonov Date: Mon, 21 Oct 2024 13:49:53 +0000 Subject: [PATCH 249/680] Fix 02967_parallel_replicas_join_algo_and_analyzer_1.sh --- ...eplicas_join_algo_and_analyzer_1.reference | 16 ++++++++++++ ...allel_replicas_join_algo_and_analyzer_1.sh | 26 +++++++++++++++++-- 2 files changed, 40 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.reference b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.reference index e1bf9c27a81..7475cc7a97e 100644 --- a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.reference +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.reference @@ -28,3 +28,19 @@ SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value` FROM `default`.` SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` GLOBAL ALL INNER JOIN `_data_` AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` GLOBAL ALL INNER JOIN `_data_` AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) DefaultCoordinator: Coordination done + +simple (global) join with analyzer and parallel replicas with local plan +4200000 4200000 4200000 -1400000 +4200006 4200006 4200006 -1400002 +4200012 4200012 4200012 -1400004 +4200018 4200018 4200018 -1400006 +4200024 4200024 4200024 -1400008 +4200030 4200030 4200030 -1400010 +4200036 4200036 4200036 -1400012 +4200042 4200042 4200042 -1400014 +4200048 4200048 4200048 -1400016 +4200054 4200054 4200054 -1400018 +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value` FROM `default`.`num_2` AS `__table1` (stage: WithMergeableState) + DefaultCoordinator: Coordination done +SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` GLOBAL ALL INNER JOIN `_data_` AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) + DefaultCoordinator: Coordination done diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh index b4271c3d29b..1d43f540138 100755 --- a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh @@ -37,7 +37,7 @@ inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0" +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0, parallel_replicas_local_plan=0" $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l @@ -45,7 +45,29 @@ inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, send_logs_level='trace', max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0" 2>&1 | +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0, parallel_replicas_local_plan=0" 2>&1 | +grep "executeQuery\|.*Coordinator: Coordination done" | +grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | +sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' + +echo +echo "simple (global) join with analyzer and parallel replicas with local plan" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2) r on l.key = r.key +order by l.key limit 10 offset 700000 +SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, +max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0, parallel_replicas_local_plan=0" + +$CLICKHOUSE_CLIENT -q " +select * from (select key, value from num_1) l +inner join (select key, value from num_2) r on l.key = r.key +order by l.key limit 10 offset 700000 +SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, send_logs_level='trace', +max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, +cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0, parallel_replicas_local_plan=1" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' From e6bae901ed9a149e81cdf50e358470326b140c32 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 21 Oct 2024 18:39:20 +0200 Subject: [PATCH 250/680] Incomplete --- .../ParallelCompressedWriteBuffer.cpp | 6 ++++- .../ParallelCompressedWriteBuffer.h | 26 ++++++------------- .../MergeTreeDataPartWriterOnDisk.cpp | 2 ++ .../MergeTree/MergeTreeDataPartWriterOnDisk.h | 1 + .../MergeTree/MergeTreeDataPartWriterWide.cpp | 19 +++++++++----- 5 files changed, 29 insertions(+), 25 deletions(-) diff --git a/src/Compression/ParallelCompressedWriteBuffer.cpp b/src/Compression/ParallelCompressedWriteBuffer.cpp index 08c2a78c80b..303e1ece68a 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.cpp +++ b/src/Compression/ParallelCompressedWriteBuffer.cpp @@ -44,6 +44,8 @@ void ParallelCompressedWriteBuffer::nextImpl() /// The buffer will be compressed and processed in the thread. current_buffer->busy = true; current_buffer->sequence_num = current_sequence_num; + current_buffer->out_callback = callback; + callback = {}; ++current_sequence_num; current_buffer->uncompressed_size = offset(); pool.scheduleOrThrowOnError([this, my_current_buffer = current_buffer, thread_group = CurrentThread::getGroup()] @@ -60,7 +62,7 @@ void ParallelCompressedWriteBuffer::nextImpl() compress(my_current_buffer); }); - const BufferPair * previous_buffer = &*current_buffer; + BufferPair * previous_buffer = &*current_buffer; ++current_buffer; if (current_buffer == buffers.end()) { @@ -153,6 +155,8 @@ void ParallelCompressedWriteBuffer::compress(Iterator buffer) } std::unique_lock lock(mutex); + if (buffer->out_callback) + buffer->out_callback(); buffer->busy = false; cond.notify_all(); } diff --git a/src/Compression/ParallelCompressedWriteBuffer.h b/src/Compression/ParallelCompressedWriteBuffer.h index 4d1dfc79797..8c5f249b06c 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.h +++ b/src/Compression/ParallelCompressedWriteBuffer.h @@ -31,24 +31,11 @@ public: ~ParallelCompressedWriteBuffer() override; - /// The amount of compressed data - size_t getCompressedBytes() + /// This function will be called once after compressing the next data and sending it to the out. + /// It can be used to fill information about marks. + void setCompletionCallback(std::function callback_) { - nextIfAtEnd(); - return out.count(); - } - - /// How many uncompressed bytes were written to the buffer - size_t getUncompressedBytes() - { - return count(); - } - - /// How many bytes are in the buffer (not yet compressed) - size_t getRemainingBytes() - { - nextIfAtEnd(); - return offset(); + callback = callback_; } private: @@ -71,15 +58,18 @@ private: Memory<> uncompressed; size_t uncompressed_size = 0; PODArray compressed; - const BufferPair * previous = nullptr; + BufferPair * previous = nullptr; size_t sequence_num = 0; bool busy = false; + std::function out_callback; }; std::mutex mutex; std::condition_variable cond; std::list buffers; + std::function callback; + using Iterator = std::list::iterator; Iterator current_buffer; size_t current_sequence_num = 0; diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp index 8047d1cf2e9..89db8174636 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp @@ -113,6 +113,8 @@ MergeTreeDataPartWriterOnDisk::Stream::Stream( max_compress_block_size_, query_write_settings.max_compression_threads, *compression_thread_pool); + + is_compressor_parallel = true; } else { diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h index 046571cb83f..cb46785ccbd 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h @@ -76,6 +76,7 @@ public: std::unique_ptr plain_file; std::optional plain_hashing; /// This could be either CompressedWriteBuffer or ParallelCompressedWriteBuffer + bool is_compressor_parallel = false; std::unique_ptr compressor; std::optional compressed_hashing; diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp index 523e4c4a31d..860722ba870 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp @@ -400,15 +400,22 @@ StreamsWithMarks MergeTreeDataPartWriterWide::getCurrentMarksForColumn( auto & stream = *column_streams[stream_name]; /// There could already be enough data to compress into the new block. + auto push_mark = [&] + { + StreamNameAndMark stream_with_mark; + stream_with_mark.stream_name = stream_name; + stream_with_mark.mark.offset_in_compressed_file = stream.plain_hashing->count(); + stream_with_mark.mark.offset_in_decompressed_block = stream.compressed_hashing->offset(); + result.push_back(stream_with_mark); + }; + if (stream.compressed_hashing->offset() >= min_compress_block_size) + { + stream.compressed_hashing->next(); + } - StreamNameAndMark stream_with_mark; - stream_with_mark.stream_name = stream_name; - stream_with_mark.mark.offset_in_compressed_file = stream.plain_hashing->count(); - stream_with_mark.mark.offset_in_decompressed_block = stream.compressed_hashing->offset(); - - result.push_back(stream_with_mark); + push_mark(); }, name_and_type.type, column_sample); return result; From fc87cd4d52a2645174bfa1c5f85520ac3bc8a667 Mon Sep 17 00:00:00 2001 From: Igor Nikonov Date: Mon, 21 Oct 2024 20:19:08 +0000 Subject: [PATCH 251/680] Update 02967_parallel_replicas_join_algo_and_analyzer_2 --- ...allel_replicas_join_algo_and_analyzer_2.sh | 28 ++++++------------- 1 file changed, 9 insertions(+), 19 deletions(-) diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.sh b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.sh index ed13bf3321b..f0118ac62df 100755 --- a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.sh +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.sh @@ -17,6 +17,8 @@ insert into num_1 select number * 2, toString(number * 2) from numbers(1e7); insert into num_2 select number * 3, -number from numbers(1.5e6); " +PARALLEL_REPLICAS_SETTINGS="enable_parallel_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join = 1" + ############## echo echo "simple (local) join with analyzer and parallel replicas" @@ -25,17 +27,13 @@ $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" +SETTINGS enable_analyzer=1, $PARALLEL_REPLICAS_SETTINGS" $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, send_logs_level='trace', -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" 2>&1 | +SETTINGS enable_analyzer=1, send_logs_level='trace', $PARALLEL_REPLICAS_SETTINGS, " 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' @@ -49,17 +47,13 @@ $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, join_algorithm='full_sorting_merge', -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" +SETTINGS enable_analyzer=1, join_algorithm='full_sorting_merge', $PARALLEL_REPLICAS_SETTINGS" $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, join_algorithm='full_sorting_merge', send_logs_level='trace', -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" 2>&1 | +SETTINGS enable_analyzer=1, join_algorithm='full_sorting_merge', send_logs_level='trace', $PARALLEL_REPLICAS_SETTINGS" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' @@ -74,7 +68,7 @@ select * from (select key, value from num_1) l inner join (select key, value from num_2 inner join (select number * 7 as key from numbers(1e5)) as nn on num_2.key = nn.key settings parallel_replicas_prefer_local_join=1) r on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1" +SETTINGS enable_analyzer=1" ############## @@ -86,18 +80,14 @@ select * from (select key, value from num_1) l inner join (select key, value from num_2 inner join (select number * 7 as key from numbers(1e5)) as nn on num_2.key = nn.key settings parallel_replicas_prefer_local_join=1) r on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" +SETTINGS enable_analyzer=1, $PARALLEL_REPLICAS_SETTINGS" $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2 inner join (select number * 7 as key from numbers(1e5)) as nn on num_2.key = nn.key settings parallel_replicas_prefer_local_join=1) r on l.key = r.key order by l.key limit 10 offset 10000 -SETTINGS allow_experimental_analyzer=1, join_algorithm='full_sorting_merge', send_logs_level='trace', -allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=1" 2>&1 | +SETTINGS enable_analyzer=1, join_algorithm='full_sorting_merge', send_logs_level='trace', $PARALLEL_REPLICAS_SETTINGS" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' From ba11a188895d0ed123a6be62007222c1b0f0cfc1 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Tue, 22 Oct 2024 01:48:17 +0000 Subject: [PATCH 252/680] run fuzzers without shell --- tests/fuzz/runner.py | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 9eac0755d78..87495dff599 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -116,21 +116,23 @@ def run_fuzzer(fuzzer: str, timeout: int): try: with open(out_path, "wb") as out: subprocess.run( - cmd_line, + cmd_line.split(), stderr=out, stdout=subprocess.DEVNULL, text=True, check=True, - shell=True, + shell=False, errors="replace", timeout=timeout, ) except subprocess.CalledProcessError: + logging.info("Fail running %s", fuzzer) with open(status_path, "w", encoding="utf-8") as status: status.write( f"FAIL\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) except subprocess.TimeoutExpired: + logging.info("Timeout running %s", fuzzer) kill_fuzzer(fuzzer) sleep(10) with open(status_path, "w", encoding="utf-8") as status: @@ -138,6 +140,7 @@ def run_fuzzer(fuzzer: str, timeout: int): f"Timeout\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) else: + logging.info("Successful running %s", fuzzer) with open(status_path, "w", encoding="utf-8") as status: status.write( f"OK\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" From be77920fc8b91dc74edc01eaf3904eb383025752 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Tue, 22 Oct 2024 02:46:28 +0000 Subject: [PATCH 253/680] fix --- tests/fuzz/runner.py | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 87495dff599..ea4aef7d92b 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -108,8 +108,6 @@ def run_fuzzer(fuzzer: str, timeout: int): if not "-dict=" in cmd_line and Path(f"{fuzzer}.dict").exists(): cmd_line += f" -dict={fuzzer}.dict" - cmd_line += " < /dev/null" - logging.info("...will execute: %s", cmd_line) stopwatch = Stopwatch() @@ -117,8 +115,9 @@ def run_fuzzer(fuzzer: str, timeout: int): with open(out_path, "wb") as out: subprocess.run( cmd_line.split(), - stderr=out, + stdin=subprocess.DEVNULL, stdout=subprocess.DEVNULL, + stderr=out, text=True, check=True, shell=False, From b02ea90727fef66bef6a238a15058024a14029f2 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Tue, 22 Oct 2024 04:25:08 +0000 Subject: [PATCH 254/680] remove fuzzer args --- tests/fuzz/runner.py | 16 ++-------------- 1 file changed, 2 insertions(+), 14 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index ea4aef7d92b..7d1d6fe6c9e 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -32,16 +32,6 @@ class Stopwatch: self.start_time_str_value = self.start_time.strftime("%Y-%m-%d %H:%M:%S") -def kill_fuzzer(fuzzer: str): - with subprocess.Popen(["ps", "-A", "u"], stdout=subprocess.PIPE) as p: - out, _ = p.communicate() - for line in out.splitlines(): - if fuzzer.encode("utf-8") in line: - pid = int(line.split(None, 2)[1]) - logging.info("Killing fuzzer %s, pid %d", fuzzer, pid) - os.kill(pid, signal.SIGKILL) - - def run_fuzzer(fuzzer: str, timeout: int): logging.info("Running fuzzer %s...", fuzzer) @@ -95,7 +85,7 @@ def run_fuzzer(fuzzer: str, timeout: int): out_path = f"{OUTPUT}/{fuzzer}.out" cmd_line = ( - f"{DEBUGGER} ./{fuzzer} {FUZZER_ARGS} {active_corpus_dir} {seed_corpus_dir}" + f"{DEBUGGER} ./{fuzzer} {active_corpus_dir} {seed_corpus_dir}" ) cmd_line += f" -exact_artifact_path={exact_artifact_path}" @@ -132,8 +122,6 @@ def run_fuzzer(fuzzer: str, timeout: int): ) except subprocess.TimeoutExpired: logging.info("Timeout running %s", fuzzer) - kill_fuzzer(fuzzer) - sleep(10) with open(status_path, "w", encoding="utf-8") as status: status.write( f"Timeout\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" @@ -152,7 +140,7 @@ def main(): subprocess.check_call("ls -al", shell=True) - timeout = 30 + timeout = 60 match = re.search(r"(^|\s+)-max_total_time=(\d+)($|\s)", FUZZER_ARGS) if match: From a742ee863cbf74b3e108bd05564b5d7c0c270fcf Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Tue, 22 Oct 2024 04:25:53 +0000 Subject: [PATCH 255/680] fix --- tests/fuzz/runner.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 7d1d6fe6c9e..b37ad81b73c 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -144,7 +144,7 @@ def main(): match = re.search(r"(^|\s+)-max_total_time=(\d+)($|\s)", FUZZER_ARGS) if match: - timeout += int(match.group(2)) + timeout = int(match.group(2)) with Path() as current: for fuzzer in current.iterdir(): From c52986bab761430cf24fe03f526da814bc339dc8 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Tue, 22 Oct 2024 04:40:34 +0000 Subject: [PATCH 256/680] fix --- tests/fuzz/runner.py | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index b37ad81b73c..00c3683e7c7 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -5,10 +5,8 @@ import datetime import logging import os import re -import signal import subprocess from pathlib import Path -from time import sleep DEBUGGER = os.getenv("DEBUGGER", "") FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") @@ -84,9 +82,7 @@ def run_fuzzer(fuzzer: str, timeout: int): status_path = f"{OUTPUT}/{fuzzer}.status" out_path = f"{OUTPUT}/{fuzzer}.out" - cmd_line = ( - f"{DEBUGGER} ./{fuzzer} {active_corpus_dir} {seed_corpus_dir}" - ) + cmd_line = f"{DEBUGGER} ./{fuzzer} {active_corpus_dir} {seed_corpus_dir}" cmd_line += f" -exact_artifact_path={exact_artifact_path}" From 9da2a68357b5c859e5fc05f46c6cfb787b12b066 Mon Sep 17 00:00:00 2001 From: Igor Nikonov Date: Tue, 22 Oct 2024 10:28:36 +0000 Subject: [PATCH 257/680] Fix 02967_parallel_replicas_join_algo_and_analyzer_2 --- ...02967_parallel_replicas_join_algo_and_analyzer_2.reference | 3 --- .../02967_parallel_replicas_join_algo_and_analyzer_2.sh | 4 ++-- 2 files changed, 2 insertions(+), 5 deletions(-) diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.reference b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.reference index 297ec311f3e..f17d9aea3d5 100644 --- a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.reference +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.reference @@ -11,7 +11,6 @@ simple (local) join with analyzer and parallel replicas 4200048 4200048 4200048 -1400016 4200054 4200054 4200054 -1400018 SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) DefaultCoordinator: Coordination done simple (local) join with analyzer and parallel replicas and full sorting merge join @@ -26,7 +25,6 @@ simple (local) join with analyzer and parallel replicas and full sorting merge j 4200048 4200048 4200048 -1400016 4200054 4200054 4200054 -1400018 SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4`) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(700000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) WithOrderCoordinator: Coordination done nested join with analyzer @@ -53,5 +51,4 @@ nested join with analyzer and parallel replicas, both local 420336 420336 420336 -140112 420378 420378 420378 -140126 SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4` ALL INNER JOIN (SELECT `__table6`.`number` * 7 AS `key` FROM numbers(100000.) AS `__table6`) AS `__table5` ON `__table4`.`key` = `__table5`.`key` SETTINGS parallel_replicas_prefer_local_join = 1) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(10000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) -SELECT `__table1`.`key` AS `key`, `__table1`.`value` AS `value`, `__table3`.`key` AS `r.key`, `__table3`.`value` AS `r.value` FROM (SELECT `__table2`.`key` AS `key`, `__table2`.`value` AS `value` FROM `default`.`num_1` AS `__table2`) AS `__table1` ALL INNER JOIN (SELECT `__table4`.`key` AS `key`, `__table4`.`value` AS `value` FROM `default`.`num_2` AS `__table4` ALL INNER JOIN (SELECT `__table6`.`number` * 7 AS `key` FROM numbers(100000.) AS `__table6`) AS `__table5` ON `__table4`.`key` = `__table5`.`key` SETTINGS parallel_replicas_prefer_local_join = 1) AS `__table3` ON `__table1`.`key` = `__table3`.`key` ORDER BY `__table1`.`key` ASC LIMIT _CAST(10000, 'UInt64'), _CAST(10, 'UInt64') (stage: WithMergeableState) WithOrderCoordinator: Coordination done diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.sh b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.sh index f0118ac62df..4768e308f1e 100755 --- a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.sh +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_2.sh @@ -17,7 +17,7 @@ insert into num_1 select number * 2, toString(number * 2) from numbers(1e7); insert into num_2 select number * 3, -number from numbers(1.5e6); " -PARALLEL_REPLICAS_SETTINGS="enable_parallel_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join = 1" +PARALLEL_REPLICAS_SETTINGS="allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join = 1, parallel_replicas_local_plan=1" ############## echo @@ -33,7 +33,7 @@ $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS enable_analyzer=1, send_logs_level='trace', $PARALLEL_REPLICAS_SETTINGS, " 2>&1 | +SETTINGS enable_analyzer=1, send_logs_level='trace', $PARALLEL_REPLICAS_SETTINGS" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' From d1426886e3a7c6f2d3b4d2f81289a005324e6a5d Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Tue, 22 Oct 2024 12:25:15 +0000 Subject: [PATCH 258/680] timeout as OK run --- tests/ci/libfuzzer_test_check.py | 10 +++++----- tests/fuzz/runner.py | 12 ++++++------ 2 files changed, 11 insertions(+), 11 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index d7e79cc26fe..17cca9a47dc 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -177,7 +177,7 @@ def read_status(status_path: Path): def process_results(result_path: Path): test_results = [] oks = 0 - timeouts = 0 + errors = 0 fails = 0 for file in result_path.glob("*.status"): fuzzer = file.stem @@ -188,8 +188,8 @@ def process_results(result_path: Path): result = TestResult(fuzzer, status[0], float(status[2])) if status[0] == "OK": oks += 1 - elif status[0] == "Timeout": - timeouts += 1 + elif status[0] == "ERROR": + errors += 1 if file_path_out.exists(): result.set_log_files(f"['{file_path_out}']") else: @@ -202,7 +202,7 @@ def process_results(result_path: Path): result.set_log_files(f"['{file_path_out}']") test_results.append(result) - return [oks, timeouts, fails, test_results] + return [oks, errors, fails, test_results] def main(): @@ -284,7 +284,7 @@ def main(): success = results[1] == 0 and results[2] == 0 JobReport( - description=f"OK: {results[0]}, Timeout: {results[1]}, FAIL: {results[2]}", + description=f"OK: {results[0]}, ERROR: {results[1]}, FAIL: {results[2]}", test_results=results[3], status=SUCCESS if success else FAILURE, start_time=stopwatch.start_time_str, diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 00c3683e7c7..59cb9877adb 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -117,17 +117,17 @@ def run_fuzzer(fuzzer: str, timeout: int): f"FAIL\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) except subprocess.TimeoutExpired: - logging.info("Timeout running %s", fuzzer) - with open(status_path, "w", encoding="utf-8") as status: - status.write( - f"Timeout\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" - ) - else: logging.info("Successful running %s", fuzzer) with open(status_path, "w", encoding="utf-8") as status: status.write( f"OK\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) + else: + logging.info("Error running %s", fuzzer) + with open(status_path, "w", encoding="utf-8") as status: + status.write( + f"ERROR\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" + ) os.remove(out_path) From 32be533290f996f859eba842911cd0f0b017f52b Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Tue, 22 Oct 2024 15:22:59 +0000 Subject: [PATCH 259/680] better diagnostics --- tests/ci/libfuzzer_test_check.py | 5 +++++ tests/fuzz/runner.py | 12 +++++++++--- 2 files changed, 14 insertions(+), 3 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 17cca9a47dc..45370b0cd00 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -184,6 +184,7 @@ def process_results(result_path: Path): file_path = file.parent / fuzzer file_path_unit = file_path.with_suffix(".unit") file_path_out = file_path.with_suffix(".out") + file_path_stdout = file_path.with_suffix(".stdout") status = read_status(file) result = TestResult(fuzzer, status[0], float(status[2])) if status[0] == "OK": @@ -192,6 +193,8 @@ def process_results(result_path: Path): errors += 1 if file_path_out.exists(): result.set_log_files(f"['{file_path_out}']") + elif file_path_stdout.exists(): + result.set_log_files(f"['{file_path_stdout}']") else: fails += 1 if file_path_out.exists(): @@ -200,6 +203,8 @@ def process_results(result_path: Path): result.set_log_files(f"['{file_path_unit}']") elif file_path_out.exists(): result.set_log_files(f"['{file_path_out}']") + elif file_path_stdout.exists(): + result.set_log_files(f"['{file_path_stdout}']") test_results.append(result) return [oks, errors, fails, test_results] diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 59cb9877adb..2c1d57ce5eb 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -81,6 +81,7 @@ def run_fuzzer(fuzzer: str, timeout: int): exact_artifact_path = f"{OUTPUT}/{fuzzer}.unit" status_path = f"{OUTPUT}/{fuzzer}.status" out_path = f"{OUTPUT}/{fuzzer}.out" + stdout_path = f"{OUTPUT}/{fuzzer}.stdout" cmd_line = f"{DEBUGGER} ./{fuzzer} {active_corpus_dir} {seed_corpus_dir}" @@ -98,11 +99,11 @@ def run_fuzzer(fuzzer: str, timeout: int): stopwatch = Stopwatch() try: - with open(out_path, "wb") as out: + with open(out_path, "wb") as out, open(stdout_path, "wb") as stdout: subprocess.run( cmd_line.split(), stdin=subprocess.DEVNULL, - stdout=subprocess.DEVNULL, + stdout=stdout, stderr=out, text=True, check=True, @@ -122,13 +123,18 @@ def run_fuzzer(fuzzer: str, timeout: int): status.write( f"OK\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) + except Exception as e: + logging.info("Unexpected exception running %s: %s", fuzzer, e) + with open(status_path, "w", encoding="utf-8") as status: + status.write( + f"ERROR\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" + ) else: logging.info("Error running %s", fuzzer) with open(status_path, "w", encoding="utf-8") as status: status.write( f"ERROR\n{stopwatch.start_time_str}\n{stopwatch.duration_seconds}\n" ) - os.remove(out_path) def main(): From e3ebe51968acf6a43922f12a9443c8e17a9cabc2 Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Wed, 23 Oct 2024 01:27:10 +0000 Subject: [PATCH 260/680] Make ParquetMetadata say whether bloom filter is present --- .../Impl/ParquetMetadataInputFormat.cpp | 5 +- .../02718_parquet_metadata_format.reference | 70 +++++++++++++++++-- .../02718_parquet_metadata_format.sh | 1 + 3 files changed, 69 insertions(+), 7 deletions(-) diff --git a/src/Processors/Formats/Impl/ParquetMetadataInputFormat.cpp b/src/Processors/Formats/Impl/ParquetMetadataInputFormat.cpp index 7fd6e93dd80..8264b565e39 100644 --- a/src/Processors/Formats/Impl/ParquetMetadataInputFormat.cpp +++ b/src/Processors/Formats/Impl/ParquetMetadataInputFormat.cpp @@ -92,8 +92,9 @@ static NamesAndTypesList getHeaderForParquetMetadata() std::make_shared(std::make_shared()), std::make_shared(std::make_shared())}, Names{"num_values", "null_count", "distinct_count", "min", "max"}), + DataTypeFactory::instance().get("Bool"), }, - Names{"name", "path", "total_compressed_size", "total_uncompressed_size", "have_statistics", "statistics"}))}, + Names{"name", "path", "total_compressed_size", "total_uncompressed_size", "have_statistics", "statistics", "have_bloom_filter"}))}, Names{"num_columns", "num_rows", "total_uncompressed_size", "total_compressed_size", "columns"}))}, }; return names_and_types; @@ -350,6 +351,8 @@ void ParquetMetadataInputFormat::fillColumnChunksMetadata(const std::unique_ptr< fillColumnStatistics(column_chunk_metadata->statistics(), tuple_column.getColumn(5), row_group_metadata->schema()->Column(column_i)->type_length()); else tuple_column.getColumn(5).insertDefault(); + bool have_bloom_filter = column_chunk_metadata->bloom_filter_offset().has_value(); + assert_cast(tuple_column.getColumn(6)).insertValue(have_bloom_filter); } array_column.getOffsets().push_back(tuple_column.size()); } diff --git a/tests/queries/0_stateless/02718_parquet_metadata_format.reference b/tests/queries/0_stateless/02718_parquet_metadata_format.reference index 1f55c29da56..815968aeba5 100644 --- a/tests/queries/0_stateless/02718_parquet_metadata_format.reference +++ b/tests/queries/0_stateless/02718_parquet_metadata_format.reference @@ -78,7 +78,8 @@ "distinct_count": null, "min": "0", "max": "999" - } + }, + "have_bloom_filter": false }, { "name": "str", @@ -92,7 +93,8 @@ "distinct_count": null, "min": "Hello0", "max": "Hello999" - } + }, + "have_bloom_filter": false }, { "name": "mod", @@ -106,7 +108,8 @@ "distinct_count": null, "min": "0", "max": "8" - } + }, + "have_bloom_filter": false } ] }, @@ -128,7 +131,8 @@ "distinct_count": null, "min": "0", "max": "999" - } + }, + "have_bloom_filter": false }, { "name": "str", @@ -142,7 +146,8 @@ "distinct_count": null, "min": "Hello0", "max": "Hello999" - } + }, + "have_bloom_filter": false }, { "name": "mod", @@ -156,7 +161,8 @@ "distinct_count": null, "min": "0", "max": "8" - } + }, + "have_bloom_filter": false } ] } @@ -223,3 +229,55 @@ } 1 1 +{ + "num_columns": "1", + "num_rows": "5", + "num_row_groups": "1", + "format_version": "1.0", + "metadata_size": "267", + "total_uncompressed_size": "105", + "total_compressed_size": "128", + "columns": [ + { + "name": "ipv6", + "path": "ipv6", + "max_definition_level": "0", + "max_repetition_level": "0", + "physical_type": "FIXED_LEN_BYTE_ARRAY", + "logical_type": "None", + "compression": "GZIP", + "total_uncompressed_size": "105", + "total_compressed_size": "128", + "space_saved": "-21.9%", + "encodings": [ + "PLAIN", + "BIT_PACKED" + ] + } + ], + "row_groups": [ + { + "num_columns": "1", + "num_rows": "5", + "total_uncompressed_size": "105", + "total_compressed_size": "128", + "columns": [ + { + "name": "ipv6", + "path": "ipv6", + "total_compressed_size": "128", + "total_uncompressed_size": "105", + "have_statistics": true, + "statistics": { + "num_values": "5", + "null_count": "0", + "distinct_count": null, + "min": "27 32 150 125 17 250 66 31 157 44 75 218 51 50 19 144 ", + "max": "154 31 90 141 15 7 68 47 190 29 121 145 188 162 234 154 " + }, + "have_bloom_filter": true + } + ] + } + ] +} diff --git a/tests/queries/0_stateless/02718_parquet_metadata_format.sh b/tests/queries/0_stateless/02718_parquet_metadata_format.sh index 94d7f453850..c6371cff7a3 100755 --- a/tests/queries/0_stateless/02718_parquet_metadata_format.sh +++ b/tests/queries/0_stateless/02718_parquet_metadata_format.sh @@ -17,3 +17,4 @@ $CLICKHOUSE_LOCAL -q "select some_column from file('$CURDIR/data_parquet/02718_d $CLICKHOUSE_LOCAL -q "select num_columns from file('$CURDIR/data_parquet/02718_data.parquet', ParquetMetadata, 'num_columns Array(UInt32)')" 2>&1 | grep -c "BAD_ARGUMENTS" +$CLICKHOUSE_LOCAL -q "select * from file('$CURDIR/data_parquet/ipv6_bloom_filter.gz.parquet', ParquetMetadata) format JSONEachRow" | python3 -m json.tool From b958dcb50fb994f6375e04196df193ea5106c1d2 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 23 Oct 2024 14:36:27 +0000 Subject: [PATCH 261/680] reorganize command line, add CI.FUZZER_ARGS option --- tests/fuzz/clickhouse_fuzzer.options | 2 ++ tests/fuzz/runner.py | 24 +++++++++++++++--------- 2 files changed, 17 insertions(+), 9 deletions(-) create mode 100644 tests/fuzz/clickhouse_fuzzer.options diff --git a/tests/fuzz/clickhouse_fuzzer.options b/tests/fuzz/clickhouse_fuzzer.options new file mode 100644 index 00000000000..a22ba7b3b88 --- /dev/null +++ b/tests/fuzz/clickhouse_fuzzer.options @@ -0,0 +1,2 @@ +[CI] +FUZZER_ARGS = true diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 2c1d57ce5eb..40b55700623 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -44,6 +44,7 @@ def run_fuzzer(fuzzer: str, timeout: int): options_file = f"{fuzzer}.options" custom_libfuzzer_options = "" fuzzer_arguments = "" + use_fuzzer_args = False with Path(options_file) as path: if path.exists() and path.is_file(): @@ -78,24 +79,28 @@ def run_fuzzer(fuzzer: str, timeout: int): for key, value in parser["fuzzer_arguments"].items() ) + use_fuzzer_args = parser.getboolean("CI", "FUZZER_ARGS", fallback=False) + exact_artifact_path = f"{OUTPUT}/{fuzzer}.unit" status_path = f"{OUTPUT}/{fuzzer}.status" out_path = f"{OUTPUT}/{fuzzer}.out" stdout_path = f"{OUTPUT}/{fuzzer}.stdout" - cmd_line = f"{DEBUGGER} ./{fuzzer} {active_corpus_dir} {seed_corpus_dir}" + if not "-dict=" in custom_libfuzzer_options and Path(f"{fuzzer}.dict").exists(): + custom_libfuzzer_options += f" -dict={fuzzer}.dict" + custom_libfuzzer_options += f" -exact_artifact_path={exact_artifact_path}" - cmd_line += f" -exact_artifact_path={exact_artifact_path}" + libfuzzer_corpora = f"{active_corpus_dir} {seed_corpus_dir}" - if custom_libfuzzer_options: - cmd_line += f" {custom_libfuzzer_options}" - if fuzzer_arguments: - cmd_line += f" {fuzzer_arguments}" + cmd_line = f"{DEBUGGER} ./{fuzzer} {fuzzer_arguments}" - if not "-dict=" in cmd_line and Path(f"{fuzzer}.dict").exists(): - cmd_line += f" -dict={fuzzer}.dict" + env = None + if use_fuzzer_args: + env = {"FUZZER_ARGS": f"{custom_libfuzzer_options} {libfuzzer_corpora}"} + else: + cmd_line += f" {custom_libfuzzer_options} {libfuzzer_corpora}" - logging.info("...will execute: %s", cmd_line) + logging.info("...will execute: %s%s", cmd_line, f" with FUZZER_ARGS {env["FUZZER_ARGS"]}" if use_fuzzer_args else "") stopwatch = Stopwatch() try: @@ -110,6 +115,7 @@ def run_fuzzer(fuzzer: str, timeout: int): shell=False, errors="replace", timeout=timeout, + env=env, ) except subprocess.CalledProcessError: logging.info("Fail running %s", fuzzer) From 19cdbf62c53070085e49657e890b74e9f5979a9f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 23 Oct 2024 14:57:05 +0000 Subject: [PATCH 262/680] fix --- tests/fuzz/runner.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 40b55700623..62f1666e77f 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -95,12 +95,14 @@ def run_fuzzer(fuzzer: str, timeout: int): cmd_line = f"{DEBUGGER} ./{fuzzer} {fuzzer_arguments}" env = None + with_fuzzer_args = "" if use_fuzzer_args: env = {"FUZZER_ARGS": f"{custom_libfuzzer_options} {libfuzzer_corpora}"} + with_fuzzer_args = f" with FUZZER_ARGS '{custom_libfuzzer_options} {libfuzzer_corpora}'" else: cmd_line += f" {custom_libfuzzer_options} {libfuzzer_corpora}" - logging.info("...will execute: %s%s", cmd_line, f" with FUZZER_ARGS {env["FUZZER_ARGS"]}" if use_fuzzer_args else "") + logging.info("...will execute: '%s'%s", cmd_line, with_fuzzer_args) stopwatch = Stopwatch() try: From a5e3f7a213c3c830ddff7ba6b937909b174ce0a1 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Wed, 23 Oct 2024 15:13:04 +0000 Subject: [PATCH 263/680] Automatic style fix --- tests/fuzz/runner.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 62f1666e77f..63f53be3766 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -98,7 +98,9 @@ def run_fuzzer(fuzzer: str, timeout: int): with_fuzzer_args = "" if use_fuzzer_args: env = {"FUZZER_ARGS": f"{custom_libfuzzer_options} {libfuzzer_corpora}"} - with_fuzzer_args = f" with FUZZER_ARGS '{custom_libfuzzer_options} {libfuzzer_corpora}'" + with_fuzzer_args = ( + f" with FUZZER_ARGS '{custom_libfuzzer_options} {libfuzzer_corpora}'" + ) else: cmd_line += f" {custom_libfuzzer_options} {libfuzzer_corpora}" From b17c6ba73ea18e0e86782966080ebd4841b893cf Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Wed, 23 Oct 2024 14:01:05 -0400 Subject: [PATCH 264/680] trigger build --- src/DataTypes/fuzzers/CMakeLists.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/src/DataTypes/fuzzers/CMakeLists.txt b/src/DataTypes/fuzzers/CMakeLists.txt index 8dedd3470e2..8940586fc70 100644 --- a/src/DataTypes/fuzzers/CMakeLists.txt +++ b/src/DataTypes/fuzzers/CMakeLists.txt @@ -1,2 +1,3 @@ clickhouse_add_executable(data_type_deserialization_fuzzer data_type_deserialization_fuzzer.cpp ${SRCS}) + target_link_libraries(data_type_deserialization_fuzzer PRIVATE clickhouse_aggregate_functions dbms) From dc1d1f080a3ecf97424138f3d6eb203a34ec3b1b Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 23 Oct 2024 20:24:54 +0000 Subject: [PATCH 265/680] fix --- tests/fuzz/runner.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 63f53be3766..5fb40173e0c 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -97,9 +97,9 @@ def run_fuzzer(fuzzer: str, timeout: int): env = None with_fuzzer_args = "" if use_fuzzer_args: - env = {"FUZZER_ARGS": f"{custom_libfuzzer_options} {libfuzzer_corpora}"} + env = {"FUZZER_ARGS": f"{custom_libfuzzer_options} {libfuzzer_corpora}".strip()} with_fuzzer_args = ( - f" with FUZZER_ARGS '{custom_libfuzzer_options} {libfuzzer_corpora}'" + f" with FUZZER_ARGS '{env['FUZZER_ARGS']}'" ) else: cmd_line += f" {custom_libfuzzer_options} {libfuzzer_corpora}" From 4c9743ca42b2806bc981d393931f05ed8ade0c99 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Wed, 23 Oct 2024 20:38:00 +0000 Subject: [PATCH 266/680] Automatic style fix --- tests/fuzz/runner.py | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 5fb40173e0c..af73a989ec3 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -98,9 +98,7 @@ def run_fuzzer(fuzzer: str, timeout: int): with_fuzzer_args = "" if use_fuzzer_args: env = {"FUZZER_ARGS": f"{custom_libfuzzer_options} {libfuzzer_corpora}".strip()} - with_fuzzer_args = ( - f" with FUZZER_ARGS '{env['FUZZER_ARGS']}'" - ) + with_fuzzer_args = f" with FUZZER_ARGS '{env['FUZZER_ARGS']}'" else: cmd_line += f" {custom_libfuzzer_options} {libfuzzer_corpora}" From f93ac138f109c6a30354231240c47324dc51541f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 23 Oct 2024 22:21:37 +0000 Subject: [PATCH 267/680] chown clickhouse data path to root --- tests/ci/libfuzzer_test_check.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 45370b0cd00..2a307d07231 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -215,6 +215,8 @@ def main(): stopwatch = Stopwatch() + os.chown("/var/lib/clickhouse", 0, 0) + temp_path = Path(TEMP_PATH) reports_path = Path(REPORT_PATH) temp_path.mkdir(parents=True, exist_ok=True) From 77c2b9e5fc2e19483a1c0f675e32757eec202a1f Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Wed, 23 Oct 2024 22:44:10 +0000 Subject: [PATCH 268/680] create clickhouse data dir --- tests/ci/libfuzzer_test_check.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 2a307d07231..7091f076b99 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -215,7 +215,8 @@ def main(): stopwatch = Stopwatch() - os.chown("/var/lib/clickhouse", 0, 0) + data_path = "/var/lib/clickhouse" + os.makedirs(data_path, exist_ok=True) temp_path = Path(TEMP_PATH) reports_path = Path(REPORT_PATH) From efd8ea7757deb9326abbe91c12e9b58629fd236c Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Thu, 24 Oct 2024 03:59:03 +0000 Subject: [PATCH 269/680] set uid gid --- tests/ci/libfuzzer_test_check.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 7091f076b99..379d681cb3e 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -81,10 +81,13 @@ def get_run_command( envs += [f"-e {e}" for e in additional_envs] env_str = " ".join(envs) + uid = os.getuid() + gid = os.getgid() return ( f"docker run " f"{ci_logs_args} " + f"--user {uid}:{gid} " f"--workdir=/fuzzers " f"--volume={fuzzers_path}:/fuzzers " f"--volume={repo_path}/tests:/usr/share/clickhouse-test " @@ -215,9 +218,6 @@ def main(): stopwatch = Stopwatch() - data_path = "/var/lib/clickhouse" - os.makedirs(data_path, exist_ok=True) - temp_path = Path(TEMP_PATH) reports_path = Path(REPORT_PATH) temp_path.mkdir(parents=True, exist_ok=True) From 32fe869e3191fb34e03e0af188d1da979a8cbea7 Mon Sep 17 00:00:00 2001 From: vdimir Date: Thu, 24 Oct 2024 12:18:47 +0000 Subject: [PATCH 270/680] reresolve conflicts --- src/Core/Settings.cpp | 10 ---- src/Core/Settings.h | 1 - src/Interpreters/executeQuery.cpp | 10 +--- src/Processors/QueryPlan/AggregatingStep.cpp | 5 +- src/Processors/QueryPlan/AggregatingStep.h | 2 + src/Processors/QueryPlan/CreatingSetsStep.h | 5 ++ src/Processors/QueryPlan/IQueryPlanStep.cpp | 17 ++++++ src/Processors/QueryPlan/IQueryPlanStep.h | 28 ++-------- src/Processors/QueryPlan/ISourceStep.h | 3 + src/Processors/QueryPlan/ITransformingStep.h | 2 - .../QueryPlan/IntersectOrExceptStep.cpp | 6 +- .../QueryPlan/IntersectOrExceptStep.h | 2 + src/Processors/QueryPlan/JoinStep.h | 2 - .../Optimizations/filterPushDown.cpp | 15 +---- src/Processors/QueryPlan/QueryPlan.cpp | 35 ------------ src/Processors/QueryPlan/UnionStep.h | 2 - src/Processors/Transforms/FilterTransform.cpp | 55 +++---------------- src/Processors/Transforms/FilterTransform.h | 2 +- 18 files changed, 55 insertions(+), 147 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 16850fe2900..09b76b7daec 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -6204,16 +6204,6 @@ std::vector Settings::getUnchangedNames() const return setting_names; } -std::vector Settings::getChangedNames() const -{ - std::vector setting_names; - for (const auto & setting : impl->allChanged()) - { - setting_names.emplace_back(setting.getName()); - } - return setting_names; -} - void Settings::dumpToSystemSettingsColumns(MutableColumnsAndConstraints & params) const { MutableColumns & res_columns = params.res_columns; diff --git a/src/Core/Settings.h b/src/Core/Settings.h index fb88f6c6ebe..77281a4c518 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -135,7 +135,6 @@ struct Settings std::vector getAllRegisteredNames() const; std::vector getChangedAndObsoleteNames() const; std::vector getUnchangedNames() const; - std::vector getChangedNames() const; void dumpToSystemSettingsColumns(MutableColumnsAndConstraints & params) const; void dumpToMapColumn(IColumn * column, bool changed_only = true) const; diff --git a/src/Interpreters/executeQuery.cpp b/src/Interpreters/executeQuery.cpp index 8949ec5bb5a..2ce921967ba 100644 --- a/src/Interpreters/executeQuery.cpp +++ b/src/Interpreters/executeQuery.cpp @@ -576,14 +576,10 @@ void logQueryFinish( if (settings[Setting::log_query_settings]) { - auto changed_settings_names = settings.getChangedNames(); - for (const auto & name : changed_settings_names) + auto changes = settings.changes(); + for (const auto & change : changes) { - Field value = settings.get(name); - String value_str = convertFieldToString(value); - - query_span->addAttribute(fmt::format("clickhouse.setting.{}", name), value_str); - + query_span->addAttribute(fmt::format("clickhouse.setting.{}", change.name), convertFieldToString(change.value)); } } query_span->finish(); diff --git a/src/Processors/QueryPlan/AggregatingStep.cpp b/src/Processors/QueryPlan/AggregatingStep.cpp index defe7d0489a..efe14edaf35 100644 --- a/src/Processors/QueryPlan/AggregatingStep.cpp +++ b/src/Processors/QueryPlan/AggregatingStep.cpp @@ -589,8 +589,11 @@ AggregatingProjectionStep::AggregatingProjectionStep( , merge_threads(merge_threads_) , temporary_data_merge_threads(temporary_data_merge_threads_) { - input_headers = std::move(input_headers_); + updateInputHeaders(std::move(input_headers_)); +} +void AggregatingProjectionStep::updateOutputHeader() +{ if (input_headers.size() != 2) throw Exception( ErrorCodes::LOGICAL_ERROR, diff --git a/src/Processors/QueryPlan/AggregatingStep.h b/src/Processors/QueryPlan/AggregatingStep.h index b1f28f17ef9..d76764f05ba 100644 --- a/src/Processors/QueryPlan/AggregatingStep.h +++ b/src/Processors/QueryPlan/AggregatingStep.h @@ -123,6 +123,8 @@ public: QueryPipelineBuilderPtr updatePipeline(QueryPipelineBuilders pipelines, const BuildQueryPipelineSettings &) override; private: + void updateOutputHeader() override; + Aggregator::Params params; bool final; size_t merge_threads; diff --git a/src/Processors/QueryPlan/CreatingSetsStep.h b/src/Processors/QueryPlan/CreatingSetsStep.h index 54548a53131..0495ca2e638 100644 --- a/src/Processors/QueryPlan/CreatingSetsStep.h +++ b/src/Processors/QueryPlan/CreatingSetsStep.h @@ -45,6 +45,9 @@ public: QueryPipelineBuilderPtr updatePipeline(QueryPipelineBuilders pipelines, const BuildQueryPipelineSettings &) override; void describePipeline(FormatSettings & settings) const override; + +private: + void updateOutputHeader() override { output_header = getInputHeaders().front(); } }; /// This is a temporary step which is converted to CreatingSetStep after plan optimization. @@ -64,6 +67,8 @@ public: PreparedSets::Subqueries detachSets() { return std::move(subqueries); } private: + void updateOutputHeader() override { output_header = getInputHeaders().front(); } + PreparedSets::Subqueries subqueries; ContextPtr context; }; diff --git a/src/Processors/QueryPlan/IQueryPlanStep.cpp b/src/Processors/QueryPlan/IQueryPlanStep.cpp index bb1451287d9..aeb94e8826d 100644 --- a/src/Processors/QueryPlan/IQueryPlanStep.cpp +++ b/src/Processors/QueryPlan/IQueryPlanStep.cpp @@ -10,6 +10,23 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; } +void IQueryPlanStep::updateInputHeaders(Headers input_headers_) +{ + input_headers = std::move(input_headers_); + updateOutputHeader(); +} + +void IQueryPlanStep::updateInputHeader(Header input_header, size_t idx) +{ + if (idx >= input_headers.size()) + throw Exception(ErrorCodes::LOGICAL_ERROR, + "Cannot update input header {} for step {} because it has only {} headers", + idx, getName(), input_headers.size()); + + input_headers[idx] = input_header; + updateOutputHeader(); +} + const Header & IQueryPlanStep::getOutputHeader() const { if (!hasOutputHeader()) diff --git a/src/Processors/QueryPlan/IQueryPlanStep.h b/src/Processors/QueryPlan/IQueryPlanStep.h index c3eeb8ebf48..36a25b8fc21 100644 --- a/src/Processors/QueryPlan/IQueryPlanStep.h +++ b/src/Processors/QueryPlan/IQueryPlanStep.h @@ -16,11 +16,6 @@ using Processors = std::vector; namespace JSONBuilder { class JSONMap; } -namespace ErrorCodes -{ - extern const int NOT_IMPLEMENTED; -} - class QueryPlan; using QueryPlanRawPtrs = std::list; @@ -82,27 +77,12 @@ public: /// Updates the input streams of the given step. Used during query plan optimizations. /// It won't do any validation of new streams, so it is your responsibility to ensure that this update doesn't break anything - /// (e.g. you update data stream traits or correctly remove / add columns). - void updateInputHeaders(Headers input_headers_) - { - chassert(canUpdateInputHeader()); - input_headers = std::move(input_headers_); - updateOutputHeader(); - } - - void updateInputHeader(Header input_header) { updateInputHeaders(Headers{input_header}); } - - void updateInputHeader(Header input_header, size_t idx) - { - chassert(canUpdateInputHeader() && idx < input_headers.size()); - input_headers[idx] = input_header; - updateOutputHeader(); - } - - virtual bool canUpdateInputHeader() const { return false; } + /// (e.g. you correctly remove / add columns). + void updateInputHeaders(Headers input_headers_); + void updateInputHeader(Header input_header, size_t idx = 0); protected: - virtual void updateOutputHeader() { throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Not implemented"); } + virtual void updateOutputHeader() = 0; Headers input_headers; std::optional

output_header; diff --git a/src/Processors/QueryPlan/ISourceStep.h b/src/Processors/QueryPlan/ISourceStep.h index 142d97fecab..d1aa900bdbe 100644 --- a/src/Processors/QueryPlan/ISourceStep.h +++ b/src/Processors/QueryPlan/ISourceStep.h @@ -15,6 +15,9 @@ public: virtual void initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings & settings) = 0; void describePipeline(FormatSettings & settings) const override; + +protected: + void updateOutputHeader() override {} }; } diff --git a/src/Processors/QueryPlan/ITransformingStep.h b/src/Processors/QueryPlan/ITransformingStep.h index f27fc189dcd..5c7a03ad575 100644 --- a/src/Processors/QueryPlan/ITransformingStep.h +++ b/src/Processors/QueryPlan/ITransformingStep.h @@ -66,8 +66,6 @@ public: throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Not implemented"); } - bool canUpdateInputHeader() const override { return true; } - protected: TransformTraits transform_traits; diff --git a/src/Processors/QueryPlan/IntersectOrExceptStep.cpp b/src/Processors/QueryPlan/IntersectOrExceptStep.cpp index 48bf5dfa192..aec69302f92 100644 --- a/src/Processors/QueryPlan/IntersectOrExceptStep.cpp +++ b/src/Processors/QueryPlan/IntersectOrExceptStep.cpp @@ -34,7 +34,11 @@ IntersectOrExceptStep::IntersectOrExceptStep( : current_operator(operator_) , max_threads(max_threads_) { - input_headers = std::move(input_headers_); + updateInputHeaders(std::move(input_headers_)); +} + +void IntersectOrExceptStep::updateOutputHeader() +{ output_header = checkHeaders(input_headers); } diff --git a/src/Processors/QueryPlan/IntersectOrExceptStep.h b/src/Processors/QueryPlan/IntersectOrExceptStep.h index a1e85e847da..cc1d6059e04 100644 --- a/src/Processors/QueryPlan/IntersectOrExceptStep.h +++ b/src/Processors/QueryPlan/IntersectOrExceptStep.h @@ -21,6 +21,8 @@ public: void describePipeline(FormatSettings & settings) const override; private: + void updateOutputHeader() override; + Operator current_operator; size_t max_threads; }; diff --git a/src/Processors/QueryPlan/JoinStep.h b/src/Processors/QueryPlan/JoinStep.h index cf1ed7e4247..1eca42c62cf 100644 --- a/src/Processors/QueryPlan/JoinStep.h +++ b/src/Processors/QueryPlan/JoinStep.h @@ -37,8 +37,6 @@ public: void setJoin(JoinPtr join_, bool swap_streams_ = false); bool allowPushDownToRight() const; - bool canUpdateInputHeader() const override { return true; } - JoinInnerTableSelectionMode inner_table_selection_mode = JoinInnerTableSelectionMode::Right; private: diff --git a/src/Processors/QueryPlan/Optimizations/filterPushDown.cpp b/src/Processors/QueryPlan/Optimizations/filterPushDown.cpp index 524baae2859..63359d039e8 100644 --- a/src/Processors/QueryPlan/Optimizations/filterPushDown.cpp +++ b/src/Processors/QueryPlan/Optimizations/filterPushDown.cpp @@ -152,20 +152,7 @@ addNewFilterStepOrThrow(QueryPlan::Node * parent_node, QueryPlan::Nodes & nodes, node.step = std::make_unique( node.children.at(0)->step->getOutputHeader(), std::move(split_filter), std::move(split_filter_column_name), can_remove_filter); - if (auto * transforming_step = dynamic_cast(child.get())) - { - transforming_step->updateInputHeader(node.step->getOutputHeader()); - } - else - { - if (auto * join = typeid_cast(child.get())) - { - join->updateInputHeader(node.step->getOutputHeader(), child_idx); - } - else - throw Exception( - ErrorCodes::LOGICAL_ERROR, "We are trying to push down a filter through a step for which we cannot update input stream"); - } + child->updateInputHeader(node.step->getOutputHeader(), child_idx); if (update_parent_filter) { diff --git a/src/Processors/QueryPlan/QueryPlan.cpp b/src/Processors/QueryPlan/QueryPlan.cpp index 2733a745622..98fd209c12a 100644 --- a/src/Processors/QueryPlan/QueryPlan.cpp +++ b/src/Processors/QueryPlan/QueryPlan.cpp @@ -458,39 +458,6 @@ void QueryPlan::explainPipeline(WriteBuffer & buffer, const ExplainPipelineOptio } } -static void updateDataStreams(QueryPlan::Node & root) -{ - class UpdateDataStreams : public QueryPlanVisitor - { - public: - explicit UpdateDataStreams(QueryPlan::Node * root_) : QueryPlanVisitor(root_) { } - - static bool visitTopDownImpl(QueryPlan::Node * /*current_node*/, QueryPlan::Node * /*parent_node*/) { return true; } - - static void visitBottomUpImpl(QueryPlan::Node * current_node, QueryPlan::Node * /*parent_node*/) - { - auto & current_step = *current_node->step; - if (!current_step.canUpdateInputHeader() || current_node->children.empty()) - return; - - for (const auto * child : current_node->children) - { - if (!child->step->hasOutputHeader()) - return; - } - - Headers headers; - headers.reserve(current_node->children.size()); - for (const auto * child : current_node->children) - headers.emplace_back(child->step->getOutputHeader()); - - current_step.updateInputHeaders(std::move(headers)); - } - }; - - UpdateDataStreams(&root).visit(); -} - void QueryPlan::optimize(const QueryPlanOptimizationSettings & optimization_settings) { /// optimization need to be applied before "mergeExpressions" optimization @@ -503,8 +470,6 @@ void QueryPlan::optimize(const QueryPlanOptimizationSettings & optimization_sett QueryPlanOptimizations::optimizeTreeSecondPass(optimization_settings, *root, nodes); if (optimization_settings.build_sets) QueryPlanOptimizations::addStepsToBuildSets(*this, *root, nodes); - - updateDataStreams(*root); } void QueryPlan::explainEstimate(MutableColumns & columns) const diff --git a/src/Processors/QueryPlan/UnionStep.h b/src/Processors/QueryPlan/UnionStep.h index a98d2ef06f3..efb8f51c7a4 100644 --- a/src/Processors/QueryPlan/UnionStep.h +++ b/src/Processors/QueryPlan/UnionStep.h @@ -19,8 +19,6 @@ public: size_t getMaxThreads() const { return max_threads; } - bool canUpdateInputHeader() const override { return true; } - private: void updateOutputHeader() override; diff --git a/src/Processors/Transforms/FilterTransform.cpp b/src/Processors/Transforms/FilterTransform.cpp index cd87019a8e0..20547439414 100644 --- a/src/Processors/Transforms/FilterTransform.cpp +++ b/src/Processors/Transforms/FilterTransform.cpp @@ -1,4 +1,3 @@ -#include #include #include @@ -16,26 +15,6 @@ namespace ErrorCodes extern const int ILLEGAL_TYPE_OF_COLUMN_FOR_FILTER; } -static void replaceFilterToConstant(Block & block, const String & filter_column_name) -{ - ConstantFilterDescription constant_filter_description; - - auto filter_column = block.getPositionByName(filter_column_name); - auto & column_elem = block.safeGetByPosition(filter_column); - - /// Isn't the filter already constant? - if (column_elem.column) - constant_filter_description = ConstantFilterDescription(*column_elem.column); - - if (!constant_filter_description.always_false - && !constant_filter_description.always_true) - { - /// Replace the filter column to a constant with value 1. - FilterDescription filter_description_check(*column_elem.column); - column_elem.column = column_elem.type->createColumnConst(block.rows(), 1u); - } -} - Block FilterTransform::transformHeader( const Block & header, const ActionsDAG * expression, const String & filter_column_name, bool remove_filter_column) { @@ -49,8 +28,6 @@ Block FilterTransform::transformHeader( if (remove_filter_column) result.erase(filter_column_name); - else - replaceFilterToConstant(result, filter_column_name); return result; } @@ -106,10 +83,10 @@ IProcessor::Status FilterTransform::prepare() } -void FilterTransform::removeFilterIfNeed(Chunk & chunk) const +void FilterTransform::removeFilterIfNeed(Columns & columns) const { - if (chunk && remove_filter_column) - chunk.erase(filter_column_position); + if (remove_filter_column) + columns.erase(columns.begin() + filter_column_position); } void FilterTransform::transform(Chunk & chunk) @@ -139,8 +116,8 @@ void FilterTransform::doTransform(Chunk & chunk) if (constant_filter_description.always_true || on_totals) { + removeFilterIfNeed(columns); chunk.setColumns(std::move(columns), num_rows_before_filtration); - removeFilterIfNeed(chunk); return; } @@ -159,8 +136,8 @@ void FilterTransform::doTransform(Chunk & chunk) if (constant_filter_description.always_true) { + removeFilterIfNeed(columns); chunk.setColumns(std::move(columns), num_rows_before_filtration); - removeFilterIfNeed(chunk); return; } @@ -208,35 +185,19 @@ void FilterTransform::doTransform(Chunk & chunk) /// If all the rows pass through the filter. if (num_filtered_rows == num_rows_before_filtration) { - if (!remove_filter_column) - { - /// Replace the column with the filter by a constant. - auto & type = transformed_header.getByPosition(filter_column_position).type; - columns[filter_column_position] = type->createColumnConst(num_filtered_rows, 1u); - } - /// No need to touch the rest of the columns. + removeFilterIfNeed(columns); chunk.setColumns(std::move(columns), num_rows_before_filtration); - removeFilterIfNeed(chunk); return; } /// Filter the rest of the columns. for (size_t i = 0; i < num_columns; ++i) { - const auto & current_type = transformed_header.safeGetByPosition(i).type; auto & current_column = columns[i]; - if (i == filter_column_position) - { - /// The column with filter itself is replaced with a column with a constant `1`, since after filtering, nothing else will remain. - /// NOTE User could pass column with something different than 0 and 1 for filter. - /// Example: - /// SELECT materialize(100) AS x WHERE x - /// will work incorrectly. - current_column = current_type->createColumnConst(num_filtered_rows, 1u); + if (i == filter_column_position && remove_filter_column) continue; - } if (i == first_non_constant_column) continue; @@ -247,8 +208,8 @@ void FilterTransform::doTransform(Chunk & chunk) current_column = filter_description->filter(*current_column, num_filtered_rows); } + removeFilterIfNeed(columns); chunk.setColumns(std::move(columns), num_filtered_rows); - removeFilterIfNeed(chunk); } diff --git a/src/Processors/Transforms/FilterTransform.h b/src/Processors/Transforms/FilterTransform.h index 23c694eed0b..78655bf9f6f 100644 --- a/src/Processors/Transforms/FilterTransform.h +++ b/src/Processors/Transforms/FilterTransform.h @@ -48,7 +48,7 @@ private: bool are_prepared_sets_initialized = false; void doTransform(Chunk & chunk); - void removeFilterIfNeed(Chunk & chunk) const; + void removeFilterIfNeed(Columns & columns) const; }; } From a228e4fa895979ee1d5bf6de71242ece82bc21e6 Mon Sep 17 00:00:00 2001 From: divanik Date: Thu, 24 Oct 2024 13:28:32 +0000 Subject: [PATCH 271/680] Fix issues with tests --- .../DataLakes/DataLakeConfiguration.h | 14 +++++++ .../ObjectStorage/StorageObjectStorage.cpp | 39 +++++++++++++++---- .../ObjectStorage/StorageObjectStorage.h | 11 ++++-- .../registerStorageObjectStorage.cpp | 22 ++++++++++- .../TableFunctionObjectStorage.cpp | 25 ++++-------- .../TableFunctionObjectStorage.h | 9 +++++ 6 files changed, 89 insertions(+), 31 deletions(-) diff --git a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h index d19b7f65640..c01e615acd9 100644 --- a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h +++ b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h @@ -10,6 +10,7 @@ # include # include # include +# include # include # include # include @@ -46,6 +47,18 @@ public: BaseStorageConfiguration::setPartitionColumns(current_metadata->getPartitionColumns()); } + std::optional tryGetTableStructureFromMetadata() const override + { + if (!current_metadata) + return std::nullopt; + auto schema_from_metadata = current_metadata->getTableSchema(); + if (!schema_from_metadata.empty()) + { + return ColumnsDescription(std::move(schema_from_metadata)); + } + return std::nullopt; + } + private: DataLakeMetadataPtr current_metadata; @@ -77,6 +90,7 @@ private: using StorageS3IcebergConfiguration = DataLakeConfiguration; using StorageAzureIcebergConfiguration = DataLakeConfiguration; using StorageLocalIcebergConfiguration = DataLakeConfiguration; +using StorageHDFSIcebergConfiguration = DataLakeConfiguration; using StorageS3DeltaLakeConfiguration = DataLakeConfiguration; using StorageS3HudiConfiguration = DataLakeConfiguration; diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index 86630b897d0..f24f152ecb4 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -14,14 +14,15 @@ #include #include -#include #include -#include -#include #include +#include #include #include -#include +#include +#include +#include +#include "Storages/ColumnsDescription.h" namespace DB @@ -252,6 +253,11 @@ ReadFromFormatInfo StorageObjectStorage::Configuration::prepareReadingFromFormat return DB::prepareReadingFromFormat(requested_columns, storage_snapshot, local_context, supports_subset_of_columns); } +std::optional StorageObjectStorage::Configuration::tryGetTableStructureFromMetadata() const +{ + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Method tryGetTableStructureFromMetadata is not implemented for basic configuration"); +} + void StorageObjectStorage::read( QueryPlan & query_plan, const Names & column_names, @@ -409,6 +415,16 @@ ColumnsDescription StorageObjectStorage::resolveSchemaFromData( std::string & sample_path, const ContextPtr & context) { + if (configuration->isDataLakeConfiguration()) + { + configuration->update(object_storage, context); + auto table_structure = configuration->tryGetTableStructureFromMetadata(); + if (table_structure) + { + return table_structure.value(); + } + } + ObjectInfos read_keys; auto iterator = createReadBufferIterator(object_storage, configuration, format_settings, read_keys, context); auto schema = readSchemaFromFormat(configuration->format, format_settings, *iterator, context); @@ -489,10 +505,17 @@ void StorageObjectStorage::Configuration::initialize( if (configuration.format == "auto") { - configuration.format = FormatFactory::instance().tryGetFormatFromFileName( - configuration.isArchive() - ? configuration.getPathInArchive() - : configuration.getPath()).value_or("auto"); + if (configuration.isDataLakeConfiguration()) + { + configuration.format = "Parquet"; + } + else + { + configuration.format + = FormatFactory::instance() + .tryGetFormatFromFileName(configuration.isArchive() ? configuration.getPathInArchive() : configuration.getPath()) + .value_or("auto"); + } } else FormatFactory::instance().checkFormatName(configuration.format); diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.h b/src/Storages/ObjectStorage/StorageObjectStorage.h index 9781d5dbe6e..21a6cdeba6f 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.h +++ b/src/Storages/ObjectStorage/StorageObjectStorage.h @@ -1,12 +1,13 @@ #pragma once -#include -#include #include -#include +#include #include -#include #include +#include #include +#include +#include +#include "Storages/ColumnsDescription.h" namespace DB { @@ -208,6 +209,8 @@ public: bool supports_subset_of_columns, ContextPtr local_context); + virtual std::optional tryGetTableStructureFromMetadata() const; + String format = "auto"; String compression_method = "auto"; String structure = "auto"; diff --git a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp index 570e888da91..1e231a8e3e4 100644 --- a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp @@ -153,6 +153,7 @@ void registerStorageObjectStorage(StorageFactory & factory) void registerStorageIceberg(StorageFactory & factory) { +#if USE_AWS_S3 factory.registerStorage( "Iceberg", [&](const StorageFactory::Arguments & args) @@ -182,7 +183,8 @@ void registerStorageIceberg(StorageFactory & factory) .supports_schema_inference = true, .source_access_type = AccessType::S3, }); - +#endif +#if USE_AZURE_BLOB_STORAGE factory.registerStorage( "IcebergAzure", [&](const StorageFactory::Arguments & args) @@ -197,7 +199,7 @@ void registerStorageIceberg(StorageFactory & factory) .supports_schema_inference = true, .source_access_type = AccessType::AZURE, }); - +#endif factory.registerStorage( "IcebergLocal", [&](const StorageFactory::Arguments & args) @@ -212,6 +214,22 @@ void registerStorageIceberg(StorageFactory & factory) .supports_schema_inference = true, .source_access_type = AccessType::FILE, }); +#if USE_HDFS + factory.registerStorage( + "IcebergHDFS", + [&](const StorageFactory::Arguments & args) + { + auto configuration = std::make_shared(); + StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); + + return createStorageObjectStorage(args, configuration, args.getLocalContext()); + }, + { + .supports_settings = false, + .supports_schema_inference = true, + .source_access_type = AccessType::HDFS, + }); +#endif } #endif diff --git a/src/TableFunctions/TableFunctionObjectStorage.cpp b/src/TableFunctions/TableFunctionObjectStorage.cpp index ecfc1e462f0..509ef92e8b2 100644 --- a/src/TableFunctions/TableFunctionObjectStorage.cpp +++ b/src/TableFunctions/TableFunctionObjectStorage.cpp @@ -251,6 +251,14 @@ void registerTableFunctionIceberg(TableFunctionFactory & factory) .categories{"DataLake"}}, .allow_readonly = false}); # endif +# if USE_HDFS + factory.registerFunction( + {.documentation + = {.description = R"(The table function can be used to read the Iceberg table stored on HDFS virtual filesystem.)", + .examples{{"icebergHDFS", "SELECT * FROM icebergHDFS(url)", ""}}, + .categories{"DataLake"}}, + .allow_readonly = false}); +# endif factory.registerFunction( {.documentation = {.description = R"(The table function can be used to read the Iceberg table stored locally.)", @@ -297,21 +305,4 @@ void registerDataLakeTableFunctions(TableFunctionFactory & factory) registerTableFunctionHudi(factory); #endif } - -#if USE_AVRO -# if USE_AWS_S3 -template class TableFunctionObjectStorage; -template class TableFunctionObjectStorage; -# endif -# if USE_AZURE_BLOB_STORAGE -template class TableFunctionObjectStorage; -# endif -template class TableFunctionObjectStorage; -#endif -#if USE_AWS_S3 -# if USE_PARQUET -template class TableFunctionObjectStorage; -# endif -template class TableFunctionObjectStorage; -#endif } diff --git a/src/TableFunctions/TableFunctionObjectStorage.h b/src/TableFunctions/TableFunctionObjectStorage.h index 3cf86f982d1..19cd637bd80 100644 --- a/src/TableFunctions/TableFunctionObjectStorage.h +++ b/src/TableFunctions/TableFunctionObjectStorage.h @@ -86,6 +86,12 @@ struct IcebergLocalDefinition static constexpr auto storage_type_name = "Local"; }; +struct IcebergHDFSDefinition +{ + static constexpr auto name = "icebergHDFS"; + static constexpr auto storage_type_name = "HDFS"; +}; + struct DeltaLakeDefinition { static constexpr auto name = "deltaLake"; @@ -184,6 +190,9 @@ using TableFunctionIcebergS3 = TableFunctionObjectStorage; # endif +# if USE_HDFS +using TableFunctionIcebergHDFS = TableFunctionObjectStorage; +# endif using TableFunctionIcebergLocal = TableFunctionObjectStorage; #endif #if USE_AWS_S3 From a3f0d27d23ebf0776304d82be1765cdcb4a122e8 Mon Sep 17 00:00:00 2001 From: divanik Date: Thu, 24 Oct 2024 13:56:26 +0000 Subject: [PATCH 272/680] Resolve some issues --- .../DataLakes/DataLakeConfiguration.h | 32 ++++++++----------- .../DataLakes/DeltaLakeMetadata.cpp | 8 ++--- .../DataLakes/DeltaLakeMetadata.h | 6 ++-- .../ObjectStorage/DataLakes/HudiMetadata.cpp | 2 +- .../ObjectStorage/DataLakes/HudiMetadata.h | 14 +++----- .../DataLakes/IcebergMetadata.cpp | 4 +-- .../ObjectStorage/DataLakes/IcebergMetadata.h | 13 ++++---- .../ObjectStorage/StorageObjectStorage.cpp | 1 + .../ObjectStorage/StorageObjectStorage.h | 2 +- .../registerStorageObjectStorage.cpp | 1 + 10 files changed, 36 insertions(+), 47 deletions(-) diff --git a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h index c01e615acd9..27599452a59 100644 --- a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h +++ b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h @@ -1,23 +1,19 @@ #pragma once -#include "config.h" +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include -#if USE_AVRO - -# include -# include -# include -# include -# include -# include -# include -# include -# include -# include -# include -# include - -# include +#include namespace DB @@ -96,5 +92,3 @@ using StorageS3HudiConfiguration = DataLakeConfigurationdata_files; } - static DataLakeMetadataPtr create(ObjectStoragePtr object_storage, ConfigurationObservePtr configuration, ContextPtr local_context) + static DataLakeMetadataPtr create(ObjectStoragePtr object_storage, ConfigurationObserverPtr configuration, ContextPtr local_context) { return std::make_unique(object_storage, configuration, local_context); } diff --git a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp index 8a93a0ea6d3..40730f6d057 100644 --- a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp +++ b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp @@ -87,7 +87,7 @@ Strings HudiMetadata::getDataFilesImpl() const return result; } -HudiMetadata::HudiMetadata(ObjectStoragePtr object_storage_, ConfigurationObservePtr configuration_, ContextPtr context_) +HudiMetadata::HudiMetadata(ObjectStoragePtr object_storage_, ConfigurationObserverPtr configuration_, ContextPtr context_) : WithContext(context_), object_storage(object_storage_), configuration(configuration_) { } diff --git a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h index b22dfacb0ad..cdab11c4277 100644 --- a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h +++ b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.h @@ -13,14 +13,11 @@ namespace DB class HudiMetadata final : public IDataLakeMetadata, private WithContext { public: - using ConfigurationObservePtr = StorageObjectStorage::ConfigurationObservePtr; + using ConfigurationObserverPtr = StorageObjectStorage::ConfigurationObserverPtr; static constexpr auto name = "Hudi"; - HudiMetadata( - ObjectStoragePtr object_storage_, - ConfigurationObservePtr configuration_, - ContextPtr context_); + HudiMetadata(ObjectStoragePtr object_storage_, ConfigurationObserverPtr configuration_, ContextPtr context_); Strings getDataFiles() const override; @@ -38,17 +35,14 @@ public: && data_files == hudi_metadata->data_files; } - static DataLakeMetadataPtr create( - ObjectStoragePtr object_storage, - ConfigurationObservePtr configuration, - ContextPtr local_context) + static DataLakeMetadataPtr create(ObjectStoragePtr object_storage, ConfigurationObserverPtr configuration, ContextPtr local_context) { return std::make_unique(object_storage, configuration, local_context); } private: const ObjectStoragePtr object_storage; - const ConfigurationObservePtr configuration; + const ConfigurationObserverPtr configuration; mutable Strings data_files; std::unordered_map column_name_to_physical_name; DataLakePartitionColumns partition_columns; diff --git a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp index 379b20ea636..f0a80a41d4e 100644 --- a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp +++ b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.cpp @@ -51,7 +51,7 @@ extern const int UNSUPPORTED_METHOD; IcebergMetadata::IcebergMetadata( ObjectStoragePtr object_storage_, - ConfigurationObservePtr configuration_, + ConfigurationObserverPtr configuration_, DB::ContextPtr context_, Int32 metadata_version_, Int32 format_version_, @@ -383,7 +383,7 @@ std::pair getMetadataFileAndVersion( } DataLakeMetadataPtr -IcebergMetadata::create(ObjectStoragePtr object_storage, ConfigurationObservePtr configuration, ContextPtr local_context) +IcebergMetadata::create(ObjectStoragePtr object_storage, ConfigurationObserverPtr configuration, ContextPtr local_context) { auto configuration_ptr = configuration.lock(); diff --git a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h index 7811bcd8b4b..eb5cac591f2 100644 --- a/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h +++ b/src/Storages/ObjectStorage/DataLakes/IcebergMetadata.h @@ -1,5 +1,7 @@ #pragma once +#include "config.h" + #if USE_AVRO /// StorageIceberg depending on Avro to parse metadata with Avro format. #include @@ -61,13 +63,13 @@ namespace DB class IcebergMetadata : public IDataLakeMetadata, private WithContext { public: - using ConfigurationObservePtr = StorageObjectStorage::ConfigurationObservePtr; + using ConfigurationObserverPtr = StorageObjectStorage::ConfigurationObserverPtr; static constexpr auto name = "Iceberg"; IcebergMetadata( ObjectStoragePtr object_storage_, - ConfigurationObservePtr configuration_, + ConfigurationObserverPtr configuration_, ContextPtr context_, Int32 metadata_version_, Int32 format_version_, @@ -92,16 +94,13 @@ public: return iceberg_metadata && getVersion() == iceberg_metadata->getVersion(); } - static DataLakeMetadataPtr create( - ObjectStoragePtr object_storage, - ConfigurationObservePtr configuration, - ContextPtr local_context); + static DataLakeMetadataPtr create(ObjectStoragePtr object_storage, ConfigurationObserverPtr configuration, ContextPtr local_context); private: size_t getVersion() const { return metadata_version; } const ObjectStoragePtr object_storage; - const ConfigurationObservePtr configuration; + const ConfigurationObserverPtr configuration; Int32 metadata_version; Int32 format_version; String manifest_list_file; diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index f24f152ecb4..a67c1628b6d 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -87,6 +87,7 @@ StorageObjectStorage::StorageObjectStorage( , distributed_processing(distributed_processing_) , log(getLogger(fmt::format("Storage{}({})", configuration->getEngineName(), table_id_.getFullTableName()))) { + configuration_->update(object_storage_, context); ColumnsDescription columns{columns_}; std::string sample_path; diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.h b/src/Storages/ObjectStorage/StorageObjectStorage.h index 21a6cdeba6f..dc461e5861d 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.h +++ b/src/Storages/ObjectStorage/StorageObjectStorage.h @@ -26,7 +26,7 @@ class StorageObjectStorage : public IStorage public: class Configuration; using ConfigurationPtr = std::shared_ptr; - using ConfigurationObservePtr = std::weak_ptr; + using ConfigurationObserverPtr = std::weak_ptr; using ObjectInfo = RelativePathWithMetadata; using ObjectInfoPtr = std::shared_ptr; using ObjectInfos = std::vector; diff --git a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp index 1e231a8e3e4..823556470b0 100644 --- a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp @@ -29,6 +29,7 @@ static std::shared_ptr createStorageObjectStorage( StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, context, false); + // Use format settings from global server context + settings from // the SETTINGS clause of the create query. Settings from current // session and user are ignored. From a457683bd016d83e4544478b3b352daeec53a6f8 Mon Sep 17 00:00:00 2001 From: vdimir Date: Thu, 24 Oct 2024 14:05:40 +0000 Subject: [PATCH 273/680] fix --- src/Processors/QueryPlan/JoinStep.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index 6925d591968..7ade437822e 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -166,6 +166,7 @@ void JoinStep::setJoin(JoinPtr join_, bool swap_streams_) join_algorithm_header.clear(); swap_streams = swap_streams_; join = std::move(join_); + updateOutputHeader(); } void JoinStep::updateOutputHeader() From 1b6979c5cd80666ba6c5164dae23c54b762a0d58 Mon Sep 17 00:00:00 2001 From: divanik Date: Thu, 24 Oct 2024 15:28:57 +0000 Subject: [PATCH 274/680] Correct ifdefs --- .../DataLakes/DataLakeConfiguration.h | 23 ++++++++++++++++--- .../DataLakes/DeltaLakeMetadata.h | 6 +++++ .../ObjectStorage/DataLakes/HudiMetadata.cpp | 9 ++++---- 3 files changed, 30 insertions(+), 8 deletions(-) diff --git a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h index 27599452a59..69968dff942 100644 --- a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h +++ b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h @@ -83,12 +83,29 @@ private: } }; +#if USE_AVRO +# if USE_AWS_S3 using StorageS3IcebergConfiguration = DataLakeConfiguration; +# endif + +# if USE_AZURE_BLOB_STORAGE using StorageAzureIcebergConfiguration = DataLakeConfiguration; -using StorageLocalIcebergConfiguration = DataLakeConfiguration; +# endif + +# if USE_HDFS using StorageHDFSIcebergConfiguration = DataLakeConfiguration; +# endif + +using StorageLocalIcebergConfiguration = DataLakeConfiguration; +#endif + +#if USE_PARQUET +# if USE_AWS_S3 using StorageS3DeltaLakeConfiguration = DataLakeConfiguration; +# endif +#endif + +#if USE_AWS_S3 using StorageS3HudiConfiguration = DataLakeConfiguration; - - +#endif } diff --git a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h index caa637cec75..031d1fb9e96 100644 --- a/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h +++ b/src/Storages/ObjectStorage/DataLakes/DeltaLakeMetadata.h @@ -1,5 +1,9 @@ #pragma once +#include "config.h" + +#if USE_PARQUET + #include #include #include @@ -46,3 +50,5 @@ private: }; } + +#endif diff --git a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp index 40730f6d057..77ef769ed0e 100644 --- a/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp +++ b/src/Storages/ObjectStorage/DataLakes/HudiMetadata.cpp @@ -1,11 +1,10 @@ -#include -#include #include -#include +#include +#include +#include #include #include -#include "config.h" -#include +#include namespace DB { From 8a0c6897f8c349d4a63d1330c226ffcce849df9e Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Thu, 24 Oct 2024 16:21:58 -0400 Subject: [PATCH 275/680] enable enable_job_stack_trace by default --- src/Core/Settings.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 1790697d03e..d3c993250fb 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -2830,7 +2830,7 @@ Limit on size of multipart/form-data content. This setting cannot be parsed from DECLARE(Bool, calculate_text_stack_trace, true, R"( Calculate text stack trace in case of exceptions during query execution. This is the default. It requires symbol lookups that may slow down fuzzing tests when a huge amount of wrong queries are executed. In normal cases, you should not disable this option. )", 0) \ - DECLARE(Bool, enable_job_stack_trace, false, R"( + DECLARE(Bool, enable_job_stack_trace, true, R"( Output stack trace of a job creator when job results in exception )", 0) \ DECLARE(Bool, allow_ddl, true, R"( From e19bf218f69448c9605f269ae7a3894bc24f0003 Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Fri, 25 Oct 2024 06:12:50 +0000 Subject: [PATCH 276/680] Fix 'Unknown executor' when reading from stdin in clickhouse local --- programs/local/LocalServer.cpp | 5 +++++ programs/local/LocalServer.h | 2 +- src/Client/ClientBase.cpp | 3 ++- src/Client/ClientBase.h | 2 ++ .../03031_clickhouse_local_input.reference | 4 +++- .../0_stateless/03031_clickhouse_local_input.sh | 17 ++++++++++++++--- 6 files changed, 27 insertions(+), 6 deletions(-) diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp index b6b67724b0a..4b861d579ab 100644 --- a/programs/local/LocalServer.cpp +++ b/programs/local/LocalServer.cpp @@ -130,6 +130,11 @@ void applySettingsOverridesForLocal(ContextMutablePtr context) context->setSettings(settings); } +LocalServer::LocalServer() +{ + is_local = true; +} + Poco::Util::LayeredConfiguration & LocalServer::getClientConfiguration() { return config(); diff --git a/programs/local/LocalServer.h b/programs/local/LocalServer.h index 7e92e92d345..ced25dbdf90 100644 --- a/programs/local/LocalServer.h +++ b/programs/local/LocalServer.h @@ -23,7 +23,7 @@ namespace DB class LocalServer : public ClientApplicationBase, public Loggers { public: - LocalServer() = default; + LocalServer(); void initialize(Poco::Util::Application & self) override; diff --git a/src/Client/ClientBase.cpp b/src/Client/ClientBase.cpp index 23aa7e841cb..b6223cf6872 100644 --- a/src/Client/ClientBase.cpp +++ b/src/Client/ClientBase.cpp @@ -1748,7 +1748,8 @@ void ClientBase::sendData(Block & sample, const ColumnsDescription & columns_des } else if (!is_interactive) { - sendDataFromStdin(sample, columns_description_for_query, parsed_query); + if (!is_local) + sendDataFromStdin(sample, columns_description_for_query, parsed_query); } else throw Exception(ErrorCodes::NO_DATA_TO_INSERT, "No data to insert"); diff --git a/src/Client/ClientBase.h b/src/Client/ClientBase.h index b06958f1d14..daf3ee7e3e4 100644 --- a/src/Client/ClientBase.h +++ b/src/Client/ClientBase.h @@ -263,6 +263,8 @@ protected: bool is_interactive = false; /// Use either interactive line editing interface or batch mode. bool delayed_interactive = false; + bool is_local = false; /// clickhouse-local, otherwise clickhouse-client + bool echo_queries = false; /// Print queries before execution in batch mode. bool ignore_error = false; /// In case of errors, don't print error message, continue to next query. Only applicable for non-interactive mode. diff --git a/tests/queries/0_stateless/03031_clickhouse_local_input.reference b/tests/queries/0_stateless/03031_clickhouse_local_input.reference index a6feeef100d..529f1832598 100644 --- a/tests/queries/0_stateless/03031_clickhouse_local_input.reference +++ b/tests/queries/0_stateless/03031_clickhouse_local_input.reference @@ -1,4 +1,6 @@ -# foo +# foo (pipe) +foo +# foo (file) foo # !foo # bar diff --git a/tests/queries/0_stateless/03031_clickhouse_local_input.sh b/tests/queries/0_stateless/03031_clickhouse_local_input.sh index e2f9cf48108..540e1203154 100755 --- a/tests/queries/0_stateless/03031_clickhouse_local_input.sh +++ b/tests/queries/0_stateless/03031_clickhouse_local_input.sh @@ -4,17 +4,28 @@ CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CUR_DIR"/../shell_config.sh -tmp_file="$CUR_DIR/$CLICKHOUSE_DATABASE.txt" -echo '# foo' +tmp_file="$CUR_DIR/03031_$CLICKHOUSE_DATABASE.txt" +tmp_input="$CUR_DIR/03031_${CLICKHOUSE_DATABASE}_in.txt" + +echo '# foo (pipe)' $CLICKHOUSE_LOCAL --engine_file_truncate_on_insert=1 -q "insert into function file('$tmp_file', 'LineAsString', 'x String') select * from input('x String') format LineAsString" << "$tmp_input" +$CLICKHOUSE_LOCAL --engine_file_truncate_on_insert=1 -q "insert into function file('$tmp_file', 'LineAsString', 'x String') select * from input('x String') format LineAsString" <"$tmp_input" +cat "$tmp_file" + echo '# !foo' $CLICKHOUSE_LOCAL --engine_file_truncate_on_insert=1 -q "insert into function file('$tmp_file', 'LineAsString', 'x String') select * from input('x String') where x != 'foo' format LineAsString" << Date: Fri, 25 Oct 2024 08:24:30 +0000 Subject: [PATCH 277/680] Also fix 'Input initializer is not set' in another query --- src/Interpreters/InterpreterInsertQuery.cpp | 3 +-- .../0_stateless/03031_clickhouse_local_input.reference | 2 ++ tests/queries/0_stateless/03031_clickhouse_local_input.sh | 4 ++++ 3 files changed, 7 insertions(+), 2 deletions(-) diff --git a/src/Interpreters/InterpreterInsertQuery.cpp b/src/Interpreters/InterpreterInsertQuery.cpp index 80b9d91a248..797895e4a93 100644 --- a/src/Interpreters/InterpreterInsertQuery.cpp +++ b/src/Interpreters/InterpreterInsertQuery.cpp @@ -121,8 +121,7 @@ StoragePtr InterpreterInsertQuery::getTable(ASTInsertQuery & query) if (current_context->getSettingsRef()[Setting::allow_experimental_analyzer]) { - InterpreterSelectQueryAnalyzer interpreter_select(query.select, current_context, select_query_options); - header_block = interpreter_select.getSampleBlock(); + header_block = InterpreterSelectQueryAnalyzer::getSampleBlock(query.select, current_context, select_query_options); } else { diff --git a/tests/queries/0_stateless/03031_clickhouse_local_input.reference b/tests/queries/0_stateless/03031_clickhouse_local_input.reference index 529f1832598..c6e6b743759 100644 --- a/tests/queries/0_stateless/03031_clickhouse_local_input.reference +++ b/tests/queries/0_stateless/03031_clickhouse_local_input.reference @@ -7,3 +7,5 @@ foo bar # defaults bam +# inferred destination table structure +foo diff --git a/tests/queries/0_stateless/03031_clickhouse_local_input.sh b/tests/queries/0_stateless/03031_clickhouse_local_input.sh index 540e1203154..cfd8c2957bb 100755 --- a/tests/queries/0_stateless/03031_clickhouse_local_input.sh +++ b/tests/queries/0_stateless/03031_clickhouse_local_input.sh @@ -28,4 +28,8 @@ echo '# defaults' $CLICKHOUSE_LOCAL --input_format_tsv_empty_as_default=1 --engine_file_truncate_on_insert=1 -q "insert into function file('$tmp_file', 'LineAsString', 'x String') select y from input('x String, y String DEFAULT \\'bam\\'') format TSV" <<<$'foo\t' cat "$tmp_file" +echo '# inferred destination table structure' +$CLICKHOUSE_LOCAL --engine_file_truncate_on_insert=1 -q "insert into function file('$tmp_file', 'TSV') select * from input('x String') format LineAsString" <"$tmp_input" +cat "$tmp_file" + rm -f "${tmp_file:?}" "${tmp_input:?}" From 3da0b2573a5e13c715562d5f8e544480ebf9cc2b Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Fri, 25 Oct 2024 09:12:52 +0000 Subject: [PATCH 278/680] Better fix --- programs/local/LocalServer.cpp | 5 ----- programs/local/LocalServer.h | 2 +- src/Client/ClientBase.cpp | 8 ++++++-- src/Client/ClientBase.h | 2 -- src/Client/IServerConnection.h | 4 ++++ src/Client/LocalConnection.cpp | 5 +++++ src/Client/LocalConnection.h | 2 ++ .../0_stateless/03031_clickhouse_local_input.reference | 4 ++++ tests/queries/0_stateless/03031_clickhouse_local_input.sh | 8 ++++++++ 9 files changed, 30 insertions(+), 10 deletions(-) diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp index 4b861d579ab..b6b67724b0a 100644 --- a/programs/local/LocalServer.cpp +++ b/programs/local/LocalServer.cpp @@ -130,11 +130,6 @@ void applySettingsOverridesForLocal(ContextMutablePtr context) context->setSettings(settings); } -LocalServer::LocalServer() -{ - is_local = true; -} - Poco::Util::LayeredConfiguration & LocalServer::getClientConfiguration() { return config(); diff --git a/programs/local/LocalServer.h b/programs/local/LocalServer.h index ced25dbdf90..7e92e92d345 100644 --- a/programs/local/LocalServer.h +++ b/programs/local/LocalServer.h @@ -23,7 +23,7 @@ namespace DB class LocalServer : public ClientApplicationBase, public Loggers { public: - LocalServer(); + LocalServer() = default; void initialize(Poco::Util::Application & self) override; diff --git a/src/Client/ClientBase.cpp b/src/Client/ClientBase.cpp index b6223cf6872..f5351b94a94 100644 --- a/src/Client/ClientBase.cpp +++ b/src/Client/ClientBase.cpp @@ -1630,6 +1630,11 @@ void ClientBase::sendData(Block & sample, const ColumnsDescription & columns_des if (!parsed_insert_query) return; + /// If it's clickhouse-local, and the input data reading is already baked into the query pipeline, + /// don't read the data again here. + if (!connection->isSendDataNeeded()) + return; + bool have_data_in_stdin = !is_interactive && !stdin_is_a_tty && isStdinNotEmptyAndValid(std_in); if (need_render_progress) @@ -1748,8 +1753,7 @@ void ClientBase::sendData(Block & sample, const ColumnsDescription & columns_des } else if (!is_interactive) { - if (!is_local) - sendDataFromStdin(sample, columns_description_for_query, parsed_query); + sendDataFromStdin(sample, columns_description_for_query, parsed_query); } else throw Exception(ErrorCodes::NO_DATA_TO_INSERT, "No data to insert"); diff --git a/src/Client/ClientBase.h b/src/Client/ClientBase.h index daf3ee7e3e4..b06958f1d14 100644 --- a/src/Client/ClientBase.h +++ b/src/Client/ClientBase.h @@ -263,8 +263,6 @@ protected: bool is_interactive = false; /// Use either interactive line editing interface or batch mode. bool delayed_interactive = false; - bool is_local = false; /// clickhouse-local, otherwise clickhouse-client - bool echo_queries = false; /// Print queries before execution in batch mode. bool ignore_error = false; /// In case of errors, don't print error message, continue to next query. Only applicable for non-interactive mode. diff --git a/src/Client/IServerConnection.h b/src/Client/IServerConnection.h index 6ab4234bca2..fe69be8788a 100644 --- a/src/Client/IServerConnection.h +++ b/src/Client/IServerConnection.h @@ -109,6 +109,10 @@ public: /// Send block of data; if name is specified, server will write it to external (temporary) table of that name. virtual void sendData(const Block & block, const String & name, bool scalar) = 0; + /// Whether the client needs to read and send the data for the INSERT. + /// False if the server will read the data through other means (in particular if clickhouse-local added input reading step directly into the query pipeline). + virtual bool isSendDataNeeded() const { return true; } + /// Send all contents of external (temporary) tables. virtual void sendExternalTablesData(ExternalTablesData & data) = 0; diff --git a/src/Client/LocalConnection.cpp b/src/Client/LocalConnection.cpp index e4915a77c83..4ca209c29c7 100644 --- a/src/Client/LocalConnection.cpp +++ b/src/Client/LocalConnection.cpp @@ -328,6 +328,11 @@ void LocalConnection::sendData(const Block & block, const String &, bool) sendProfileEvents(); } +bool LocalConnection::isSendDataNeeded() const +{ + return !state || state->input_pipeline == nullptr; +} + void LocalConnection::sendCancel() { state->is_cancelled = true; diff --git a/src/Client/LocalConnection.h b/src/Client/LocalConnection.h index b424c5b5aa3..a70ed6ffa7e 100644 --- a/src/Client/LocalConnection.h +++ b/src/Client/LocalConnection.h @@ -120,6 +120,8 @@ public: void sendData(const Block & block, const String & name/* = "" */, bool scalar/* = false */) override; + bool isSendDataNeeded() const override; + void sendExternalTablesData(ExternalTablesData &) override; void sendMergeTreeReadTaskResponse(const ParallelReadResponse & response) override; diff --git a/tests/queries/0_stateless/03031_clickhouse_local_input.reference b/tests/queries/0_stateless/03031_clickhouse_local_input.reference index c6e6b743759..bbb57da94ce 100644 --- a/tests/queries/0_stateless/03031_clickhouse_local_input.reference +++ b/tests/queries/0_stateless/03031_clickhouse_local_input.reference @@ -9,3 +9,7 @@ bar bam # inferred destination table structure foo +# direct +foo +# infile +foo diff --git a/tests/queries/0_stateless/03031_clickhouse_local_input.sh b/tests/queries/0_stateless/03031_clickhouse_local_input.sh index cfd8c2957bb..f271a5184fd 100755 --- a/tests/queries/0_stateless/03031_clickhouse_local_input.sh +++ b/tests/queries/0_stateless/03031_clickhouse_local_input.sh @@ -32,4 +32,12 @@ echo '# inferred destination table structure' $CLICKHOUSE_LOCAL --engine_file_truncate_on_insert=1 -q "insert into function file('$tmp_file', 'TSV') select * from input('x String') format LineAsString" <"$tmp_input" cat "$tmp_file" +echo '# direct' +$CLICKHOUSE_LOCAL --engine_file_truncate_on_insert=1 -q "insert into function file('$tmp_file', 'LineAsString', 'x String') format LineAsString" <"$tmp_input" +cat "$tmp_file" + +echo '# infile' +$CLICKHOUSE_LOCAL --engine_file_truncate_on_insert=1 -q "insert into function file('$tmp_file', 'LineAsString', 'x String') from infile '$tmp_input' format LineAsString" +cat "$tmp_file" + rm -f "${tmp_file:?}" "${tmp_input:?}" From 45e23584f4cee58bf9c0f0612e4799076c0d21e8 Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Fri, 25 Oct 2024 09:15:53 +0000 Subject: [PATCH 279/680] Comment --- src/Client/ClientBase.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Client/ClientBase.cpp b/src/Client/ClientBase.cpp index f5351b94a94..6475d682b65 100644 --- a/src/Client/ClientBase.cpp +++ b/src/Client/ClientBase.cpp @@ -1631,7 +1631,7 @@ void ClientBase::sendData(Block & sample, const ColumnsDescription & columns_des return; /// If it's clickhouse-local, and the input data reading is already baked into the query pipeline, - /// don't read the data again here. + /// don't read the data again here. This happens in some cases (e.g. input() table function) but not others (e.g. INFILE). if (!connection->isSendDataNeeded()) return; From 31490438d95f514e8ff285b80345c55872b2b485 Mon Sep 17 00:00:00 2001 From: divanik Date: Fri, 25 Oct 2024 11:09:03 +0000 Subject: [PATCH 280/680] Corrected smoe ifdef issues --- .../registerStorageObjectStorage.cpp | 31 ++++++++++--------- 1 file changed, 17 insertions(+), 14 deletions(-) diff --git a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp index 823556470b0..b0122de3bf7 100644 --- a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp @@ -201,20 +201,6 @@ void registerStorageIceberg(StorageFactory & factory) .source_access_type = AccessType::AZURE, }); #endif - factory.registerStorage( - "IcebergLocal", - [&](const StorageFactory::Arguments & args) - { - auto configuration = std::make_shared(); - StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); - - return createStorageObjectStorage(args, configuration, args.getLocalContext()); - }, - { - .supports_settings = false, - .supports_schema_inference = true, - .source_access_type = AccessType::FILE, - }); #if USE_HDFS factory.registerStorage( "IcebergHDFS", @@ -231,10 +217,26 @@ void registerStorageIceberg(StorageFactory & factory) .source_access_type = AccessType::HDFS, }); #endif + factory.registerStorage( + "IcebergLocal", + [&](const StorageFactory::Arguments & args) + { + auto configuration = std::make_shared(); + StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, args.getLocalContext(), false); + + return createStorageObjectStorage(args, configuration, args.getLocalContext()); + }, + { + .supports_settings = false, + .supports_schema_inference = true, + .source_access_type = AccessType::FILE, + }); } #endif + +#if USE_AWS_S3 #if USE_PARQUET void registerStorageDeltaLake(StorageFactory & factory) { @@ -272,4 +274,5 @@ void registerStorageHudi(StorageFactory & factory) .source_access_type = AccessType::S3, }); } +#endif } From ca040906c3bca0e283fc5df57451d4d0805336b3 Mon Sep 17 00:00:00 2001 From: divanik Date: Fri, 25 Oct 2024 13:37:12 +0000 Subject: [PATCH 281/680] Fix some ifdef issues --- .../DataLakes/DataLakeConfiguration.h | 8 +++--- .../registerStorageObjectStorage.cpp | 10 +++---- src/Storages/registerStorages.cpp | 3 +- .../TableFunctionObjectStorage.cpp | 28 +++++++++---------- 4 files changed, 24 insertions(+), 25 deletions(-) diff --git a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h index 69968dff942..866ef24aa91 100644 --- a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h +++ b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h @@ -84,15 +84,15 @@ private: }; #if USE_AVRO -# if USE_AWS_S3 +#if USE_AWS_S3 using StorageS3IcebergConfiguration = DataLakeConfiguration; # endif -# if USE_AZURE_BLOB_STORAGE +#if USE_AZURE_BLOB_STORAGE using StorageAzureIcebergConfiguration = DataLakeConfiguration; # endif -# if USE_HDFS +#if USE_HDFS using StorageHDFSIcebergConfiguration = DataLakeConfiguration; # endif @@ -100,7 +100,7 @@ using StorageLocalIcebergConfiguration = DataLakeConfiguration; # endif #endif diff --git a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp index b0122de3bf7..cb1826b2976 100644 --- a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp @@ -11,8 +11,6 @@ namespace DB { -#if USE_AWS_S3 || USE_AZURE_BLOB_STORAGE || USE_HDFS - namespace ErrorCodes { extern const int BAD_ARGUMENTS; @@ -65,8 +63,6 @@ static std::shared_ptr createStorageObjectStorage( partition_by); } -#endif - #if USE_AZURE_BLOB_STORAGE void registerStorageAzure(StorageFactory & factory) { @@ -236,10 +232,10 @@ void registerStorageIceberg(StorageFactory & factory) #endif -#if USE_AWS_S3 #if USE_PARQUET void registerStorageDeltaLake(StorageFactory & factory) { +#if USE_AWS_S3 factory.registerStorage( "DeltaLake", [&](const StorageFactory::Arguments & args) @@ -254,11 +250,13 @@ void registerStorageDeltaLake(StorageFactory & factory) .supports_schema_inference = true, .source_access_type = AccessType::S3, }); +#endif } #endif void registerStorageHudi(StorageFactory & factory) { +#if USE_AWS_S3 factory.registerStorage( "Hudi", [&](const StorageFactory::Arguments & args) @@ -273,6 +271,6 @@ void registerStorageHudi(StorageFactory & factory) .supports_schema_inference = true, .source_access_type = AccessType::S3, }); -} #endif } +} diff --git a/src/Storages/registerStorages.cpp b/src/Storages/registerStorages.cpp index cfd406ccbe2..4eb90955a6c 100644 --- a/src/Storages/registerStorages.cpp +++ b/src/Storages/registerStorages.cpp @@ -41,10 +41,11 @@ void registerStorageS3Queue(StorageFactory & factory); #if USE_PARQUET void registerStorageDeltaLake(StorageFactory & factory); #endif +#endif + #if USE_AVRO void registerStorageIceberg(StorageFactory & factory); #endif -#endif #if USE_AZURE_BLOB_STORAGE void registerStorageAzureQueue(StorageFactory & factory); diff --git a/src/TableFunctions/TableFunctionObjectStorage.cpp b/src/TableFunctions/TableFunctionObjectStorage.cpp index 509ef92e8b2..66c90b15c0b 100644 --- a/src/TableFunctions/TableFunctionObjectStorage.cpp +++ b/src/TableFunctions/TableFunctionObjectStorage.cpp @@ -228,7 +228,7 @@ template class TableFunctionObjectStorage( {.documentation = {.description = R"(The table function can be used to read the Iceberg table stored on S3 object store. Alias to icebergS3)", @@ -242,23 +242,23 @@ void registerTableFunctionIceberg(TableFunctionFactory & factory) .categories{"DataLake"}}, .allow_readonly = false}); -# endif -# if USE_AZURE_BLOB_STORAGE +#endif +#if USE_AZURE_BLOB_STORAGE factory.registerFunction( {.documentation = {.description = R"(The table function can be used to read the Iceberg table stored on Azure object store.)", .examples{{"icebergAzure", "SELECT * FROM icebergAzure(url, access_key_id, secret_access_key)", ""}}, .categories{"DataLake"}}, .allow_readonly = false}); -# endif -# if USE_HDFS +#endif +#if USE_HDFS factory.registerFunction( {.documentation = {.description = R"(The table function can be used to read the Iceberg table stored on HDFS virtual filesystem.)", .examples{{"icebergHDFS", "SELECT * FROM icebergHDFS(url)", ""}}, .categories{"DataLake"}}, .allow_readonly = false}); -# endif +#endif factory.registerFunction( {.documentation = {.description = R"(The table function can be used to read the Iceberg table stored locally.)", @@ -268,29 +268,31 @@ void registerTableFunctionIceberg(TableFunctionFactory & factory) } #endif -#if USE_AWS_S3 -# if USE_PARQUET +#if USE_PARQUET void registerTableFunctionDeltaLake(TableFunctionFactory & factory) { +#if USE_AWS_S3 factory.registerFunction( {.documentation = {.description = R"(The table function can be used to read the DeltaLake table stored on object store.)", .examples{{"deltaLake", "SELECT * FROM deltaLake(url, access_key_id, secret_access_key)", ""}}, .categories{"DataLake"}}, .allow_readonly = false}); +#endif } -# endif +#endif void registerTableFunctionHudi(TableFunctionFactory & factory) { +#if USE_AWS_S3 factory.registerFunction( {.documentation = {.description = R"(The table function can be used to read the Hudi table stored on object store.)", .examples{{"hudi", "SELECT * FROM hudi(url, access_key_id, secret_access_key)", ""}}, .categories{"DataLake"}}, .allow_readonly = false}); -} #endif +} void registerDataLakeTableFunctions(TableFunctionFactory & factory) { @@ -298,11 +300,9 @@ void registerDataLakeTableFunctions(TableFunctionFactory & factory) #if USE_AVRO registerTableFunctionIceberg(factory); #endif -#if USE_AWS_S3 -# if USE_PARQUET +#if USE_PARQUET registerTableFunctionDeltaLake(factory); -# endif - registerTableFunctionHudi(factory); #endif + registerTableFunctionHudi(factory); } } From a7b23292f962eada087b2b7518c231b57ca71493 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Fri, 25 Oct 2024 17:58:43 +0000 Subject: [PATCH 282/680] add staleness to sql --- src/Analyzer/QueryTreeBuilder.cpp | 2 ++ src/Analyzer/Resolve/QueryAnalyzer.cpp | 43 ++++++++++++++++++++++-- src/Analyzer/Resolve/QueryAnalyzer.h | 3 +- src/Analyzer/SortNode.cpp | 8 +++++ src/Analyzer/SortNode.h | 21 +++++++++++- src/Parsers/ASTOrderByElement.cpp | 5 +++ src/Parsers/ASTOrderByElement.h | 3 ++ src/Parsers/CommonParsers.h | 1 + src/Parsers/ExpressionElementParsers.cpp | 6 ++++ src/Planner/Planner.cpp | 3 ++ src/Planner/PlannerActionsVisitor.cpp | 3 ++ src/Planner/PlannerSorting.cpp | 24 +++++++++++-- 12 files changed, 115 insertions(+), 7 deletions(-) diff --git a/src/Analyzer/QueryTreeBuilder.cpp b/src/Analyzer/QueryTreeBuilder.cpp index 39c59d27e2c..d3c88d39213 100644 --- a/src/Analyzer/QueryTreeBuilder.cpp +++ b/src/Analyzer/QueryTreeBuilder.cpp @@ -498,6 +498,8 @@ QueryTreeNodePtr QueryTreeBuilder::buildSortList(const ASTPtr & order_by_express sort_node->getFillTo() = buildExpression(order_by_element.getFillTo(), context); if (order_by_element.getFillStep()) sort_node->getFillStep() = buildExpression(order_by_element.getFillStep(), context); + if (order_by_element.getFillStaleness()) + sort_node->getFillStaleness() = buildExpression(order_by_element.getFillStaleness(), context); list_node->getNodes().push_back(std::move(sort_node)); } diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index 381edee607d..ab29373f5fb 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -432,8 +432,13 @@ ProjectionName QueryAnalyzer::calculateWindowProjectionName(const QueryTreeNodeP return buffer.str(); } -ProjectionName QueryAnalyzer::calculateSortColumnProjectionName(const QueryTreeNodePtr & sort_column_node, const ProjectionName & sort_expression_projection_name, - const ProjectionName & fill_from_expression_projection_name, const ProjectionName & fill_to_expression_projection_name, const ProjectionName & fill_step_expression_projection_name) +ProjectionName QueryAnalyzer::calculateSortColumnProjectionName( + const QueryTreeNodePtr & sort_column_node, + const ProjectionName & sort_expression_projection_name, + const ProjectionName & fill_from_expression_projection_name, + const ProjectionName & fill_to_expression_projection_name, + const ProjectionName & fill_step_expression_projection_name, + const ProjectionName & fill_staleness_expression_projection_name) { auto & sort_node_typed = sort_column_node->as(); @@ -463,6 +468,9 @@ ProjectionName QueryAnalyzer::calculateSortColumnProjectionName(const QueryTreeN if (sort_node_typed.hasFillStep()) sort_column_projection_name_buffer << " STEP " << fill_step_expression_projection_name; + + if (sort_node_typed.hasFillStaleness()) + sort_column_projection_name_buffer << " STALENESS " << fill_staleness_expression_projection_name; } return sort_column_projection_name_buffer.str(); @@ -3993,6 +4001,7 @@ ProjectionNames QueryAnalyzer::resolveSortNodeList(QueryTreeNodePtr & sort_node_ ProjectionNames fill_from_expression_projection_names; ProjectionNames fill_to_expression_projection_names; ProjectionNames fill_step_expression_projection_names; + ProjectionNames fill_staleness_expression_projection_names; auto & sort_node_list_typed = sort_node_list->as(); for (auto & node : sort_node_list_typed.getNodes()) @@ -4083,11 +4092,38 @@ ProjectionNames QueryAnalyzer::resolveSortNodeList(QueryTreeNodePtr & sort_node_ fill_step_expression_projection_names_size); } + if (sort_node.hasFillStaleness()) + { + fill_staleness_expression_projection_names = resolveExpressionNode(sort_node.getFillStaleness(), scope, false /*allow_lambda_expression*/, false /*allow_table_expression*/); + + const auto * constant_node = sort_node.getFillStaleness()->as(); + if (!constant_node) + throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, + "Sort FILL STALENESS expression must be constant with numeric or interval type. Actual {}. In scope {}", + sort_node.getFillStaleness()->formatASTForErrorMessage(), + scope.scope_node->formatASTForErrorMessage()); + + bool is_number = isColumnedAsNumber(constant_node->getResultType()); + bool is_interval = WhichDataType(constant_node->getResultType()).isInterval(); + if (!is_number && !is_interval) + throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, + "Sort FILL STALENESS expression must be constant with numeric or interval type. Actual {}. In scope {}", + sort_node.getFillStaleness()->formatASTForErrorMessage(), + scope.scope_node->formatASTForErrorMessage()); + + size_t fill_staleness_expression_projection_names_size = fill_staleness_expression_projection_names.size(); + if (fill_staleness_expression_projection_names_size != 1) + throw Exception(ErrorCodes::LOGICAL_ERROR, + "Sort FILL STALENESS expression expected 1 projection name. Actual {}", + fill_staleness_expression_projection_names_size); + } + auto sort_column_projection_name = calculateSortColumnProjectionName(node, sort_expression_projection_names[0], fill_from_expression_projection_names.empty() ? "" : fill_from_expression_projection_names.front(), fill_to_expression_projection_names.empty() ? "" : fill_to_expression_projection_names.front(), - fill_step_expression_projection_names.empty() ? "" : fill_step_expression_projection_names.front()); + fill_step_expression_projection_names.empty() ? "" : fill_step_expression_projection_names.front(), + fill_staleness_expression_projection_names.empty() ? "" : fill_staleness_expression_projection_names.front()); result_projection_names.push_back(std::move(sort_column_projection_name)); @@ -4095,6 +4131,7 @@ ProjectionNames QueryAnalyzer::resolveSortNodeList(QueryTreeNodePtr & sort_node_ fill_from_expression_projection_names.clear(); fill_to_expression_projection_names.clear(); fill_step_expression_projection_names.clear(); + fill_staleness_expression_projection_names.clear(); } return result_projection_names; diff --git a/src/Analyzer/Resolve/QueryAnalyzer.h b/src/Analyzer/Resolve/QueryAnalyzer.h index 0d4309843e6..d24bede561e 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.h +++ b/src/Analyzer/Resolve/QueryAnalyzer.h @@ -140,7 +140,8 @@ private: const ProjectionName & sort_expression_projection_name, const ProjectionName & fill_from_expression_projection_name, const ProjectionName & fill_to_expression_projection_name, - const ProjectionName & fill_step_expression_projection_name); + const ProjectionName & fill_step_expression_projection_name, + const ProjectionName & fill_staleness_expression_projection_name); QueryTreeNodePtr tryGetLambdaFromSQLUserDefinedFunctions(const std::string & function_name, ContextPtr context); diff --git a/src/Analyzer/SortNode.cpp b/src/Analyzer/SortNode.cpp index e891046626a..42c010e4784 100644 --- a/src/Analyzer/SortNode.cpp +++ b/src/Analyzer/SortNode.cpp @@ -69,6 +69,12 @@ void SortNode::dumpTreeImpl(WriteBuffer & buffer, FormatState & format_state, si buffer << '\n' << std::string(indent + 2, ' ') << "FILL STEP\n"; getFillStep()->dumpTreeImpl(buffer, format_state, indent + 4); } + + if (hasFillStaleness()) + { + buffer << '\n' << std::string(indent + 2, ' ') << "FILL STALENESS\n"; + getFillStaleness()->dumpTreeImpl(buffer, format_state, indent + 4); + } } bool SortNode::isEqualImpl(const IQueryTreeNode & rhs, CompareOptions) const @@ -132,6 +138,8 @@ ASTPtr SortNode::toASTImpl(const ConvertToASTOptions & options) const result->setFillTo(getFillTo()->toAST(options)); if (hasFillStep()) result->setFillStep(getFillStep()->toAST(options)); + if (hasFillStaleness()) + result->setFillStaleness(getFillStaleness()->toAST(options)); return result; } diff --git a/src/Analyzer/SortNode.h b/src/Analyzer/SortNode.h index 0ebdde61912..d9086dc9ed7 100644 --- a/src/Analyzer/SortNode.h +++ b/src/Analyzer/SortNode.h @@ -105,6 +105,24 @@ public: return children[fill_step_child_index]; } + /// Returns true if sort node has fill step, false otherwise + bool hasFillStaleness() const + { + return children[fill_staleness_child_index] != nullptr; + } + + /// Get fill step + const QueryTreeNodePtr & getFillStaleness() const + { + return children[fill_staleness_child_index]; + } + + /// Get fill step + QueryTreeNodePtr & getFillStaleness() + { + return children[fill_staleness_child_index]; + } + /// Get collator const std::shared_ptr & getCollator() const { @@ -144,7 +162,8 @@ private: static constexpr size_t fill_from_child_index = 1; static constexpr size_t fill_to_child_index = 2; static constexpr size_t fill_step_child_index = 3; - static constexpr size_t children_size = fill_step_child_index + 1; + static constexpr size_t fill_staleness_child_index = 4; + static constexpr size_t children_size = fill_staleness_child_index + 1; SortDirection sort_direction = SortDirection::ASCENDING; std::optional nulls_sort_direction; diff --git a/src/Parsers/ASTOrderByElement.cpp b/src/Parsers/ASTOrderByElement.cpp index 09193a8b5e1..d87c296d398 100644 --- a/src/Parsers/ASTOrderByElement.cpp +++ b/src/Parsers/ASTOrderByElement.cpp @@ -54,6 +54,11 @@ void ASTOrderByElement::formatImpl(const FormatSettings & settings, FormatState settings.ostr << (settings.hilite ? hilite_keyword : "") << " STEP " << (settings.hilite ? hilite_none : ""); fill_step->formatImpl(settings, state, frame); } + if (auto fill_staleness = getFillStaleness()) + { + settings.ostr << (settings.hilite ? hilite_keyword : "") << " STALENESS " << (settings.hilite ? hilite_none : ""); + fill_staleness->formatImpl(settings, state, frame); + } } } diff --git a/src/Parsers/ASTOrderByElement.h b/src/Parsers/ASTOrderByElement.h index 6edf84d7bde..4dc35dac217 100644 --- a/src/Parsers/ASTOrderByElement.h +++ b/src/Parsers/ASTOrderByElement.h @@ -18,6 +18,7 @@ private: FILL_FROM, FILL_TO, FILL_STEP, + FILL_STALENESS, }; public: @@ -32,12 +33,14 @@ public: void setFillFrom(ASTPtr node) { setChild(Child::FILL_FROM, node); } void setFillTo(ASTPtr node) { setChild(Child::FILL_TO, node); } void setFillStep(ASTPtr node) { setChild(Child::FILL_STEP, node); } + void setFillStaleness(ASTPtr node) { setChild(Child::FILL_STALENESS, node); } /** Collation for locale-specific string comparison. If empty, then sorting done by bytes. */ ASTPtr getCollation() const { return getChild(Child::COLLATION); } ASTPtr getFillFrom() const { return getChild(Child::FILL_FROM); } ASTPtr getFillTo() const { return getChild(Child::FILL_TO); } ASTPtr getFillStep() const { return getChild(Child::FILL_STEP); } + ASTPtr getFillStaleness() const { return getChild(Child::FILL_STALENESS); } String getID(char) const override { return "OrderByElement"; } diff --git a/src/Parsers/CommonParsers.h b/src/Parsers/CommonParsers.h index 8ea9fb12b86..c10e4879214 100644 --- a/src/Parsers/CommonParsers.h +++ b/src/Parsers/CommonParsers.h @@ -541,6 +541,7 @@ namespace DB MR_MACROS(YY, "YY") \ MR_MACROS(YYYY, "YYYY") \ MR_MACROS(ZKPATH, "ZKPATH") \ + MR_MACROS(STALENESS, "STALENESS") \ /// The list of keywords where underscore is intentional #define APPLY_FOR_PARSER_KEYWORDS_WITH_UNDERSCORES(MR_MACROS) \ diff --git a/src/Parsers/ExpressionElementParsers.cpp b/src/Parsers/ExpressionElementParsers.cpp index 31efcb16f02..ad062d27a37 100644 --- a/src/Parsers/ExpressionElementParsers.cpp +++ b/src/Parsers/ExpressionElementParsers.cpp @@ -2178,6 +2178,7 @@ bool ParserOrderByElement::parseImpl(Pos & pos, ASTPtr & node, Expected & expect ParserKeyword from(Keyword::FROM); ParserKeyword to(Keyword::TO); ParserKeyword step(Keyword::STEP); + ParserKeyword staleness(Keyword::STALENESS); ParserStringLiteral collate_locale_parser; ParserExpressionWithOptionalAlias exp_parser(false); @@ -2219,6 +2220,7 @@ bool ParserOrderByElement::parseImpl(Pos & pos, ASTPtr & node, Expected & expect ASTPtr fill_from; ASTPtr fill_to; ASTPtr fill_step; + ASTPtr fill_staleness; if (with_fill.ignore(pos, expected)) { has_with_fill = true; @@ -2230,6 +2232,9 @@ bool ParserOrderByElement::parseImpl(Pos & pos, ASTPtr & node, Expected & expect if (step.ignore(pos, expected) && !exp_parser.parse(pos, fill_step, expected)) return false; + + if (staleness.ignore(pos, expected) && !exp_parser.parse(pos, fill_staleness, expected)) + return false; } auto elem = std::make_shared(); @@ -2244,6 +2249,7 @@ bool ParserOrderByElement::parseImpl(Pos & pos, ASTPtr & node, Expected & expect elem->setFillFrom(fill_from); elem->setFillTo(fill_to); elem->setFillStep(fill_step); + elem->setFillStaleness(fill_staleness); node = elem; diff --git a/src/Planner/Planner.cpp b/src/Planner/Planner.cpp index 8d3c75fdabb..f1c752aecd0 100644 --- a/src/Planner/Planner.cpp +++ b/src/Planner/Planner.cpp @@ -847,6 +847,9 @@ void addWithFillStepIfNeeded(QueryPlan & query_plan, interpolate_description = std::make_shared(std::move(interpolate_actions_dag), empty_aliases); } + if (interpolate_description) + LOG_DEBUG(getLogger("addWithFillStepIfNeeded"), "InterpolateDescription: {}", interpolate_description->actions.dumpDAG()); + const auto & query_context = planner_context->getQueryContext(); const Settings & settings = query_context->getSettingsRef(); auto filling_step = std::make_unique( diff --git a/src/Planner/PlannerActionsVisitor.cpp b/src/Planner/PlannerActionsVisitor.cpp index aea304e0ecc..aa233109fa9 100644 --- a/src/Planner/PlannerActionsVisitor.cpp +++ b/src/Planner/PlannerActionsVisitor.cpp @@ -391,6 +391,9 @@ public: if (sort_node.hasFillStep()) buffer << " STEP " << calculateActionNodeName(sort_node.getFillStep()); + + if (sort_node.hasFillStaleness()) + buffer << " STALENESS " << calculateActionNodeName(sort_node.getFillStaleness()); } if (i + 1 != order_by_nodes_size) diff --git a/src/Planner/PlannerSorting.cpp b/src/Planner/PlannerSorting.cpp index af51afdef13..0a33e2f0828 100644 --- a/src/Planner/PlannerSorting.cpp +++ b/src/Planner/PlannerSorting.cpp @@ -43,7 +43,7 @@ std::pair extractWithFillValue(const QueryTreeNodePtr & node return result; } -std::pair> extractWithFillStepValue(const QueryTreeNodePtr & node) +std::pair> extractWithFillValueWithIntervalKind(const QueryTreeNodePtr & node) { const auto & constant_node = node->as(); @@ -77,7 +77,7 @@ FillColumnDescription extractWithFillDescription(const SortNode & sort_node) if (sort_node.hasFillStep()) { - auto extract_result = extractWithFillStepValue(sort_node.getFillStep()); + auto extract_result = extractWithFillValueWithIntervalKind(sort_node.getFillStep()); fill_column_description.fill_step = std::move(extract_result.first); fill_column_description.step_kind = std::move(extract_result.second); } @@ -87,10 +87,30 @@ FillColumnDescription extractWithFillDescription(const SortNode & sort_node) fill_column_description.fill_step = Field(direction_value); } + if (sort_node.getFillStaleness()) + { + auto extract_result = extractWithFillValueWithIntervalKind(sort_node.getFillStaleness()); + fill_column_description.fill_staleness = std::move(extract_result.first); + fill_column_description.staleness_kind = std::move(extract_result.second); + } + + /////////////////////////////////// + if (applyVisitor(FieldVisitorAccurateEquals(), fill_column_description.fill_step, Field{0})) throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, "WITH FILL STEP value cannot be zero"); + if (sort_node.hasFillStaleness()) + { + if (sort_node.hasFillFrom()) + throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, + "WITH FILL STALENESS cannot be used together with WITH FILL FROM"); + + if (applyVisitor(FieldVisitorAccurateLessOrEqual(), fill_column_description.fill_staleness, Field{0})) + throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, + "WITH FILL STALENESS value cannot be less or equal zero"); + } + if (sort_node.getSortDirection() == SortDirection::ASCENDING) { if (applyVisitor(FieldVisitorAccurateLess(), fill_column_description.fill_step, Field{0})) From c58afb753c3a9b394f6f88cc8fad3e13897c5e57 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 27 Oct 2024 00:29:36 +0200 Subject: [PATCH 283/680] Retry more errors from S3 --- src/IO/S3/Client.cpp | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/src/IO/S3/Client.cpp b/src/IO/S3/Client.cpp index 9a0eccd8783..088087458c7 100644 --- a/src/IO/S3/Client.cpp +++ b/src/IO/S3/Client.cpp @@ -645,7 +645,7 @@ Client::doRequestWithRetryNetworkErrors(RequestType & request, RequestFn request try { /// S3 does retries network errors actually. - /// But it is matter when errors occur. + /// But it does matter when errors occur. /// This code retries a specific case when /// network error happens when XML document is being read from the response body. /// Hence, the response body is a stream, network errors are possible at reading. @@ -656,8 +656,9 @@ Client::doRequestWithRetryNetworkErrors(RequestType & request, RequestFn request /// Requests that expose the response stream as an answer are not retried with that code. E.g. GetObject. return request_fn_(request_); } - catch (Poco::Net::ConnectionResetException &) + catch (Poco::Net::NetException &) { + /// This includes "connection reset", "malformed message", and possibly other exceptions. if constexpr (IsReadMethod) { From 8807fe3bb5ff125e3a907354757552957e52b646 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 28 Oct 2024 00:57:13 +0100 Subject: [PATCH 284/680] Better log messages --- src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp index 8b3c7bdf3fb..c0464946752 100644 --- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp +++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp @@ -254,7 +254,8 @@ MergeTreeDataMergerMutator::PartitionIdsHint MergeTreeDataMergerMutator::getPart if (status == SelectPartsDecision::SELECTED) res.insert(all_partition_ids[i]); else - LOG_TEST(log, "Nothing to merge in partition {}: {}", all_partition_ids[i], out_disable_reason.text); + LOG_TEST(log, "Nothing to merge in partition {} with max_total_size_to_merge = {} (looked up {} ranges): {}", + all_partition_ids[i], ReadableSize(max_total_size_to_merge), ranges_per_partition[i].size(), out_disable_reason.text); } String best_partition_id_to_optimize = getBestPartitionToOptimizeEntire(info.partitions_info); From 07508cb3819a89fec7e63604e2de64ff1bd4904a Mon Sep 17 00:00:00 2001 From: divanik Date: Mon, 28 Oct 2024 11:47:01 +0000 Subject: [PATCH 285/680] Handle some problems with tests --- src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp | 3 +-- src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h | 3 +++ src/Storages/ObjectStorage/StorageObjectStorage.cpp | 3 ++- src/Storages/ObjectStorage/registerStorageObjectStorage.cpp | 3 ++- 4 files changed, 8 insertions(+), 4 deletions(-) diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp index cd36429d0a2..4e6d0d985dd 100644 --- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp +++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp @@ -500,8 +500,7 @@ void S3ObjectStorage::applyNewSettings( } auto current_settings = s3_settings.get(); - if (options.allow_client_change - && (current_settings->auth_settings.hasUpdates(modified_settings->auth_settings) || for_disk_s3)) + if (options.allow_client_change && (current_settings->auth_settings.hasUpdates(modified_settings->auth_settings) || for_disk_s3)) { auto new_client = getClient(uri, *modified_settings, context, for_disk_s3); client.set(std::move(new_client)); diff --git a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h index 866ef24aa91..18ff6d93c46 100644 --- a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h +++ b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h @@ -30,10 +30,13 @@ public: bool isDataLakeConfiguration() const override { return true; } + bool isStaticConfiguration() const override { return false; } + std::string getEngineName() const override { return DataLakeMetadata::name; } void update(ObjectStoragePtr object_storage, ContextPtr local_context) override { + BaseStorageConfiguration::update(object_storage, local_context); auto new_metadata = DataLakeMetadata::create(object_storage, weak_from_this(), local_context); if (current_metadata && *current_metadata == *new_metadata) return; diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index a67c1628b6d..ddc6276a8a1 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -87,8 +87,9 @@ StorageObjectStorage::StorageObjectStorage( , distributed_processing(distributed_processing_) , log(getLogger(fmt::format("Storage{}({})", configuration->getEngineName(), table_id_.getFullTableName()))) { - configuration_->update(object_storage_, context); ColumnsDescription columns{columns_}; + LOG_DEBUG(&Poco::Logger::get("StorageObjectStorage Creation"), "Columns size {}", columns.size()); + configuration->update(object_storage, context); std::string sample_path; resolveSchemaAndFormat(columns, configuration->format, object_storage, configuration, format_settings, sample_path, context); diff --git a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp index cb1826b2976..9a525b4e21a 100644 --- a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp @@ -27,7 +27,6 @@ static std::shared_ptr createStorageObjectStorage( StorageObjectStorage::Configuration::initialize(*configuration, args.engine_args, context, false); - // Use format settings from global server context + settings from // the SETTINGS clause of the create query. Settings from current // session and user are ignored. @@ -251,6 +250,7 @@ void registerStorageDeltaLake(StorageFactory & factory) .source_access_type = AccessType::S3, }); #endif + UNUSED(factory); } #endif @@ -272,5 +272,6 @@ void registerStorageHudi(StorageFactory & factory) .source_access_type = AccessType::S3, }); #endif + UNUSED(factory); } } From 7ff2d5c98114d5d364e33cc5d0db88f5a1a06b8e Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Mon, 28 Oct 2024 14:01:37 +0000 Subject: [PATCH 286/680] add baseline --- src/Common/FieldVisitorMul.cpp | 50 ++++++ src/Common/FieldVisitorMul.h | 53 ++++++ src/Core/Field.h | 8 + src/Core/SortDescription.h | 5 +- src/Interpreters/FillingRow.cpp | 94 +++++++++-- src/Interpreters/FillingRow.h | 9 +- .../Transforms/FillingTransform.cpp | 159 +++++++++++------- 7 files changed, 306 insertions(+), 72 deletions(-) create mode 100644 src/Common/FieldVisitorMul.cpp create mode 100644 src/Common/FieldVisitorMul.h diff --git a/src/Common/FieldVisitorMul.cpp b/src/Common/FieldVisitorMul.cpp new file mode 100644 index 00000000000..36c32c40c05 --- /dev/null +++ b/src/Common/FieldVisitorMul.cpp @@ -0,0 +1,50 @@ +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + + +FieldVisitorMul::FieldVisitorMul(const Field & rhs_) : rhs(rhs_) {} + +// We can add all ints as unsigned regardless of their actual signedness. +bool FieldVisitorMul::operator() (Int64 & x) const { return this->operator()(reinterpret_cast(x)); } +bool FieldVisitorMul::operator() (UInt64 & x) const +{ + x *= applyVisitor(FieldVisitorConvertToNumber(), rhs); + return x != 0; +} + +bool FieldVisitorMul::operator() (Float64 & x) const { + x *= rhs.safeGet(); + return x != 0; +} + +bool FieldVisitorMul::operator() (Null &) const +{ + /// Do not add anything + return false; +} + +bool FieldVisitorMul::operator() (String &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Strings"); } +bool FieldVisitorMul::operator() (Array &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Arrays"); } +bool FieldVisitorMul::operator() (Tuple &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Tuples"); } +bool FieldVisitorMul::operator() (Map &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Maps"); } +bool FieldVisitorMul::operator() (Object &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Objects"); } +bool FieldVisitorMul::operator() (UUID &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply UUIDs"); } +bool FieldVisitorMul::operator() (IPv4 &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply IPv4s"); } +bool FieldVisitorMul::operator() (IPv6 &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply IPv6s"); } +bool FieldVisitorMul::operator() (CustomType & x) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply custom type {}", x.getTypeName()); } + +bool FieldVisitorMul::operator() (AggregateFunctionStateData &) const +{ + throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply AggregateFunctionStates"); +} + +bool FieldVisitorMul::operator() (bool &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Bools"); } + +} diff --git a/src/Common/FieldVisitorMul.h b/src/Common/FieldVisitorMul.h new file mode 100644 index 00000000000..5bce41f1e71 --- /dev/null +++ b/src/Common/FieldVisitorMul.h @@ -0,0 +1,53 @@ +#pragma once + +#include +#include + + +namespace DB +{ + +/** Implements `*=` operation. + * Returns false if the result is zero. + */ +class FieldVisitorMul : public StaticVisitor +{ +private: + const Field & rhs; +public: + explicit FieldVisitorMul(const Field & rhs_); + + // We can add all ints as unsigned regardless of their actual signedness. + bool operator() (Int64 & x) const; + bool operator() (UInt64 & x) const; + bool operator() (Float64 & x) const; + bool operator() (Null &) const; + bool operator() (String &) const; + bool operator() (Array &) const; + bool operator() (Tuple &) const; + bool operator() (Map &) const; + bool operator() (Object &) const; + bool operator() (UUID &) const; + bool operator() (IPv4 &) const; + bool operator() (IPv6 &) const; + bool operator() (AggregateFunctionStateData &) const; + bool operator() (CustomType &) const; + bool operator() (bool &) const; + + template + bool operator() (DecimalField & x) const + { + x *= rhs.safeGet>(); + return x.getValue() != T(0); + } + + template + requires is_big_int_v + bool operator() (T & x) const + { + x *= applyVisitor(FieldVisitorConvertToNumber(), rhs); + return x != T(0); + } +}; + +} diff --git a/src/Core/Field.h b/src/Core/Field.h index 7b916d30646..47df5c2907e 100644 --- a/src/Core/Field.h +++ b/src/Core/Field.h @@ -185,6 +185,14 @@ public: return *this; } + const DecimalField & operator *= (const DecimalField & r) + { + if (scale != r.getScale()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Multiply different decimal fields"); + dec *= r.getValue(); + return *this; + } + const DecimalField & operator -= (const DecimalField & r) { if (scale != r.getScale()) diff --git a/src/Core/SortDescription.h b/src/Core/SortDescription.h index 5c6f3e3150a..7a7c92f3b53 100644 --- a/src/Core/SortDescription.h +++ b/src/Core/SortDescription.h @@ -33,9 +33,12 @@ struct FillColumnDescription DataTypePtr fill_to_type; Field fill_step; /// Default = +1 or -1 according to direction std::optional step_kind; + Field fill_staleness; /// Default = Null - should not be considered + std::optional staleness_kind; - using StepFunction = std::function; + using StepFunction = std::function; StepFunction step_func; + StepFunction staleness_step_func; }; /// Description of the sorting rule by one column. diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index 21b5b04bca3..1d3eae03ddd 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -28,6 +28,7 @@ FillingRow::FillingRow(const SortDescription & sort_description_) : sort_description(sort_description_) { row.resize(sort_description.size()); + staleness_base_row.resize(sort_description.size()); } bool FillingRow::operator<(const FillingRow & other) const @@ -63,7 +64,53 @@ bool FillingRow::isNull() const return true; } -std::pair FillingRow::next(const FillingRow & to_row) +std::optional FillingRow::doJump(const FillColumnDescription& descr, size_t column_ind) +{ + Field next_value = row[column_ind]; + descr.step_func(next_value, 1); + + if (!descr.fill_to.isNull() && less(descr.fill_to, next_value, getDirection(column_ind))) + return std::nullopt; + + if (!descr.fill_staleness.isNull()) { + Field staleness_border = staleness_base_row[column_ind]; + descr.staleness_step_func(staleness_border, 1); + + if (less(next_value, staleness_border, getDirection(column_ind))) + return next_value; + else + return std::nullopt; + } + + return next_value; +} + +std::optional FillingRow::doLongJump(const FillColumnDescription & descr, size_t column_ind, const Field & to) +{ + Field shifted_value = row[column_ind]; + + if (less(to, shifted_value, getDirection(column_ind))) + return std::nullopt; + + for (int32_t step_len = 1, step_no = 0; step_no < 100; ++step_no) { + Field next_value = shifted_value; + descr.step_func(next_value, step_len); + + if (less(next_value, to, getDirection(0))) + { + shifted_value = std::move(next_value); + step_len *= 2; + } + else + { + step_len /= 2; + } + } + + return shifted_value; +} + +std::pair FillingRow::next(const FillingRow & to_row, bool long_jump) { const size_t row_size = size(); size_t pos = 0; @@ -85,23 +132,43 @@ std::pair FillingRow::next(const FillingRow & to_row) if (fill_column_desc.fill_to.isNull() || row[i].isNull()) continue; - Field next_value = row[i]; - fill_column_desc.step_func(next_value); - if (less(next_value, fill_column_desc.fill_to, getDirection(i))) + auto next_value = doJump(fill_column_desc, i); + if (next_value.has_value() && !equals(next_value.value(), fill_column_desc.fill_to)) { - row[i] = next_value; + row[i] = std::move(next_value.value()); initFromDefaults(i + 1); return {true, true}; } } - auto next_value = row[pos]; - getFillDescription(pos).step_func(next_value); + auto & fill_column_desc = getFillDescription(pos); + std::optional next_value; - if (less(to_row.row[pos], next_value, getDirection(pos)) || equals(next_value, getFillDescription(pos).fill_to)) - return {false, false}; + if (long_jump) + { + next_value = doLongJump(fill_column_desc, pos, to_row[pos]); - row[pos] = next_value; + if (!next_value.has_value()) + return {false, false}; + + Field calibration_jump_value = next_value.value(); + fill_column_desc.step_func(calibration_jump_value, 1); + + if (equals(calibration_jump_value, to_row[pos])) + next_value = calibration_jump_value; + + if (!next_value.has_value() || less(to_row.row[pos], next_value.value(), getDirection(pos)) || equals(next_value.value(), getFillDescription(pos).fill_to)) + return {false, false}; + } + else + { + next_value = doJump(fill_column_desc, pos); + + if (!next_value.has_value() || less(to_row.row[pos], next_value.value(), getDirection(pos)) || equals(next_value.value(), getFillDescription(pos).fill_to)) + return {false, false}; + } + + row[pos] = std::move(next_value.value()); if (equals(row[pos], to_row.row[pos])) { bool is_less = false; @@ -128,6 +195,13 @@ void FillingRow::initFromDefaults(size_t from_pos) row[i] = getFillDescription(i).fill_from; } +void FillingRow::initStalenessRow(const Columns& base_row, size_t row_ind) +{ + for (size_t i = 0; i < size(); ++i) { + staleness_base_row[i] = (*base_row[i])[row_ind]; + } +} + String FillingRow::dump() const { WriteBufferFromOwnString out; diff --git a/src/Interpreters/FillingRow.h b/src/Interpreters/FillingRow.h index 004b417542c..14b6034ce35 100644 --- a/src/Interpreters/FillingRow.h +++ b/src/Interpreters/FillingRow.h @@ -1,6 +1,6 @@ #pragma once -#include +#include namespace DB { @@ -15,6 +15,9 @@ bool equals(const Field & lhs, const Field & rhs); */ class FillingRow { + std::optional doJump(const FillColumnDescription & descr, size_t column_ind); + std::optional doLongJump(const FillColumnDescription & descr, size_t column_ind, const Field & to); + public: explicit FillingRow(const SortDescription & sort_description); @@ -22,9 +25,10 @@ public: /// Return pair of boolean /// apply - true if filling values should be inserted into result set /// value_changed - true if filling row value was changed - std::pair next(const FillingRow & to_row); + std::pair next(const FillingRow & to_row, bool long_jump); void initFromDefaults(size_t from_pos = 0); + void initStalenessRow(const Columns& base_row, size_t row_ind); Field & operator[](size_t index) { return row[index]; } const Field & operator[](size_t index) const { return row[index]; } @@ -42,6 +46,7 @@ public: private: Row row; + Row staleness_base_row; SortDescription sort_description; }; diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index 95f4a674ebb..1d68f73e8c2 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -7,15 +7,17 @@ #include #include #include +#include #include #include #include +#include namespace DB { -constexpr bool debug_logging_enabled = false; +constexpr bool debug_logging_enabled = true; template void logDebug(String key, const T & value, const char * separator = " : ") @@ -60,15 +62,78 @@ static FillColumnDescription::StepFunction getStepFunction( { #define DECLARE_CASE(NAME) \ case IntervalKind::Kind::NAME: \ - return [step, scale, &date_lut](Field & field) { \ + return [step, scale, &date_lut](Field & field, Int32 jumps_count) { \ field = Add##NAME##sImpl::execute(static_cast(\ - field.safeGet()), static_cast(step), date_lut, utc_time_zone, scale); }; + field.safeGet()), static_cast(step) * jumps_count, date_lut, utc_time_zone, scale); }; FOR_EACH_INTERVAL_KIND(DECLARE_CASE) #undef DECLARE_CASE } } +static FillColumnDescription::StepFunction getStepFunction(const Field & step, const std::optional & step_kind, const DataTypePtr & type) +{ + WhichDataType which(type); + + if (step_kind) + { + if (which.isDate() || which.isDate32()) + { + Int64 avg_seconds = step.safeGet() * step_kind->toAvgSeconds(); + if (std::abs(avg_seconds) < 86400) + throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, + "Value of step is to low ({} seconds). Must be >= 1 day", std::abs(avg_seconds)); + } + + if (which.isDate()) + return getStepFunction(step_kind.value(), step.safeGet(), DateLUT::instance()); + else if (which.isDate32()) + return getStepFunction(step_kind.value(), step.safeGet(), DateLUT::instance()); + else if (const auto * date_time = checkAndGetDataType(type.get())) + return getStepFunction(step_kind.value(), step.safeGet(), date_time->getTimeZone()); + else if (const auto * date_time64 = checkAndGetDataType(type.get())) + { + const auto & step_dec = step.safeGet &>(); + Int64 converted_step = DecimalUtils::convertTo(step_dec.getValue(), step_dec.getScale()); + static const DateLUTImpl & utc_time_zone = DateLUT::instance("UTC"); + + switch (step_kind.value()) // NOLINT(bugprone-switch-missing-default-case) + { +#define DECLARE_CASE(NAME) \ + case IntervalKind::Kind::NAME: \ + return [converted_step, &time_zone = date_time64->getTimeZone()](Field & field, Int32 jumps_count) \ + { \ + auto field_decimal = field.safeGet>(); \ + auto res = Add##NAME##sImpl::execute(field_decimal.getValue(), converted_step * jumps_count, time_zone, utc_time_zone, field_decimal.getScale()); \ + field = DecimalField(res, field_decimal.getScale()); \ + }; \ + break; + + FOR_EACH_INTERVAL_KIND(DECLARE_CASE) +#undef DECLARE_CASE + } + } + else + throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, + "STEP of Interval type can be used only with Date/DateTime types, but got {}", type->getName()); + } + else + { + return [step](Field & field, Int32 jumps_count) + { + auto shifted_step = step; + if (jumps_count != 1) + applyVisitor(FieldVisitorMul(jumps_count), shifted_step); + + logDebug("field", field.dump()); + logDebug("step", step.dump()); + logDebug("shifted field", shifted_step.dump()); + + applyVisitor(FieldVisitorSum(shifted_step), field); + }; + } +} + static bool tryConvertFields(FillColumnDescription & descr, const DataTypePtr & type) { auto max_type = Field::Types::Null; @@ -125,7 +190,8 @@ static bool tryConvertFields(FillColumnDescription & descr, const DataTypePtr & if (descr.fill_from.getType() > max_type || descr.fill_to.getType() > max_type - || descr.fill_step.getType() > max_type) + || descr.fill_step.getType() > max_type + || descr.fill_staleness.getType() > max_type) return false; if (!descr.fill_from.isNull()) @@ -134,56 +200,11 @@ static bool tryConvertFields(FillColumnDescription & descr, const DataTypePtr & descr.fill_to = convertFieldToTypeOrThrow(descr.fill_to, *to_type); if (!descr.fill_step.isNull()) descr.fill_step = convertFieldToTypeOrThrow(descr.fill_step, *to_type); + if (!descr.fill_staleness.isNull()) + descr.fill_staleness = convertFieldToTypeOrThrow(descr.fill_staleness, *to_type); - if (descr.step_kind) - { - if (which.isDate() || which.isDate32()) - { - Int64 avg_seconds = descr.fill_step.safeGet() * descr.step_kind->toAvgSeconds(); - if (std::abs(avg_seconds) < 86400) - throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, - "Value of step is to low ({} seconds). Must be >= 1 day", std::abs(avg_seconds)); - } - - if (which.isDate()) - descr.step_func = getStepFunction(*descr.step_kind, descr.fill_step.safeGet(), DateLUT::instance()); - else if (which.isDate32()) - descr.step_func = getStepFunction(*descr.step_kind, descr.fill_step.safeGet(), DateLUT::instance()); - else if (const auto * date_time = checkAndGetDataType(type.get())) - descr.step_func = getStepFunction(*descr.step_kind, descr.fill_step.safeGet(), date_time->getTimeZone()); - else if (const auto * date_time64 = checkAndGetDataType(type.get())) - { - const auto & step_dec = descr.fill_step.safeGet &>(); - Int64 step = DecimalUtils::convertTo(step_dec.getValue(), step_dec.getScale()); - static const DateLUTImpl & utc_time_zone = DateLUT::instance("UTC"); - - switch (*descr.step_kind) // NOLINT(bugprone-switch-missing-default-case) - { -#define DECLARE_CASE(NAME) \ - case IntervalKind::Kind::NAME: \ - descr.step_func = [step, &time_zone = date_time64->getTimeZone()](Field & field) \ - { \ - auto field_decimal = field.safeGet>(); \ - auto res = Add##NAME##sImpl::execute(field_decimal.getValue(), step, time_zone, utc_time_zone, field_decimal.getScale()); \ - field = DecimalField(res, field_decimal.getScale()); \ - }; \ - break; - - FOR_EACH_INTERVAL_KIND(DECLARE_CASE) -#undef DECLARE_CASE - } - } - else - throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, - "STEP of Interval type can be used only with Date/DateTime types, but got {}", type->getName()); - } - else - { - descr.step_func = [step = descr.fill_step](Field & field) - { - applyVisitor(FieldVisitorSum(step), field); - }; - } + descr.step_func = getStepFunction(descr.fill_step, descr.step_kind, type); + descr.staleness_step_func = getStepFunction(descr.fill_staleness, descr.staleness_kind, type); return true; } @@ -482,8 +503,8 @@ bool FillingTransform::generateSuffixIfNeeded( MutableColumnRawPtrs res_sort_prefix_columns, MutableColumnRawPtrs res_other_columns) { - logDebug("generateSuffixIfNeeded() filling_row", filling_row); - logDebug("generateSuffixIfNeeded() next_row", next_row); + logDebug("generateSuffixIfNeeded filling_row", filling_row); + logDebug("generateSuffixIfNeeded next_row", next_row); /// Determines if we should insert filling row before start generating next rows bool should_insert_first = (next_row < filling_row && !filling_row_inserted) || next_row.isNull(); @@ -492,11 +513,11 @@ bool FillingTransform::generateSuffixIfNeeded( for (size_t i = 0, size = filling_row.size(); i < size; ++i) next_row[i] = filling_row.getFillDescription(i).fill_to; - logDebug("generateSuffixIfNeeded() next_row updated", next_row); + logDebug("generateSuffixIfNeeded next_row updated", next_row); if (filling_row >= next_row) { - logDebug("generateSuffixIfNeeded()", "no need to generate suffix"); + logDebug("generateSuffixIfNeeded", "no need to generate suffix"); return false; } @@ -516,7 +537,7 @@ bool FillingTransform::generateSuffixIfNeeded( bool filling_row_changed = false; while (true) { - const auto [apply, changed] = filling_row.next(next_row); + const auto [apply, changed] = filling_row.next(next_row, /*long_jump=*/false); filling_row_changed = changed; if (!apply) break; @@ -593,6 +614,9 @@ void FillingTransform::transformRange( const auto current_value = (*input_fill_columns[i])[range_begin]; const auto & fill_from = filling_row.getFillDescription(i).fill_from; + logDebug("current value", current_value.dump()); + logDebug("fill from", fill_from.dump()); + if (!fill_from.isNull() && !equals(current_value, fill_from)) { filling_row.initFromDefaults(i); @@ -609,6 +633,9 @@ void FillingTransform::transformRange( } } + /// Init staleness first interval + filling_row.initStalenessRow(input_fill_columns, range_begin); + for (size_t row_ind = range_begin; row_ind < range_end; ++row_ind) { logDebug("row", row_ind); @@ -623,6 +650,9 @@ void FillingTransform::transformRange( const auto current_value = (*input_fill_columns[i])[row_ind]; const auto & fill_to = filling_row.getFillDescription(i).fill_to; + logDebug("current value", current_value.dump()); + logDebug("fill to", fill_to.dump()); + if (fill_to.isNull() || less(current_value, fill_to, filling_row.getDirection(i))) next_row[i] = current_value; else @@ -643,7 +673,7 @@ void FillingTransform::transformRange( bool filling_row_changed = false; while (true) { - const auto [apply, changed] = filling_row.next(next_row); + const auto [apply, changed] = filling_row.next(next_row, /*long_jump=*/false); filling_row_changed = changed; if (!apply) break; @@ -652,6 +682,14 @@ void FillingTransform::transformRange( insertFromFillingRow(res_fill_columns, res_interpolate_columns, res_other_columns, interpolate_block); copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); } + + const auto [apply, changed] = filling_row.next(next_row, /*long_jump=*/true); + logDebug("apply", apply); + logDebug("changed", changed); + + if (changed) + filling_row_changed = true; + /// new valid filling row was generated but not inserted, will use it during suffix generation if (filling_row_changed) filling_row_inserted = false; @@ -662,6 +700,9 @@ void FillingTransform::transformRange( copyRowFromColumns(res_interpolate_columns, input_interpolate_columns, row_ind); copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); copyRowFromColumns(res_other_columns, input_other_columns, row_ind); + + /// Init next staleness interval with current row, because we have already made the long jump to it + filling_row.initStalenessRow(input_fill_columns, row_ind); } /// save sort prefix of last row in the range, it's used to generate suffix From 8f9d577c453573d82a529186fde60697d509e6f2 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Mon, 28 Oct 2024 10:12:59 -0400 Subject: [PATCH 287/680] add enable_job_stack_trace to change history --- src/Core/SettingsChangesHistory.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index d958d091975..02601f12d56 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -68,6 +68,7 @@ static std::initializer_list Date: Mon, 28 Oct 2024 15:13:33 +0000 Subject: [PATCH 288/680] change mul to scale --- src/Common/FieldVisitorMul.cpp | 50 ----------------- src/Common/FieldVisitorMul.h | 53 ------------------- src/Common/FieldVisitorScale.cpp | 30 +++++++++++ src/Common/FieldVisitorScale.h | 46 ++++++++++++++++ .../Transforms/FillingTransform.cpp | 4 +- 5 files changed, 78 insertions(+), 105 deletions(-) delete mode 100644 src/Common/FieldVisitorMul.cpp delete mode 100644 src/Common/FieldVisitorMul.h create mode 100644 src/Common/FieldVisitorScale.cpp create mode 100644 src/Common/FieldVisitorScale.h diff --git a/src/Common/FieldVisitorMul.cpp b/src/Common/FieldVisitorMul.cpp deleted file mode 100644 index 36c32c40c05..00000000000 --- a/src/Common/FieldVisitorMul.cpp +++ /dev/null @@ -1,50 +0,0 @@ -#include - -namespace DB -{ - -namespace ErrorCodes -{ - extern const int LOGICAL_ERROR; -} - - -FieldVisitorMul::FieldVisitorMul(const Field & rhs_) : rhs(rhs_) {} - -// We can add all ints as unsigned regardless of their actual signedness. -bool FieldVisitorMul::operator() (Int64 & x) const { return this->operator()(reinterpret_cast(x)); } -bool FieldVisitorMul::operator() (UInt64 & x) const -{ - x *= applyVisitor(FieldVisitorConvertToNumber(), rhs); - return x != 0; -} - -bool FieldVisitorMul::operator() (Float64 & x) const { - x *= rhs.safeGet(); - return x != 0; -} - -bool FieldVisitorMul::operator() (Null &) const -{ - /// Do not add anything - return false; -} - -bool FieldVisitorMul::operator() (String &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Strings"); } -bool FieldVisitorMul::operator() (Array &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Arrays"); } -bool FieldVisitorMul::operator() (Tuple &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Tuples"); } -bool FieldVisitorMul::operator() (Map &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Maps"); } -bool FieldVisitorMul::operator() (Object &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Objects"); } -bool FieldVisitorMul::operator() (UUID &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply UUIDs"); } -bool FieldVisitorMul::operator() (IPv4 &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply IPv4s"); } -bool FieldVisitorMul::operator() (IPv6 &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply IPv6s"); } -bool FieldVisitorMul::operator() (CustomType & x) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply custom type {}", x.getTypeName()); } - -bool FieldVisitorMul::operator() (AggregateFunctionStateData &) const -{ - throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply AggregateFunctionStates"); -} - -bool FieldVisitorMul::operator() (bool &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Bools"); } - -} diff --git a/src/Common/FieldVisitorMul.h b/src/Common/FieldVisitorMul.h deleted file mode 100644 index 5bce41f1e71..00000000000 --- a/src/Common/FieldVisitorMul.h +++ /dev/null @@ -1,53 +0,0 @@ -#pragma once - -#include -#include - - -namespace DB -{ - -/** Implements `*=` operation. - * Returns false if the result is zero. - */ -class FieldVisitorMul : public StaticVisitor -{ -private: - const Field & rhs; -public: - explicit FieldVisitorMul(const Field & rhs_); - - // We can add all ints as unsigned regardless of their actual signedness. - bool operator() (Int64 & x) const; - bool operator() (UInt64 & x) const; - bool operator() (Float64 & x) const; - bool operator() (Null &) const; - bool operator() (String &) const; - bool operator() (Array &) const; - bool operator() (Tuple &) const; - bool operator() (Map &) const; - bool operator() (Object &) const; - bool operator() (UUID &) const; - bool operator() (IPv4 &) const; - bool operator() (IPv6 &) const; - bool operator() (AggregateFunctionStateData &) const; - bool operator() (CustomType &) const; - bool operator() (bool &) const; - - template - bool operator() (DecimalField & x) const - { - x *= rhs.safeGet>(); - return x.getValue() != T(0); - } - - template - requires is_big_int_v - bool operator() (T & x) const - { - x *= applyVisitor(FieldVisitorConvertToNumber(), rhs); - return x != T(0); - } -}; - -} diff --git a/src/Common/FieldVisitorScale.cpp b/src/Common/FieldVisitorScale.cpp new file mode 100644 index 00000000000..fdb566007c3 --- /dev/null +++ b/src/Common/FieldVisitorScale.cpp @@ -0,0 +1,30 @@ +#include + +namespace DB +{ + +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +} + +FieldVisitorScale::FieldVisitorScale(Int32 rhs_) : rhs(rhs_) {} + +void FieldVisitorScale::operator() (Int64 & x) const { x *= rhs; } +void FieldVisitorScale::operator() (UInt64 & x) const { x *= rhs; } +void FieldVisitorScale::operator() (Float64 & x) const { x *= rhs; } +void FieldVisitorScale::operator() (Null &) const { /*Do not scale anything*/ } + +void FieldVisitorScale::operator() (String &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Strings"); } +void FieldVisitorScale::operator() (Array &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Arrays"); } +void FieldVisitorScale::operator() (Tuple &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Tuples"); } +void FieldVisitorScale::operator() (Map &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Maps"); } +void FieldVisitorScale::operator() (Object &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Objects"); } +void FieldVisitorScale::operator() (UUID &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply UUIDs"); } +void FieldVisitorScale::operator() (IPv4 &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply IPv4s"); } +void FieldVisitorScale::operator() (IPv6 &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply IPv6s"); } +void FieldVisitorScale::operator() (CustomType & x) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply custom type {}", x.getTypeName()); } +void FieldVisitorScale::operator() (AggregateFunctionStateData &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply AggregateFunctionStates"); } +void FieldVisitorScale::operator() (bool &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Bools"); } + +} diff --git a/src/Common/FieldVisitorScale.h b/src/Common/FieldVisitorScale.h new file mode 100644 index 00000000000..45bacdccc9c --- /dev/null +++ b/src/Common/FieldVisitorScale.h @@ -0,0 +1,46 @@ +#pragma once + +#include +#include +#include +#include "base/Decimal.h" +#include "base/extended_types.h" + +namespace DB +{ + +/** Implements `*=` operation by number + */ +class FieldVisitorScale : public StaticVisitor +{ +private: + Int32 rhs; + +public: + explicit FieldVisitorScale(Int32 rhs_); + + void operator() (Int64 & x) const; + void operator() (UInt64 & x) const; + void operator() (Float64 & x) const; + void operator() (Null &) const; + [[noreturn]] void operator() (String &) const; + [[noreturn]] void operator() (Array &) const; + [[noreturn]] void operator() (Tuple &) const; + [[noreturn]] void operator() (Map &) const; + [[noreturn]] void operator() (Object &) const; + [[noreturn]] void operator() (UUID &) const; + [[noreturn]] void operator() (IPv4 &) const; + [[noreturn]] void operator() (IPv6 &) const; + [[noreturn]] void operator() (AggregateFunctionStateData &) const; + [[noreturn]] void operator() (CustomType &) const; + [[noreturn]] void operator() (bool &) const; + + template + void operator() (DecimalField & x) const { x = DecimalField(x.getValue() * T(rhs), x.getScale()); } + + template + requires is_big_int_v + void operator() (T & x) const { x *= rhs; } +}; + +} diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index 1d68f73e8c2..54331186302 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -7,7 +7,7 @@ #include #include #include -#include +#include #include #include #include @@ -123,7 +123,7 @@ static FillColumnDescription::StepFunction getStepFunction(const Field & step, c { auto shifted_step = step; if (jumps_count != 1) - applyVisitor(FieldVisitorMul(jumps_count), shifted_step); + applyVisitor(FieldVisitorScale(jumps_count), shifted_step); logDebug("field", field.dump()); logDebug("step", step.dump()); From 2c3363e40e1856f7b5ce8eb23c301ee2ee403f36 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Mon, 28 Oct 2024 19:00:37 +0300 Subject: [PATCH 289/680] Hard limit on replicated tables, dicts, views --- src/Common/CurrentMetrics.cpp | 1 + src/Core/ServerSettings.cpp | 3 + src/Databases/DatabasesCommon.cpp | 6 +- src/Interpreters/InterpreterCreateQuery.cpp | 63 ++++++++++++++++--- src/Interpreters/InterpreterCreateQuery.h | 2 + src/Storages/Utils.cpp | 16 +++-- src/Storages/Utils.h | 2 +- .../test_table_db_num_limit/config/config.xml | 12 ++++ .../test_table_db_num_limit/test.py | 43 ++++++++++--- 9 files changed, 121 insertions(+), 27 deletions(-) diff --git a/src/Common/CurrentMetrics.cpp b/src/Common/CurrentMetrics.cpp index e9d5e07c914..542838813de 100644 --- a/src/Common/CurrentMetrics.cpp +++ b/src/Common/CurrentMetrics.cpp @@ -242,6 +242,7 @@ M(PartsActive, "Active data part, used by current and upcoming SELECTs.") \ M(AttachedDatabase, "Active databases.") \ M(AttachedTable, "Active tables.") \ + M(AttachedReplicatedTable, "Active replicated tables.") \ M(AttachedView, "Active views.") \ M(AttachedDictionary, "Active dictionaries.") \ M(PartsOutdated, "Not active data part, but could be used by only current SELECTs, could be deleted after SELECTs finishes.") \ diff --git a/src/Core/ServerSettings.cpp b/src/Core/ServerSettings.cpp index 8c0864e78b7..2240b45a49f 100644 --- a/src/Core/ServerSettings.cpp +++ b/src/Core/ServerSettings.cpp @@ -128,7 +128,10 @@ namespace DB M(UInt64, max_database_num_to_warn, 1000lu, "If the number of databases is greater than this value, the server will create a warning that will displayed to user.", 0) \ M(UInt64, max_part_num_to_warn, 100000lu, "If the number of parts is greater than this value, the server will create a warning that will displayed to user.", 0) \ M(UInt64, max_table_num_to_throw, 0lu, "If number of tables is greater than this value, server will throw an exception. 0 means no limitation. View, remote tables, dictionary, system tables are not counted. Only count table in Atomic/Ordinary/Replicated/Lazy database engine.", 0) \ + M(UInt64, max_replicated_table_num_to_throw, 0lu, "If number of replicated tables is greater than this value, server will throw an exception. 0 means no limitation. Only count table in Atomic/Ordinary/Replicated/Lazy database engine.", 0) \ M(UInt64, max_database_num_to_throw, 0lu, "If number of databases is greater than this value, server will throw an exception. 0 means no limitation.", 0) \ + M(UInt64, max_dictionary_num_to_throw, 0lu, "If number of dictionaries is greater than this value, server will throw an exception. 0 means no limitation. Only count table in Atomic/Ordinary/Replicated/Lazy database engine.", 0) \ + M(UInt64, max_view_num_to_throw, 0lu, "If number of views is greater than this value, server will throw an exception. 0 means no limitation. Only count table in Atomic/Ordinary/Replicated/Lazy database engine.", 0) \ M(UInt64, max_authentication_methods_per_user, 100, "The maximum number of authentication methods a user can be created with or altered. Changing this setting does not affect existing users. Zero means unlimited", 0) \ M(UInt64, concurrent_threads_soft_limit_num, 0, "Sets how many concurrent thread can be allocated before applying CPU pressure. Zero means unlimited.", 0) \ M(UInt64, concurrent_threads_soft_limit_ratio_to_cores, 0, "Same as concurrent_threads_soft_limit_num, but with ratio to cores.", 0) \ diff --git a/src/Databases/DatabasesCommon.cpp b/src/Databases/DatabasesCommon.cpp index d26ec9d6eec..23d199cd160 100644 --- a/src/Databases/DatabasesCommon.cpp +++ b/src/Databases/DatabasesCommon.cpp @@ -382,7 +382,8 @@ StoragePtr DatabaseWithOwnTablesBase::detachTableUnlocked(const String & table_n if (!table_storage->isSystemStorage() && !DatabaseCatalog::isPredefinedDatabase(database_name)) { LOG_TEST(log, "Counting detached table {} to database {}", table_name, database_name); - CurrentMetrics::sub(getAttachedCounterForStorage(table_storage)); + for (auto metric : getAttachedCountersForStorage(table_storage)) + CurrentMetrics::sub(metric); } auto table_id = table_storage->getStorageID(); @@ -430,7 +431,8 @@ void DatabaseWithOwnTablesBase::attachTableUnlocked(const String & table_name, c if (!table->isSystemStorage() && !DatabaseCatalog::isPredefinedDatabase(database_name)) { LOG_TEST(log, "Counting attached table {} to database {}", table_name, database_name); - CurrentMetrics::add(getAttachedCounterForStorage(table)); + for (auto metric : getAttachedCountersForStorage(table)) + CurrentMetrics::add(metric); } } diff --git a/src/Interpreters/InterpreterCreateQuery.cpp b/src/Interpreters/InterpreterCreateQuery.cpp index 6057afefd02..f8e85733911 100644 --- a/src/Interpreters/InterpreterCreateQuery.cpp +++ b/src/Interpreters/InterpreterCreateQuery.cpp @@ -98,6 +98,9 @@ namespace CurrentMetrics { extern const Metric AttachedTable; + extern const Metric AttachedReplicatedTable; + extern const Metric AttachedDictionary; + extern const Metric AttachedView; } namespace DB @@ -146,7 +149,10 @@ namespace ServerSetting { extern const ServerSettingsBool ignore_empty_sql_security_in_create_view_query; extern const ServerSettingsUInt64 max_database_num_to_throw; + extern const ServerSettingsUInt64 max_dictionary_num_to_throw; extern const ServerSettingsUInt64 max_table_num_to_throw; + extern const ServerSettingsUInt64 max_replicated_table_num_to_throw; + extern const ServerSettingsUInt64 max_view_num_to_throw; } namespace ErrorCodes @@ -1914,16 +1920,8 @@ bool InterpreterCreateQuery::doCreateTable(ASTCreateQuery & create, } } - UInt64 table_num_limit = getContext()->getGlobalContext()->getServerSettings()[ServerSetting::max_table_num_to_throw]; - if (table_num_limit > 0 && !internal) - { - UInt64 table_count = CurrentMetrics::get(CurrentMetrics::AttachedTable); - if (table_count >= table_num_limit) - throw Exception(ErrorCodes::TOO_MANY_TABLES, - "Too many tables. " - "The limit (server configuration parameter `max_table_num_to_throw`) is set to {}, the current number of tables is {}", - table_num_limit, table_count); - } + if (!internal) + throwIfTooManyEntities(create, res); database->createTable(getContext(), create.getTable(), res, query_ptr); @@ -1950,6 +1948,51 @@ bool InterpreterCreateQuery::doCreateTable(ASTCreateQuery & create, } +void InterpreterCreateQuery::throwIfTooManyEntities(ASTCreateQuery & create, StoragePtr storage) const +{ + if (auto * replicated_storage = typeid_cast(storage.get())) + { + UInt64 num_limit = getContext()->getGlobalContext()->getServerSettings()[ServerSetting::max_replicated_table_num_to_throw]; + UInt64 attached_count = CurrentMetrics::get(CurrentMetrics::AttachedReplicatedTable); + if (attached_count >= num_limit) + throw Exception(ErrorCodes::TOO_MANY_TABLES, + "Too many replicated tables. " + "The limit (server configuration parameter `max_replicated_table_num_to_throw`) is set to {}, the current number is {}", + num_limit, attached_count); + } + else if (create.is_dictionary) + { + UInt64 num_limit = getContext()->getGlobalContext()->getServerSettings()[ServerSetting::max_dictionary_num_to_throw]; + UInt64 attached_count = CurrentMetrics::get(CurrentMetrics::AttachedDictionary); + if (attached_count >= num_limit) + throw Exception(ErrorCodes::TOO_MANY_TABLES, + "Too many dictionaries. " + "The limit (server configuration parameter `max_dictionary_num_to_throw`) is set to {}, the current number is {}", + num_limit, attached_count); + } + else if (create.isView()) + { + UInt64 num_limit = getContext()->getGlobalContext()->getServerSettings()[ServerSetting::max_view_num_to_throw]; + UInt64 attached_count = CurrentMetrics::get(CurrentMetrics::AttachedView); + if (attached_count >= num_limit) + throw Exception(ErrorCodes::TOO_MANY_TABLES, + "Too many views. " + "The limit (server configuration parameter `max_view_num_to_throw`) is set to {}, the current number is {}", + num_limit, attached_count); + } + else + { + UInt64 num_limit = getContext()->getGlobalContext()->getServerSettings()[ServerSetting::max_table_num_to_throw]; + UInt64 attached_count = CurrentMetrics::get(CurrentMetrics::AttachedTable); + if (attached_count >= num_limit) + throw Exception(ErrorCodes::TOO_MANY_TABLES, + "Too many tables. " + "The limit (server configuration parameter `max_table_num_to_throw`) is set to {}, the current number is {}", + num_limit, attached_count); + } +} + + BlockIO InterpreterCreateQuery::doCreateOrReplaceTable(ASTCreateQuery & create, const InterpreterCreateQuery::TableProperties & properties, LoadingStrictnessLevel mode) { diff --git a/src/Interpreters/InterpreterCreateQuery.h b/src/Interpreters/InterpreterCreateQuery.h index cb7af25383e..24cf308951c 100644 --- a/src/Interpreters/InterpreterCreateQuery.h +++ b/src/Interpreters/InterpreterCreateQuery.h @@ -122,6 +122,8 @@ private: BlockIO executeQueryOnCluster(ASTCreateQuery & create); + void throwIfTooManyEntities(ASTCreateQuery & create, StoragePtr storage) const; + ASTPtr query_ptr; /// Skip safety threshold when loading tables. diff --git a/src/Storages/Utils.cpp b/src/Storages/Utils.cpp index bd03a96c7cc..72aeb0d158d 100644 --- a/src/Storages/Utils.cpp +++ b/src/Storages/Utils.cpp @@ -1,10 +1,13 @@ +#include #include #include +#include namespace CurrentMetrics { extern const Metric AttachedTable; + extern const Metric AttachedReplicatedTable; extern const Metric AttachedView; extern const Metric AttachedDictionary; } @@ -12,17 +15,20 @@ namespace CurrentMetrics namespace DB { - CurrentMetrics::Metric getAttachedCounterForStorage(const StoragePtr & storage) + std::vector getAttachedCountersForStorage(const StoragePtr & storage) { if (storage->isView()) { - return CurrentMetrics::AttachedView; + return {CurrentMetrics::AttachedView}; } if (storage->isDictionary()) { - return CurrentMetrics::AttachedDictionary; + return {CurrentMetrics::AttachedDictionary}; } - - return CurrentMetrics::AttachedTable; + if (auto * replicated_storage = typeid_cast(storage.get())) + { + return {CurrentMetrics::AttachedTable, CurrentMetrics::AttachedReplicatedTable}; + } + return {CurrentMetrics::AttachedTable}; } } diff --git a/src/Storages/Utils.h b/src/Storages/Utils.h index c86c2a4c341..eb302178485 100644 --- a/src/Storages/Utils.h +++ b/src/Storages/Utils.h @@ -6,5 +6,5 @@ namespace DB { - CurrentMetrics::Metric getAttachedCounterForStorage(const StoragePtr & storage); + std::vector getAttachedCountersForStorage(const StoragePtr & storage); } diff --git a/tests/integration/test_table_db_num_limit/config/config.xml b/tests/integration/test_table_db_num_limit/config/config.xml index 9a573b158fe..a4246c79694 100644 --- a/tests/integration/test_table_db_num_limit/config/config.xml +++ b/tests/integration/test_table_db_num_limit/config/config.xml @@ -1,5 +1,17 @@ + + + + + node1 + 9000 + + + + + 10 + 5 10 diff --git a/tests/integration/test_table_db_num_limit/test.py b/tests/integration/test_table_db_num_limit/test.py index b3aff6ddca2..ce981ffca3c 100644 --- a/tests/integration/test_table_db_num_limit/test.py +++ b/tests/integration/test_table_db_num_limit/test.py @@ -1,11 +1,14 @@ import pytest -from helpers.client import QueryRuntimeException from helpers.cluster import ClickHouseCluster cluster = ClickHouseCluster(__file__) -node = cluster.add_instance("node", main_configs=["config/config.xml"]) +node = cluster.add_instance( + "node1", + with_zookeeper=True, + main_configs=["config/config.xml"], +) @pytest.fixture(scope="module") @@ -24,10 +27,9 @@ def test_table_db_limit(started_cluster): for i in range(9): node.query("create database db{}".format(i)) - with pytest.raises(QueryRuntimeException) as exp_info: - node.query("create database db_exp".format(i)) - - assert "TOO_MANY_DATABASES" in str(exp_info) + assert "TOO_MANY_DATABASES" in node.query_and_get_error( + "create database db_exp".format(i) + ) for i in range(10): node.query("create table t{} (a Int32) Engine = Log".format(i)) @@ -35,13 +37,36 @@ def test_table_db_limit(started_cluster): # This checks that system tables are not accounted in the number of tables. node.query("system flush logs") + # Regular tables for i in range(10): node.query("drop table t{}".format(i)) for i in range(10): node.query("create table t{} (a Int32) Engine = Log".format(i)) - with pytest.raises(QueryRuntimeException) as exp_info: - node.query("create table default.tx (a Int32) Engine = Log") + assert "TOO_MANY_TABLES" in node.query_and_get_error( + "create table default.tx (a Int32) Engine = Log" + ) - assert "TOO_MANY_TABLES" in str(exp_info) + # Replicated tables + for i in range(10): + node.query("drop table t{}".format(i)) + + for i in range(5): + node.query( + "create table t{} (a Int32) Engine = ReplicatedMergeTree('/clickhouse/tables/t{}', 'r1') order by a".format( + i, i + ) + ) + + assert "Too many replicated tables" in node.query_and_get_error( + "create table tx (a Int32) Engine = ReplicatedMergeTree('/clickhouse/tables/tx', 'r1') order by a" + ) + + # Checks that replicated tables are also counted as regular tables + for i in range(5, 10): + node.query("create table t{} (a Int32) Engine = Log".format(i)) + + assert "TOO_MANY_TABLES" in node.query_and_get_error( + "create table tx (a Int32) Engine = Log" + ) From 2d7de40ba70d6609f6fd79c5ef8534002803b707 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Mon, 28 Oct 2024 17:24:03 +0000 Subject: [PATCH 290/680] fix sparse tables --- src/Processors/Transforms/FillingTransform.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index 54331186302..635b46de3ee 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -458,7 +458,7 @@ void FillingTransform::initColumns( non_const_columns.reserve(input_columns.size()); for (const auto & column : input_columns) - non_const_columns.push_back(column->convertToFullColumnIfConst()); + non_const_columns.push_back(column->convertToFullColumnIfConst()->convertToFullColumnIfSparse()); for (const auto & column : non_const_columns) output_columns.push_back(column->cloneEmpty()->assumeMutable()); From 37f691bf9d1168431500c39c47432722a441a29e Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Mon, 28 Oct 2024 17:42:52 +0000 Subject: [PATCH 291/680] add test --- .../03266_with_fill_staleness.reference | 28 +++++++++++++++++ .../0_stateless/03266_with_fill_staleness.sql | 31 +++++++++++++++++++ 2 files changed, 59 insertions(+) create mode 100644 tests/queries/0_stateless/03266_with_fill_staleness.reference create mode 100644 tests/queries/0_stateless/03266_with_fill_staleness.sql diff --git a/tests/queries/0_stateless/03266_with_fill_staleness.reference b/tests/queries/0_stateless/03266_with_fill_staleness.reference new file mode 100644 index 00000000000..6061ecfe400 --- /dev/null +++ b/tests/queries/0_stateless/03266_with_fill_staleness.reference @@ -0,0 +1,28 @@ +add samples +regular with fill +2016-06-15 23:00:00 0 +2016-06-15 23:00:01 0 +2016-06-15 23:00:02 0 +2016-06-15 23:00:03 0 +2016-06-15 23:00:04 0 +2016-06-15 23:00:05 5 +2016-06-15 23:00:06 5 +2016-06-15 23:00:07 5 +2016-06-15 23:00:08 5 +2016-06-15 23:00:09 5 +2016-06-15 23:00:10 10 +2016-06-15 23:00:11 10 +2016-06-15 23:00:12 10 +2016-06-15 23:00:13 10 +2016-06-15 23:00:14 10 +2016-06-15 23:00:15 15 +2016-06-15 23:00:16 15 +2016-06-15 23:00:17 15 +2016-06-15 23:00:18 15 +2016-06-15 23:00:19 15 +2016-06-15 23:00:20 20 +2016-06-15 23:00:21 20 +2016-06-15 23:00:22 20 +2016-06-15 23:00:23 20 +2016-06-15 23:00:24 20 +2016-06-15 23:00:25 25 diff --git a/tests/queries/0_stateless/03266_with_fill_staleness.sql b/tests/queries/0_stateless/03266_with_fill_staleness.sql new file mode 100644 index 00000000000..3ab9be63a08 --- /dev/null +++ b/tests/queries/0_stateless/03266_with_fill_staleness.sql @@ -0,0 +1,31 @@ +DROP TABLE IF EXISTS with_fill_staleness; +CREATE TABLE with_fill_staleness (a DateTime, b DateTime, c UInt64) ENGINE = MergeTree ORDER BY a; + +SELECT 'add samples'; + +INSERT INTO with_fill_staleness +SELECT + toDateTime('2016-06-15 23:00:00') + number AS a, a as b, number as c +FROM numbers(30) +WHERE (number % 5) == 0; + +SELECT 'regular with fill'; +SELECT a, c, 'original' as original FROM with_fill_staleness ORDER BY a ASC WITH FILL INTERPOLATE (c); + +SELECT 'staleness 1 seconds'; +SELECT a, c, 'original' as original FROM with_fill_staleness ORDER BY a ASC WITH FILL STALENESS INTERVAL 1 SECOND INTERPOLATE (c); + +SELECT 'staleness 3 seconds'; +SELECT a, c, 'original' as original FROM with_fill_staleness ORDER BY a ASC WITH FILL STALENESS INTERVAL 3 SECOND INTERPOLATE (c); + +SELECT 'descending order'; +SELECT a, c, 'original' as original FROM with_fill_staleness ORDER BY a DESC WITH FILL STALENESS INTERVAL -2 SECOND INTERPOLATE (c); + +SELECT 'staleness with to and step'; +SELECT a, c, 'original' as original FROM with_fill_staleness ORDER BY a ASC WITH FILL TO toDateTime('2016-06-15 23:00:40') STEP 3 STALENESS INTERVAL 7 SECOND INTERPOLATE (c); + +SELECT 'staleness with another regular with fill'; +SELECT a, b, c, 'original' as original FROM with_fill_staleness ORDER BY a ASC WITH FILL STALENESS INTERVAL 2 SECOND, b ASC WITH FILL FROM 0 TO 3 INTERPOLATE (c); + +SELECT 'double staleness'; +SELECT a, b, c, 'original' as original FROM with_fill_staleness ORDER BY a ASC WITH FILL STALENESS INTERVAL 2 SECOND, b ASC WITH FILL TO toDateTime('2016-06-15 23:01:00') STEP 2 STALENESS 5 INTERPOLATE (c); From 9760d39efe82339403de7a7177706c42c8d8c5a5 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Mon, 28 Oct 2024 17:43:15 +0000 Subject: [PATCH 292/680] allow negative staleness for descending order --- src/Planner/PlannerSorting.cpp | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/src/Planner/PlannerSorting.cpp b/src/Planner/PlannerSorting.cpp index 0a33e2f0828..9476ae348c5 100644 --- a/src/Planner/PlannerSorting.cpp +++ b/src/Planner/PlannerSorting.cpp @@ -105,10 +105,6 @@ FillColumnDescription extractWithFillDescription(const SortNode & sort_node) if (sort_node.hasFillFrom()) throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, "WITH FILL STALENESS cannot be used together with WITH FILL FROM"); - - if (applyVisitor(FieldVisitorAccurateLessOrEqual(), fill_column_description.fill_staleness, Field{0})) - throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, - "WITH FILL STALENESS value cannot be less or equal zero"); } if (sort_node.getSortDirection() == SortDirection::ASCENDING) @@ -117,6 +113,10 @@ FillColumnDescription extractWithFillDescription(const SortNode & sort_node) throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, "WITH FILL STEP value cannot be negative for sorting in ascending direction"); + if (applyVisitor(FieldVisitorAccurateLess(), fill_column_description.fill_staleness, Field{0})) + throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, + "WITH FILL STALENESS value cannot be negative for sorting in ascending direction"); + if (!fill_column_description.fill_from.isNull() && !fill_column_description.fill_to.isNull() && applyVisitor(FieldVisitorAccurateLess(), fill_column_description.fill_to, fill_column_description.fill_from)) { @@ -130,6 +130,10 @@ FillColumnDescription extractWithFillDescription(const SortNode & sort_node) throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, "WITH FILL STEP value cannot be positive for sorting in descending direction"); + if (applyVisitor(FieldVisitorAccurateLess(), Field{0}, fill_column_description.fill_staleness)) + throw Exception(ErrorCodes::INVALID_WITH_FILL_EXPRESSION, + "WITH FILL STALENESS value cannot be positive for sorting in descending direction"); + if (!fill_column_description.fill_from.isNull() && !fill_column_description.fill_to.isNull() && applyVisitor(FieldVisitorAccurateLess(), fill_column_description.fill_from, fill_column_description.fill_to)) { From fc33593ff05ab3c5ca4271b79ba4eb39957fa057 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Mon, 28 Oct 2024 17:45:02 +0000 Subject: [PATCH 293/680] fix style --- src/Interpreters/FillingRow.cpp | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index 1d3eae03ddd..fdd3b55b66b 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -72,7 +72,8 @@ std::optional FillingRow::doJump(const FillColumnDescription& descr, size if (!descr.fill_to.isNull() && less(descr.fill_to, next_value, getDirection(column_ind))) return std::nullopt; - if (!descr.fill_staleness.isNull()) { + if (!descr.fill_staleness.isNull()) + { Field staleness_border = staleness_base_row[column_ind]; descr.staleness_step_func(staleness_border, 1); @@ -92,7 +93,8 @@ std::optional FillingRow::doLongJump(const FillColumnDescription & descr, if (less(to, shifted_value, getDirection(column_ind))) return std::nullopt; - for (int32_t step_len = 1, step_no = 0; step_no < 100; ++step_no) { + for (int32_t step_len = 1, step_no = 0; step_no < 100; ++step_no) + { Field next_value = shifted_value; descr.step_func(next_value, step_len); @@ -197,9 +199,8 @@ void FillingRow::initFromDefaults(size_t from_pos) void FillingRow::initStalenessRow(const Columns& base_row, size_t row_ind) { - for (size_t i = 0; i < size(); ++i) { + for (size_t i = 0; i < size(); ++i) staleness_base_row[i] = (*base_row[i])[row_ind]; - } } String FillingRow::dump() const From 4c9d865e7592985507accd7aa805647ef9335d72 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Mon, 28 Oct 2024 17:45:27 +0000 Subject: [PATCH 294/680] disable debug logs --- src/Processors/Transforms/FillingTransform.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index 635b46de3ee..7f81b86697c 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -17,7 +17,7 @@ namespace DB { -constexpr bool debug_logging_enabled = true; +constexpr bool debug_logging_enabled = false; template void logDebug(String key, const T & value, const char * separator = " : ") From 83844841b4f00a24a654ac7ce9f665c321b4df85 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Mon, 28 Oct 2024 18:04:00 +0000 Subject: [PATCH 295/680] fix test timezone --- .../03266_with_fill_staleness.reference | 163 +++++++++++++++--- .../0_stateless/03266_with_fill_staleness.sql | 2 + 2 files changed, 139 insertions(+), 26 deletions(-) diff --git a/tests/queries/0_stateless/03266_with_fill_staleness.reference b/tests/queries/0_stateless/03266_with_fill_staleness.reference index 6061ecfe400..6b090443359 100644 --- a/tests/queries/0_stateless/03266_with_fill_staleness.reference +++ b/tests/queries/0_stateless/03266_with_fill_staleness.reference @@ -1,28 +1,139 @@ add samples regular with fill -2016-06-15 23:00:00 0 -2016-06-15 23:00:01 0 -2016-06-15 23:00:02 0 -2016-06-15 23:00:03 0 -2016-06-15 23:00:04 0 -2016-06-15 23:00:05 5 -2016-06-15 23:00:06 5 -2016-06-15 23:00:07 5 -2016-06-15 23:00:08 5 -2016-06-15 23:00:09 5 -2016-06-15 23:00:10 10 -2016-06-15 23:00:11 10 -2016-06-15 23:00:12 10 -2016-06-15 23:00:13 10 -2016-06-15 23:00:14 10 -2016-06-15 23:00:15 15 -2016-06-15 23:00:16 15 -2016-06-15 23:00:17 15 -2016-06-15 23:00:18 15 -2016-06-15 23:00:19 15 -2016-06-15 23:00:20 20 -2016-06-15 23:00:21 20 -2016-06-15 23:00:22 20 -2016-06-15 23:00:23 20 -2016-06-15 23:00:24 20 -2016-06-15 23:00:25 25 +2016-06-15 23:00:00 0 original +2016-06-15 23:00:01 0 +2016-06-15 23:00:02 0 +2016-06-15 23:00:03 0 +2016-06-15 23:00:04 0 +2016-06-15 23:00:05 5 original +2016-06-15 23:00:06 5 +2016-06-15 23:00:07 5 +2016-06-15 23:00:08 5 +2016-06-15 23:00:09 5 +2016-06-15 23:00:10 10 original +2016-06-15 23:00:11 10 +2016-06-15 23:00:12 10 +2016-06-15 23:00:13 10 +2016-06-15 23:00:14 10 +2016-06-15 23:00:15 15 original +2016-06-15 23:00:16 15 +2016-06-15 23:00:17 15 +2016-06-15 23:00:18 15 +2016-06-15 23:00:19 15 +2016-06-15 23:00:20 20 original +2016-06-15 23:00:21 20 +2016-06-15 23:00:22 20 +2016-06-15 23:00:23 20 +2016-06-15 23:00:24 20 +2016-06-15 23:00:25 25 original +staleness 1 seconds +2016-06-15 23:00:00 0 original +2016-06-15 23:00:05 5 original +2016-06-15 23:00:10 10 original +2016-06-15 23:00:15 15 original +2016-06-15 23:00:20 20 original +2016-06-15 23:00:25 25 original +staleness 3 seconds +2016-06-15 23:00:00 0 original +2016-06-15 23:00:01 0 +2016-06-15 23:00:02 0 +2016-06-15 23:00:05 5 original +2016-06-15 23:00:06 5 +2016-06-15 23:00:07 5 +2016-06-15 23:00:10 10 original +2016-06-15 23:00:11 10 +2016-06-15 23:00:12 10 +2016-06-15 23:00:15 15 original +2016-06-15 23:00:16 15 +2016-06-15 23:00:17 15 +2016-06-15 23:00:20 20 original +2016-06-15 23:00:21 20 +2016-06-15 23:00:22 20 +2016-06-15 23:00:25 25 original +descending order +2016-06-15 23:00:25 25 original +2016-06-15 23:00:24 25 +2016-06-15 23:00:20 20 original +2016-06-15 23:00:19 20 +2016-06-15 23:00:15 15 original +2016-06-15 23:00:14 15 +2016-06-15 23:00:10 10 original +2016-06-15 23:00:09 10 +2016-06-15 23:00:05 5 original +2016-06-15 23:00:04 5 +2016-06-15 23:00:00 0 original +staleness with to and step +2016-06-15 23:00:00 0 original +2016-06-15 23:00:03 0 +2016-06-15 23:00:05 5 original +2016-06-15 23:00:06 5 +2016-06-15 23:00:09 5 +2016-06-15 23:00:10 10 original +2016-06-15 23:00:12 10 +2016-06-15 23:00:15 15 original +2016-06-15 23:00:18 15 +2016-06-15 23:00:20 20 original +2016-06-15 23:00:21 20 +2016-06-15 23:00:24 20 +2016-06-15 23:00:25 25 original +2016-06-15 23:00:27 25 +2016-06-15 23:00:30 25 +staleness with another regular with fill +2016-06-15 23:00:00 1970-01-01 01:00:00 0 +2016-06-15 23:00:00 1970-01-01 01:00:01 0 +2016-06-15 23:00:00 1970-01-01 01:00:02 0 +2016-06-15 23:00:00 2016-06-15 23:00:00 0 original +2016-06-15 23:00:01 1970-01-01 01:00:00 0 +2016-06-15 23:00:01 1970-01-01 01:00:01 0 +2016-06-15 23:00:01 1970-01-01 01:00:02 0 +2016-06-15 23:00:05 2016-06-15 23:00:05 5 original +2016-06-15 23:00:05 1970-01-01 01:00:01 5 +2016-06-15 23:00:05 1970-01-01 01:00:02 5 +2016-06-15 23:00:06 1970-01-01 01:00:00 5 +2016-06-15 23:00:06 1970-01-01 01:00:01 5 +2016-06-15 23:00:06 1970-01-01 01:00:02 5 +2016-06-15 23:00:10 2016-06-15 23:00:10 10 original +2016-06-15 23:00:10 1970-01-01 01:00:01 10 +2016-06-15 23:00:10 1970-01-01 01:00:02 10 +2016-06-15 23:00:11 1970-01-01 01:00:00 10 +2016-06-15 23:00:11 1970-01-01 01:00:01 10 +2016-06-15 23:00:11 1970-01-01 01:00:02 10 +2016-06-15 23:00:15 2016-06-15 23:00:15 15 original +2016-06-15 23:00:15 1970-01-01 01:00:01 15 +2016-06-15 23:00:15 1970-01-01 01:00:02 15 +2016-06-15 23:00:16 1970-01-01 01:00:00 15 +2016-06-15 23:00:16 1970-01-01 01:00:01 15 +2016-06-15 23:00:16 1970-01-01 01:00:02 15 +2016-06-15 23:00:20 2016-06-15 23:00:20 20 original +2016-06-15 23:00:20 1970-01-01 01:00:01 20 +2016-06-15 23:00:20 1970-01-01 01:00:02 20 +2016-06-15 23:00:21 1970-01-01 01:00:00 20 +2016-06-15 23:00:21 1970-01-01 01:00:01 20 +2016-06-15 23:00:21 1970-01-01 01:00:02 20 +2016-06-15 23:00:25 2016-06-15 23:00:25 25 original +2016-06-15 23:00:25 1970-01-01 01:00:01 25 +2016-06-15 23:00:25 1970-01-01 01:00:02 25 +double staleness +2016-06-15 23:00:00 2016-06-15 23:00:00 0 original +2016-06-15 23:00:00 2016-06-15 23:00:02 0 +2016-06-15 23:00:00 2016-06-15 23:00:04 0 +2016-06-15 23:00:01 1970-01-01 01:00:00 0 +2016-06-15 23:00:05 2016-06-15 23:00:05 5 original +2016-06-15 23:00:05 2016-06-15 23:00:07 5 +2016-06-15 23:00:05 2016-06-15 23:00:09 5 +2016-06-15 23:00:06 1970-01-01 01:00:00 5 +2016-06-15 23:00:10 2016-06-15 23:00:10 10 original +2016-06-15 23:00:10 2016-06-15 23:00:12 10 +2016-06-15 23:00:10 2016-06-15 23:00:14 10 +2016-06-15 23:00:11 1970-01-01 01:00:00 10 +2016-06-15 23:00:15 2016-06-15 23:00:15 15 original +2016-06-15 23:00:15 2016-06-15 23:00:17 15 +2016-06-15 23:00:15 2016-06-15 23:00:19 15 +2016-06-15 23:00:16 1970-01-01 01:00:00 15 +2016-06-15 23:00:20 2016-06-15 23:00:20 20 original +2016-06-15 23:00:20 2016-06-15 23:00:22 20 +2016-06-15 23:00:20 2016-06-15 23:00:24 20 +2016-06-15 23:00:21 1970-01-01 01:00:00 20 +2016-06-15 23:00:25 2016-06-15 23:00:25 25 original +2016-06-15 23:00:25 2016-06-15 23:00:27 25 +2016-06-15 23:00:25 2016-06-15 23:00:29 25 diff --git a/tests/queries/0_stateless/03266_with_fill_staleness.sql b/tests/queries/0_stateless/03266_with_fill_staleness.sql index 3ab9be63a08..fff702ffd83 100644 --- a/tests/queries/0_stateless/03266_with_fill_staleness.sql +++ b/tests/queries/0_stateless/03266_with_fill_staleness.sql @@ -1,3 +1,5 @@ +SET session_timezone='Europe/Amsterdam'; + DROP TABLE IF EXISTS with_fill_staleness; CREATE TABLE with_fill_staleness (a DateTime, b DateTime, c UInt64) ENGINE = MergeTree ORDER BY a; From 60f0efa67689c28bd5b155eefd3266f385822b94 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Mon, 28 Oct 2024 18:08:25 +0000 Subject: [PATCH 296/680] remove debug log --- src/Planner/Planner.cpp | 3 --- 1 file changed, 3 deletions(-) diff --git a/src/Planner/Planner.cpp b/src/Planner/Planner.cpp index f1c752aecd0..8d3c75fdabb 100644 --- a/src/Planner/Planner.cpp +++ b/src/Planner/Planner.cpp @@ -847,9 +847,6 @@ void addWithFillStepIfNeeded(QueryPlan & query_plan, interpolate_description = std::make_shared(std::move(interpolate_actions_dag), empty_aliases); } - if (interpolate_description) - LOG_DEBUG(getLogger("addWithFillStepIfNeeded"), "InterpolateDescription: {}", interpolate_description->actions.dumpDAG()); - const auto & query_context = planner_context->getQueryContext(); const Settings & settings = query_context->getSettingsRef(); auto filling_step = std::make_unique( From 64d038c4408f500ae58a6a3cdd68e99c2901faa0 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Mon, 28 Oct 2024 18:14:56 +0000 Subject: [PATCH 297/680] cleanup --- src/Analyzer/SortNode.h | 6 ++--- src/Common/FieldVisitorScale.cpp | 22 +++++++++---------- src/Common/FieldVisitorScale.h | 3 --- src/Core/Field.h | 8 ------- .../Transforms/FillingTransform.cpp | 8 ++----- 5 files changed, 16 insertions(+), 31 deletions(-) diff --git a/src/Analyzer/SortNode.h b/src/Analyzer/SortNode.h index d9086dc9ed7..6f0010abdaa 100644 --- a/src/Analyzer/SortNode.h +++ b/src/Analyzer/SortNode.h @@ -105,19 +105,19 @@ public: return children[fill_step_child_index]; } - /// Returns true if sort node has fill step, false otherwise + /// Returns true if sort node has fill staleness, false otherwise bool hasFillStaleness() const { return children[fill_staleness_child_index] != nullptr; } - /// Get fill step + /// Get fill staleness const QueryTreeNodePtr & getFillStaleness() const { return children[fill_staleness_child_index]; } - /// Get fill step + /// Get fill staleness QueryTreeNodePtr & getFillStaleness() { return children[fill_staleness_child_index]; diff --git a/src/Common/FieldVisitorScale.cpp b/src/Common/FieldVisitorScale.cpp index fdb566007c3..a6c0f6d0c5b 100644 --- a/src/Common/FieldVisitorScale.cpp +++ b/src/Common/FieldVisitorScale.cpp @@ -15,16 +15,16 @@ void FieldVisitorScale::operator() (UInt64 & x) const { x *= rhs; } void FieldVisitorScale::operator() (Float64 & x) const { x *= rhs; } void FieldVisitorScale::operator() (Null &) const { /*Do not scale anything*/ } -void FieldVisitorScale::operator() (String &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Strings"); } -void FieldVisitorScale::operator() (Array &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Arrays"); } -void FieldVisitorScale::operator() (Tuple &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Tuples"); } -void FieldVisitorScale::operator() (Map &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Maps"); } -void FieldVisitorScale::operator() (Object &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Objects"); } -void FieldVisitorScale::operator() (UUID &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply UUIDs"); } -void FieldVisitorScale::operator() (IPv4 &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply IPv4s"); } -void FieldVisitorScale::operator() (IPv6 &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply IPv6s"); } -void FieldVisitorScale::operator() (CustomType & x) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply custom type {}", x.getTypeName()); } -void FieldVisitorScale::operator() (AggregateFunctionStateData &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply AggregateFunctionStates"); } -void FieldVisitorScale::operator() (bool &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot multiply Bools"); } +void FieldVisitorScale::operator() (String &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot scale Strings"); } +void FieldVisitorScale::operator() (Array &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot scale Arrays"); } +void FieldVisitorScale::operator() (Tuple &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot scale Tuples"); } +void FieldVisitorScale::operator() (Map &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot scale Maps"); } +void FieldVisitorScale::operator() (Object &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot scale Objects"); } +void FieldVisitorScale::operator() (UUID &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot scale UUIDs"); } +void FieldVisitorScale::operator() (IPv4 &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot scale IPv4s"); } +void FieldVisitorScale::operator() (IPv6 &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot scale IPv6s"); } +void FieldVisitorScale::operator() (CustomType & x) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot scale custom type {}", x.getTypeName()); } +void FieldVisitorScale::operator() (AggregateFunctionStateData &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot scale AggregateFunctionStates"); } +void FieldVisitorScale::operator() (bool &) const { throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot scale Bools"); } } diff --git a/src/Common/FieldVisitorScale.h b/src/Common/FieldVisitorScale.h index 45bacdccc9c..90d86cc53bd 100644 --- a/src/Common/FieldVisitorScale.h +++ b/src/Common/FieldVisitorScale.h @@ -1,10 +1,7 @@ #pragma once -#include #include #include -#include "base/Decimal.h" -#include "base/extended_types.h" namespace DB { diff --git a/src/Core/Field.h b/src/Core/Field.h index 47df5c2907e..7b916d30646 100644 --- a/src/Core/Field.h +++ b/src/Core/Field.h @@ -185,14 +185,6 @@ public: return *this; } - const DecimalField & operator *= (const DecimalField & r) - { - if (scale != r.getScale()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Multiply different decimal fields"); - dec *= r.getValue(); - return *this; - } - const DecimalField & operator -= (const DecimalField & r) { if (scale != r.getScale()) diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index 7f81b86697c..46a670394a5 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -125,10 +125,6 @@ static FillColumnDescription::StepFunction getStepFunction(const Field & step, c if (jumps_count != 1) applyVisitor(FieldVisitorScale(jumps_count), shifted_step); - logDebug("field", field.dump()); - logDebug("step", step.dump()); - logDebug("shifted field", shifted_step.dump()); - applyVisitor(FieldVisitorSum(shifted_step), field); }; } @@ -684,8 +680,8 @@ void FillingTransform::transformRange( } const auto [apply, changed] = filling_row.next(next_row, /*long_jump=*/true); - logDebug("apply", apply); - logDebug("changed", changed); + logDebug("long jump apply", apply); + logDebug("long jump changed", changed); if (changed) filling_row_changed = true; From f905c804f5b5aa0c0b14e9aaab1034fa8fbbef03 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Mon, 28 Oct 2024 19:58:53 +0000 Subject: [PATCH 298/680] fix calibration jump --- src/Interpreters/FillingRow.cpp | 16 +++++----------- 1 file changed, 5 insertions(+), 11 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index fdd3b55b66b..49ee558cb20 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -153,23 +153,17 @@ std::pair FillingRow::next(const FillingRow & to_row, bool long_jump if (!next_value.has_value()) return {false, false}; - Field calibration_jump_value = next_value.value(); - fill_column_desc.step_func(calibration_jump_value, 1); - - if (equals(calibration_jump_value, to_row[pos])) - next_value = calibration_jump_value; - - if (!next_value.has_value() || less(to_row.row[pos], next_value.value(), getDirection(pos)) || equals(next_value.value(), getFillDescription(pos).fill_to)) - return {false, false}; + /// We need value >= to_row[pos] + fill_column_desc.step_func(next_value.value(), 1); } else { next_value = doJump(fill_column_desc, pos); - - if (!next_value.has_value() || less(to_row.row[pos], next_value.value(), getDirection(pos)) || equals(next_value.value(), getFillDescription(pos).fill_to)) - return {false, false}; } + if (!next_value.has_value() || less(to_row.row[pos], next_value.value(), getDirection(pos)) || equals(next_value.value(), getFillDescription(pos).fill_to)) + return {false, false}; + row[pos] = std::move(next_value.value()); if (equals(row[pos], to_row.row[pos])) { From 6772d3fe6623f73edb4509a7d6e9cbdc5e9883f9 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Mon, 28 Oct 2024 22:08:38 +0000 Subject: [PATCH 299/680] little improvement --- src/Interpreters/FillingRow.cpp | 17 ++++++++++------- src/Interpreters/FillingRow.h | 2 +- 2 files changed, 11 insertions(+), 8 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index 49ee558cb20..8c5f102bcd6 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -28,7 +28,7 @@ FillingRow::FillingRow(const SortDescription & sort_description_) : sort_description(sort_description_) { row.resize(sort_description.size()); - staleness_base_row.resize(sort_description.size()); + staleness_border.resize(sort_description.size()); } bool FillingRow::operator<(const FillingRow & other) const @@ -74,10 +74,7 @@ std::optional FillingRow::doJump(const FillColumnDescription& descr, size if (!descr.fill_staleness.isNull()) { - Field staleness_border = staleness_base_row[column_ind]; - descr.staleness_step_func(staleness_border, 1); - - if (less(next_value, staleness_border, getDirection(column_ind))) + if (less(next_value, staleness_border[column_ind], getDirection(column_ind))) return next_value; else return std::nullopt; @@ -93,7 +90,7 @@ std::optional FillingRow::doLongJump(const FillColumnDescription & descr, if (less(to, shifted_value, getDirection(column_ind))) return std::nullopt; - for (int32_t step_len = 1, step_no = 0; step_no < 100; ++step_no) + for (int32_t step_len = 1, step_no = 0; step_no < 100 && step_len > 0; ++step_no) { Field next_value = shifted_value; descr.step_func(next_value, step_len); @@ -194,7 +191,13 @@ void FillingRow::initFromDefaults(size_t from_pos) void FillingRow::initStalenessRow(const Columns& base_row, size_t row_ind) { for (size_t i = 0; i < size(); ++i) - staleness_base_row[i] = (*base_row[i])[row_ind]; + { + staleness_border[i] = (*base_row[i])[row_ind]; + + const auto& descr = getFillDescription(i); + if (!descr.fill_staleness.isNull()) + descr.staleness_step_func(staleness_border[i], 1); + } } String FillingRow::dump() const diff --git a/src/Interpreters/FillingRow.h b/src/Interpreters/FillingRow.h index 14b6034ce35..dc787173191 100644 --- a/src/Interpreters/FillingRow.h +++ b/src/Interpreters/FillingRow.h @@ -46,7 +46,7 @@ public: private: Row row; - Row staleness_base_row; + Row staleness_border; SortDescription sort_description; }; From 219cc4e5d241201d8bb4838cc440735ec5c905ea Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Tue, 29 Oct 2024 12:15:13 +0800 Subject: [PATCH 300/680] fix mismatched aggreage function name of quantileExactWeightedInterpolated --- .../AggregateFunctionQuantileExactWeighted.cpp | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/AggregateFunctions/AggregateFunctionQuantileExactWeighted.cpp b/src/AggregateFunctions/AggregateFunctionQuantileExactWeighted.cpp index 58b3b75b056..116b04bf4ba 100644 --- a/src/AggregateFunctions/AggregateFunctionQuantileExactWeighted.cpp +++ b/src/AggregateFunctions/AggregateFunctionQuantileExactWeighted.cpp @@ -387,7 +387,7 @@ template using FuncQuantileExactWeighted = AggregateFunctionQuantile< Value, QuantileExactWeighted, - NameQuantileExactWeighted, + std::conditional_t, true, std::conditional_t, false, @@ -396,7 +396,7 @@ template using FuncQuantilesExactWeighted = AggregateFunctionQuantile< Value, QuantileExactWeighted, - NameQuantilesExactWeighted, + std::conditional_t, true, std::conditional_t, true, From 190703b603fe8bfef6d92cc883f9e0107fdce83c Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Tue, 29 Oct 2024 05:32:52 +0100 Subject: [PATCH 301/680] Close #8687 --- .../03258_multiple_array_joins.reference | 8 +++++++ .../03258_multiple_array_joins.sql | 24 +++++++++++++++++++ 2 files changed, 32 insertions(+) create mode 100644 tests/queries/0_stateless/03258_multiple_array_joins.reference create mode 100644 tests/queries/0_stateless/03258_multiple_array_joins.sql diff --git a/tests/queries/0_stateless/03258_multiple_array_joins.reference b/tests/queries/0_stateless/03258_multiple_array_joins.reference new file mode 100644 index 00000000000..4d357c8ac80 --- /dev/null +++ b/tests/queries/0_stateless/03258_multiple_array_joins.reference @@ -0,0 +1,8 @@ +1 Michel Foucault alive no +1 Michel Foucault profession philosopher +1 Thomas Aquinas alive no +1 Thomas Aquinas profession philosopher +2 Nicola Tesla alive no +2 Nicola Tesla profession inventor +2 Thomas Edison alive no +2 Thomas Edison profession inventor diff --git a/tests/queries/0_stateless/03258_multiple_array_joins.sql b/tests/queries/0_stateless/03258_multiple_array_joins.sql new file mode 100644 index 00000000000..5afe7725d3f --- /dev/null +++ b/tests/queries/0_stateless/03258_multiple_array_joins.sql @@ -0,0 +1,24 @@ +DROP TABLE IF EXISTS test_multiple_array_join; + +CREATE TABLE test_multiple_array_join ( + id UInt64, + person Nested ( + name String, + surname String + ), + properties Nested ( + key String, + value String + ) +) Engine=MergeTree ORDER BY id; + +INSERT INTO test_multiple_array_join VALUES (1, ['Thomas', 'Michel'], ['Aquinas', 'Foucault'], ['profession', 'alive'], ['philosopher', 'no']); +INSERT INTO test_multiple_array_join VALUES (2, ['Thomas', 'Nicola'], ['Edison', 'Tesla'], ['profession', 'alive'], ['inventor', 'no']); + +SELECT * +FROM test_multiple_array_join +ARRAY JOIN person +ARRAY JOIN properties +ORDER BY ALL; + +DROP TABLE test_multiple_array_join; From aaba95ca8ceac01fcc22416a400d52c8a169cafd Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Tue, 29 Oct 2024 11:41:37 +0300 Subject: [PATCH 302/680] Simplify and fix limit check --- src/Interpreters/InterpreterCreateQuery.cpp | 51 ++++++--------------- 1 file changed, 15 insertions(+), 36 deletions(-) diff --git a/src/Interpreters/InterpreterCreateQuery.cpp b/src/Interpreters/InterpreterCreateQuery.cpp index f8e85733911..3a6e7bc1653 100644 --- a/src/Interpreters/InterpreterCreateQuery.cpp +++ b/src/Interpreters/InterpreterCreateQuery.cpp @@ -1950,46 +1950,25 @@ bool InterpreterCreateQuery::doCreateTable(ASTCreateQuery & create, void InterpreterCreateQuery::throwIfTooManyEntities(ASTCreateQuery & create, StoragePtr storage) const { + auto check_and_throw = [&](auto setting, CurrentMetrics::Metric metric, String setting_name, String entity_name) + { + UInt64 num_limit = getContext()->getGlobalContext()->getServerSettings()[setting]; + UInt64 attached_count = CurrentMetrics::get(metric); + if (num_limit > 0 && attached_count >= num_limit) + throw Exception(ErrorCodes::TOO_MANY_TABLES, + "Too many {}. " + "The limit (server configuration parameter `{}`) is set to {}, the current number is {}", + entity_name, setting_name, num_limit, attached_count); + }; + if (auto * replicated_storage = typeid_cast(storage.get())) - { - UInt64 num_limit = getContext()->getGlobalContext()->getServerSettings()[ServerSetting::max_replicated_table_num_to_throw]; - UInt64 attached_count = CurrentMetrics::get(CurrentMetrics::AttachedReplicatedTable); - if (attached_count >= num_limit) - throw Exception(ErrorCodes::TOO_MANY_TABLES, - "Too many replicated tables. " - "The limit (server configuration parameter `max_replicated_table_num_to_throw`) is set to {}, the current number is {}", - num_limit, attached_count); - } + check_and_throw(ServerSetting::max_replicated_table_num_to_throw, CurrentMetrics::AttachedReplicatedTable, "max_replicated_table_num_to_throw", "replicated tables"); else if (create.is_dictionary) - { - UInt64 num_limit = getContext()->getGlobalContext()->getServerSettings()[ServerSetting::max_dictionary_num_to_throw]; - UInt64 attached_count = CurrentMetrics::get(CurrentMetrics::AttachedDictionary); - if (attached_count >= num_limit) - throw Exception(ErrorCodes::TOO_MANY_TABLES, - "Too many dictionaries. " - "The limit (server configuration parameter `max_dictionary_num_to_throw`) is set to {}, the current number is {}", - num_limit, attached_count); - } + check_and_throw(ServerSetting::max_dictionary_num_to_throw, CurrentMetrics::AttachedDictionary, "max_dictionary_num_to_throw", "dictionaries"); else if (create.isView()) - { - UInt64 num_limit = getContext()->getGlobalContext()->getServerSettings()[ServerSetting::max_view_num_to_throw]; - UInt64 attached_count = CurrentMetrics::get(CurrentMetrics::AttachedView); - if (attached_count >= num_limit) - throw Exception(ErrorCodes::TOO_MANY_TABLES, - "Too many views. " - "The limit (server configuration parameter `max_view_num_to_throw`) is set to {}, the current number is {}", - num_limit, attached_count); - } + check_and_throw(ServerSetting::max_view_num_to_throw, CurrentMetrics::AttachedView, "max_view_num_to_throw", "views"); else - { - UInt64 num_limit = getContext()->getGlobalContext()->getServerSettings()[ServerSetting::max_table_num_to_throw]; - UInt64 attached_count = CurrentMetrics::get(CurrentMetrics::AttachedTable); - if (attached_count >= num_limit) - throw Exception(ErrorCodes::TOO_MANY_TABLES, - "Too many tables. " - "The limit (server configuration parameter `max_table_num_to_throw`) is set to {}, the current number is {}", - num_limit, attached_count); - } + check_and_throw(ServerSetting::max_table_num_to_throw, CurrentMetrics::AttachedTable, "max_table_num_to_throw", "tables"); } From 19c95b2f0e52bd3794d160605e24c59abc5101b1 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Tue, 29 Oct 2024 11:44:50 +0300 Subject: [PATCH 303/680] Test dictionaries --- .../test_table_db_num_limit/config/config.xml | 1 + tests/integration/test_table_db_num_limit/test.py | 12 ++++++++++++ 2 files changed, 13 insertions(+) diff --git a/tests/integration/test_table_db_num_limit/config/config.xml b/tests/integration/test_table_db_num_limit/config/config.xml index a4246c79694..bfe50325d3f 100644 --- a/tests/integration/test_table_db_num_limit/config/config.xml +++ b/tests/integration/test_table_db_num_limit/config/config.xml @@ -10,6 +10,7 @@ + 10 10 5 10 diff --git a/tests/integration/test_table_db_num_limit/test.py b/tests/integration/test_table_db_num_limit/test.py index ce981ffca3c..bcfa60e48cd 100644 --- a/tests/integration/test_table_db_num_limit/test.py +++ b/tests/integration/test_table_db_num_limit/test.py @@ -48,6 +48,18 @@ def test_table_db_limit(started_cluster): "create table default.tx (a Int32) Engine = Log" ) + # Dictionaries + for i in range(10): + node.query( + "create dictionary d{} (a Int32) primary key a source(null()) layout(flat()) lifetime(1000)".format( + i + ) + ) + + assert "TOO_MANY_TABLES" in node.query_and_get_error( + "create dictionary dx (a Int32) primary key a source(null()) layout(flat()) lifetime(1000)" + ) + # Replicated tables for i in range(10): node.query("drop table t{}".format(i)) From eac5e9883a24af86c277b674c63700763ee8c9a7 Mon Sep 17 00:00:00 2001 From: flynn Date: Tue, 29 Oct 2024 08:57:37 +0000 Subject: [PATCH 304/680] Remove StorageExternalDistributed --- src/Storages/StorageExternalDistributed.cpp | 233 -------------------- src/Storages/StorageExternalDistributed.h | 43 ---- src/Storages/registerStorages.cpp | 8 - src/TableFunctions/TableFunctionURL.cpp | 1 - 4 files changed, 285 deletions(-) delete mode 100644 src/Storages/StorageExternalDistributed.cpp delete mode 100644 src/Storages/StorageExternalDistributed.h diff --git a/src/Storages/StorageExternalDistributed.cpp b/src/Storages/StorageExternalDistributed.cpp deleted file mode 100644 index ac560b58962..00000000000 --- a/src/Storages/StorageExternalDistributed.cpp +++ /dev/null @@ -1,233 +0,0 @@ -#include - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - - -namespace DB -{ -namespace Setting -{ - extern const SettingsUInt64 glob_expansion_max_elements; - extern const SettingsUInt64 postgresql_connection_attempt_timeout; - extern const SettingsBool postgresql_connection_pool_auto_close_connection; - extern const SettingsUInt64 postgresql_connection_pool_retries; - extern const SettingsUInt64 postgresql_connection_pool_size; - extern const SettingsUInt64 postgresql_connection_pool_wait_timeout; -} - -namespace ErrorCodes -{ - extern const int BAD_ARGUMENTS; -} - -StorageExternalDistributed::StorageExternalDistributed( - const StorageID & table_id_, - std::unordered_set && shards_, - const ColumnsDescription & columns_, - const ConstraintsDescription & constraints_, - const String & comment) - : IStorage(table_id_) - , shards(shards_) -{ - StorageInMemoryMetadata storage_metadata; - storage_metadata.setColumns(columns_); - storage_metadata.setConstraints(constraints_); - storage_metadata.setComment(comment); - setInMemoryMetadata(storage_metadata); -} - -void StorageExternalDistributed::read( - QueryPlan & query_plan, - const Names & column_names, - const StorageSnapshotPtr & storage_snapshot, - SelectQueryInfo & query_info, - ContextPtr context, - QueryProcessingStage::Enum processed_stage, - size_t max_block_size, - size_t num_streams) -{ - std::vector> plans; - for (const auto & shard : shards) - { - plans.emplace_back(std::make_unique()); - shard->read( - *plans.back(), - column_names, - storage_snapshot, - query_info, - context, - processed_stage, - max_block_size, - num_streams - ); - } - - if (plans.empty()) - { - auto header = storage_snapshot->getSampleBlockForColumns(column_names); - InterpreterSelectQuery::addEmptySourceToQueryPlan(query_plan, header, query_info); - } - - if (plans.size() == 1) - { - query_plan = std::move(*plans.front()); - return; - } - - Headers input_headers; - input_headers.reserve(plans.size()); - for (auto & plan : plans) - input_headers.emplace_back(plan->getCurrentHeader()); - - auto union_step = std::make_unique(std::move(input_headers)); - query_plan.unitePlans(std::move(union_step), std::move(plans)); -} - -void registerStorageExternalDistributed(StorageFactory & factory) -{ - factory.registerStorage("ExternalDistributed", [](const StorageFactory::Arguments & args) - { - ASTs & engine_args = args.engine_args; - if (engine_args.size() < 2) - throw Exception(ErrorCodes::BAD_ARGUMENTS, - "Engine ExternalDistributed must have at least 2 arguments: " - "engine_name, named_collection and/or description"); - - auto context = args.getLocalContext(); - const auto & settings = context->getSettingsRef(); - size_t max_addresses = settings[Setting::glob_expansion_max_elements]; - auto get_addresses = [&](const std::string addresses_expr) - { - return parseRemoteDescription(addresses_expr, 0, addresses_expr.size(), ',', max_addresses); - }; - - std::unordered_set shards; - ASTs inner_engine_args(engine_args.begin() + 1, engine_args.end()); - - ASTPtr * address_arg = nullptr; - - /// If there is a named collection argument, named `addresses_expr` - for (auto & node : inner_engine_args) - { - if (ASTFunction * func = node->as(); func && func->name == "equals" && func->arguments) - { - if (ASTExpressionList * func_args = func->arguments->as(); func_args && func_args->children.size() == 2) - { - if (ASTIdentifier * arg_name = func_args->children[0]->as(); arg_name && arg_name->name() == "addresses_expr") - { - address_arg = &func_args->children[1]; - break; - } - } - } - } - - /// Otherwise it is the first argument. - if (!address_arg) - address_arg = &inner_engine_args.at(0); - - String addresses_expr = checkAndGetLiteralArgument(*address_arg, "addresses"); - Strings shards_addresses = get_addresses(addresses_expr); - - auto engine_name = checkAndGetLiteralArgument(engine_args[0], "engine_name"); - if (engine_name == "URL") - { - auto format_settings = StorageURL::getFormatSettingsFromArgs(args); - for (const auto & shard_address : shards_addresses) - { - *address_arg = std::make_shared(shard_address); - auto configuration = StorageURL::getConfiguration(inner_engine_args, context); - auto uri_options = parseRemoteDescription(shard_address, 0, shard_address.size(), '|', max_addresses); - if (uri_options.size() > 1) - { - shards.insert( - std::make_shared( - uri_options, args.table_id, configuration.format, format_settings, - args.columns, args.constraints, context, configuration.compression_method)); - } - else - { - shards.insert(std::make_shared( - shard_address, args.table_id, configuration.format, format_settings, - args.columns, args.constraints, String{}, context, configuration.compression_method)); - } - } - } -#if USE_MYSQL - else if (engine_name == "MySQL") - { - MySQLSettings mysql_settings; - for (const auto & shard_address : shards_addresses) - { - *address_arg = std::make_shared(shard_address); - auto configuration = StorageMySQL::getConfiguration(inner_engine_args, context, mysql_settings); - configuration.addresses = parseRemoteDescriptionForExternalDatabase(shard_address, max_addresses, 3306); - auto pool = createMySQLPoolWithFailover(configuration, mysql_settings); - shards.insert(std::make_shared( - args.table_id, std::move(pool), configuration.database, configuration.table, - /* replace_query = */ false, /* on_duplicate_clause = */ "", - args.columns, args.constraints, String{}, context, mysql_settings)); - } - } -#endif -#if USE_LIBPQXX - else if (engine_name == "PostgreSQL") - { - for (const auto & shard_address : shards_addresses) - { - *address_arg = std::make_shared(shard_address); - auto configuration = StoragePostgreSQL::getConfiguration(inner_engine_args, context); - configuration.addresses = parseRemoteDescriptionForExternalDatabase(shard_address, max_addresses, 5432); - auto pool = std::make_shared( - configuration, - settings[Setting::postgresql_connection_pool_size], - settings[Setting::postgresql_connection_pool_wait_timeout], - settings[Setting::postgresql_connection_pool_retries], - settings[Setting::postgresql_connection_pool_auto_close_connection], - settings[Setting::postgresql_connection_attempt_timeout]); - shards.insert(std::make_shared( - args.table_id, std::move(pool), configuration.table, args.columns, args.constraints, String{}, context)); - } - } -#endif - else - { - throw Exception( - ErrorCodes::BAD_ARGUMENTS, - "External storage engine {} is not supported for StorageExternalDistributed. " - "Supported engines are: MySQL, PostgreSQL, URL", - engine_name); - } - - return std::make_shared( - args.table_id, - std::move(shards), - args.columns, - args.constraints, - args.comment); - }, - { - .source_access_type = AccessType::SOURCES, - }); -} - -} diff --git a/src/Storages/StorageExternalDistributed.h b/src/Storages/StorageExternalDistributed.h deleted file mode 100644 index 56c7fe86f34..00000000000 --- a/src/Storages/StorageExternalDistributed.h +++ /dev/null @@ -1,43 +0,0 @@ -#pragma once - -#include "config.h" - -#include - - -namespace DB -{ - -/// Storages MySQL and PostgreSQL use ConnectionPoolWithFailover and support multiple replicas. -/// This class unites multiple storages with replicas into multiple shards with replicas. -/// A query to external database is passed to one replica on each shard, the result is united. -/// Replicas on each shard have the same priority, traversed replicas are moved to the end of the queue. -/// Similar approach is used for URL storage. -class StorageExternalDistributed final : public DB::IStorage -{ -public: - StorageExternalDistributed( - const StorageID & table_id_, - std::unordered_set && shards_, - const ColumnsDescription & columns_, - const ConstraintsDescription & constraints_, - const String & comment); - - std::string getName() const override { return "ExternalDistributed"; } - - void read( - QueryPlan & query_plan, - const Names & column_names, - const StorageSnapshotPtr & storage_snapshot, - SelectQueryInfo & query_info, - ContextPtr context, - QueryProcessingStage::Enum processed_stage, - size_t max_block_size, - size_t num_streams) override; - -private: - using Shards = std::unordered_set; - Shards shards; -}; - -} diff --git a/src/Storages/registerStorages.cpp b/src/Storages/registerStorages.cpp index cfd406ccbe2..d2c445c8706 100644 --- a/src/Storages/registerStorages.cpp +++ b/src/Storages/registerStorages.cpp @@ -93,10 +93,6 @@ void registerStoragePostgreSQL(StorageFactory & factory); void registerStorageMaterializedPostgreSQL(StorageFactory & factory); #endif -#if USE_MYSQL || USE_LIBPQXX -void registerStorageExternalDistributed(StorageFactory & factory); -#endif - #if USE_FILELOG void registerStorageFileLog(StorageFactory & factory); #endif @@ -205,10 +201,6 @@ void registerStorages(bool use_legacy_mongodb_integration [[maybe_unused]]) registerStorageMaterializedPostgreSQL(factory); #endif - #if USE_MYSQL || USE_LIBPQXX - registerStorageExternalDistributed(factory); - #endif - #if USE_SQLITE registerStorageSQLite(factory); #endif diff --git a/src/TableFunctions/TableFunctionURL.cpp b/src/TableFunctions/TableFunctionURL.cpp index 2bdc0b449e0..8f4841a992b 100644 --- a/src/TableFunctions/TableFunctionURL.cpp +++ b/src/TableFunctions/TableFunctionURL.cpp @@ -6,7 +6,6 @@ #include #include #include -#include #include #include #include From af7aa7de568063c53d849150be83ee625413dc7d Mon Sep 17 00:00:00 2001 From: divanik Date: Tue, 29 Oct 2024 10:03:02 +0000 Subject: [PATCH 305/680] Fix some bugs --- .../ObjectStorage/DataLakes/Common.cpp | 7 +++ .../DataLakes/DataLakeConfiguration.h | 2 +- .../ObjectStorage/StorageObjectStorage.cpp | 46 +++++++++++++++++-- .../ObjectStorage/StorageObjectStorage.h | 2 + .../registerStorageObjectStorage.cpp | 3 +- .../TableFunctionObjectStorage.cpp | 5 +- .../TableFunctionObjectStorageCluster.cpp | 7 +-- .../configs/config.d/filesystem_caches.xml | 1 + .../integration/test_storage_iceberg/test.py | 14 ++++-- 9 files changed, 74 insertions(+), 13 deletions(-) diff --git a/src/Storages/ObjectStorage/DataLakes/Common.cpp b/src/Storages/ObjectStorage/DataLakes/Common.cpp index 4830cc52a90..c21c0486eca 100644 --- a/src/Storages/ObjectStorage/DataLakes/Common.cpp +++ b/src/Storages/ObjectStorage/DataLakes/Common.cpp @@ -1,6 +1,9 @@ #include "Common.h" #include #include +#include +#include +#include #include namespace DB @@ -13,6 +16,10 @@ std::vector listFiles( { auto key = std::filesystem::path(configuration.getPath()) / prefix; RelativePathsWithMetadata files_with_metadata; + // time_t now = time(nullptr); + Poco::DateTime now; + std::string formatted = Poco::DateTimeFormatter::format(now, Poco::DateTimeFormat::ISO8601_FORMAT); + LOG_ERROR(&Poco::Logger::get("Inside listFiles"), "Time of files listing: {}", formatted); object_storage.listObjects(key, files_with_metadata, 0); Strings res; for (const auto & file_with_metadata : files_with_metadata) diff --git a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h index 18ff6d93c46..8a4147308f3 100644 --- a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h +++ b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h @@ -36,7 +36,7 @@ public: void update(ObjectStoragePtr object_storage, ContextPtr local_context) override { - BaseStorageConfiguration::update(object_storage, local_context); + // BaseStorageConfiguration::update(object_storage, local_context); auto new_metadata = DataLakeMetadata::create(object_storage, weak_from_this(), local_context); if (current_metadata && *current_metadata == *new_metadata) return; diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index ddc6276a8a1..6f4c0787e81 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -22,6 +22,7 @@ #include #include #include +#include "Databases/LoadingStrictnessLevel.h" #include "Storages/ColumnsDescription.h" @@ -68,6 +69,27 @@ String StorageObjectStorage::getPathSample(StorageInMemoryMetadata metadata, Con return ""; } +void printConfiguration(const Poco::Util::AbstractConfiguration & config, std::string log_name, const std::string & prefix = "") +{ + Poco::Util::AbstractConfiguration::Keys keys; + config.keys(prefix, keys); + + for (const auto & key : keys) + { + std::string fullKey = prefix.empty() ? key : (prefix + "." + key); + + if (config.hasProperty(fullKey)) + { + std::string value = config.getString(fullKey); + LOG_DEBUG(&Poco::Logger::get(log_name), "{} = {}", fullKey, value); + } + + // Recursively print sub-configurations + printConfiguration(config, fullKey, log_name); + } +} + + StorageObjectStorage::StorageObjectStorage( ConfigurationPtr configuration_, ObjectStoragePtr object_storage_, @@ -77,6 +99,7 @@ StorageObjectStorage::StorageObjectStorage( const ConstraintsDescription & constraints_, const String & comment, std::optional format_settings_, + LoadingStrictnessLevel mode, bool distributed_processing_, ASTPtr partition_by_) : IStorage(table_id_) @@ -87,11 +110,27 @@ StorageObjectStorage::StorageObjectStorage( , distributed_processing(distributed_processing_) , log(getLogger(fmt::format("Storage{}({})", configuration->getEngineName(), table_id_.getFullTableName()))) { - ColumnsDescription columns{columns_}; - LOG_DEBUG(&Poco::Logger::get("StorageObjectStorage Creation"), "Columns size {}", columns.size()); - configuration->update(object_storage, context); + // LOG_DEBUG(&Poco::Logger::get("StorageObjectStorage Creation"), "Columns size {}", columns.size()); + printConfiguration(context->getConfigRef(), "Storage create"); + try + { + // configuration->update(object_storage, context); + } + catch (...) + { + if (mode <= LoadingStrictnessLevel::CREATE) + { + throw; + } + else + { + tryLogCurrentException(__PRETTY_FUNCTION__); + return; + } + } std::string sample_path; + ColumnsDescription columns{columns_}; resolveSchemaAndFormat(columns, configuration->format, object_storage, configuration, format_settings, sample_path, context); configuration->check(context); @@ -271,6 +310,7 @@ void StorageObjectStorage::read( size_t num_streams) { configuration->update(object_storage, local_context); + printConfiguration(local_context->getConfigRef(), "Select query"); if (partition_by && configuration->withPartitionWildcard()) { throw Exception(ErrorCodes::NOT_IMPLEMENTED, diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.h b/src/Storages/ObjectStorage/StorageObjectStorage.h index dc461e5861d..6ca1613e65c 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.h +++ b/src/Storages/ObjectStorage/StorageObjectStorage.h @@ -57,6 +57,7 @@ public: const ConstraintsDescription & constraints_, const String & comment, std::optional format_settings_, + LoadingStrictnessLevel mode, bool distributed_processing_ = false, ASTPtr partition_by_ = nullptr); @@ -217,6 +218,7 @@ public: virtual void update(ObjectStoragePtr object_storage, ContextPtr local_context); + protected: virtual void fromNamedCollection(const NamedCollection & collection, ContextPtr context) = 0; virtual void fromAST(ASTs & args, ContextPtr context, bool with_structure) = 0; diff --git a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp index 9a525b4e21a..a0393ea3e6a 100644 --- a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp @@ -51,13 +51,14 @@ static std::shared_ptr createStorageObjectStorage( return std::make_shared( configuration, - configuration->createObjectStorage(context, /* is_readonly */false), + configuration->createObjectStorage(context, /* is_readonly */ false), args.getContext(), args.table_id, args.columns, args.constraints, args.comment, format_settings, + args.mode, /* distributed_processing */ false, partition_by); } diff --git a/src/TableFunctions/TableFunctionObjectStorage.cpp b/src/TableFunctions/TableFunctionObjectStorage.cpp index 66c90b15c0b..6d81269f2d7 100644 --- a/src/TableFunctions/TableFunctionObjectStorage.cpp +++ b/src/TableFunctions/TableFunctionObjectStorage.cpp @@ -117,8 +117,9 @@ StoragePtr TableFunctionObjectStorage::executeImpl( columns, ConstraintsDescription{}, String{}, - /* format_settings */std::nullopt, - /* distributed_processing */false, + /* format_settings */ std::nullopt, + /* mode */ LoadingStrictnessLevel::CREATE, + /* distributed_processing */ false, nullptr); storage->startup(); diff --git a/src/TableFunctions/TableFunctionObjectStorageCluster.cpp b/src/TableFunctions/TableFunctionObjectStorageCluster.cpp index 449bd2c8c49..5ca26aabe32 100644 --- a/src/TableFunctions/TableFunctionObjectStorageCluster.cpp +++ b/src/TableFunctions/TableFunctionObjectStorageCluster.cpp @@ -41,9 +41,10 @@ StoragePtr TableFunctionObjectStorageCluster::execute StorageID(Base::getDatabaseName(), table_name), columns, ConstraintsDescription{}, - /* comment */String{}, - /* format_settings */std::nullopt, /// No format_settings - /* distributed_processing */true, + /* comment */ String{}, + /* format_settings */ std::nullopt, /// No format_settings + /* mode */ LoadingStrictnessLevel::CREATE, + /* distributed_processing */ true, /*partition_by_=*/nullptr); } else diff --git a/tests/integration/test_storage_iceberg/configs/config.d/filesystem_caches.xml b/tests/integration/test_storage_iceberg/configs/config.d/filesystem_caches.xml index e91362640fe..3b1b2aeb37e 100644 --- a/tests/integration/test_storage_iceberg/configs/config.d/filesystem_caches.xml +++ b/tests/integration/test_storage_iceberg/configs/config.d/filesystem_caches.xml @@ -5,4 +5,5 @@ cache1 + diff --git a/tests/integration/test_storage_iceberg/test.py b/tests/integration/test_storage_iceberg/test.py index 36aba550dbd..ca78fbea667 100644 --- a/tests/integration/test_storage_iceberg/test.py +++ b/tests/integration/test_storage_iceberg/test.py @@ -6,6 +6,8 @@ import time import uuid from datetime import datetime +from logging import log + import pyspark import pytest from azure.storage.blob import BlobServiceClient @@ -856,14 +858,20 @@ def test_restart_broken_s3(started_cluster): ) minio_client.remove_bucket(bucket) + print("Before restart: ", datetime.now()) + instance.restart_clickhouse() - assert "NoSuchBucket" in instance.query_and_get_error( - f"SELECT count() FROM {TABLE_NAME}" - ) + # assert "NoSuchBucket" in instance.query_and_get_error( + # f"SELECT count() FROM {TABLE_NAME}" + # ) + + time.sleep(10) minio_client.make_bucket(bucket) + print("Before successful select: ", datetime.now()) + files = default_upload_directory( started_cluster, "s3", From b5e3df977b3799f2eaaa2590293b0271eeadc073 Mon Sep 17 00:00:00 2001 From: vdimir Date: Tue, 29 Oct 2024 12:48:44 +0000 Subject: [PATCH 306/680] finishing --- src/Interpreters/ConcurrentHashJoin.h | 6 +++++- src/Interpreters/FullSortingMergeJoin.h | 2 +- src/Interpreters/HashJoin/HashJoin.h | 2 +- src/Processors/QueryPlan/UnionStep.cpp | 3 +-- tests/clickhouse-test | 3 +-- 5 files changed, 9 insertions(+), 7 deletions(-) diff --git a/src/Interpreters/ConcurrentHashJoin.h b/src/Interpreters/ConcurrentHashJoin.h index 355218554ce..b377727a134 100644 --- a/src/Interpreters/ConcurrentHashJoin.h +++ b/src/Interpreters/ConcurrentHashJoin.h @@ -61,7 +61,11 @@ public: getNonJoinedBlocks(const Block & left_sample_block, const Block & result_sample_block, UInt64 max_block_size) const override; - bool isCloneSupported() const override { return true; } + bool isCloneSupported() const override + { + return !getTotals() && getTotalRowCount() == 0; + } + std::shared_ptr clone(const std::shared_ptr & table_join_, const Block &, const Block & right_sample_block_) const override { return std::make_shared(context, table_join_, slots, right_sample_block_, stats_collecting_params); diff --git a/src/Interpreters/FullSortingMergeJoin.h b/src/Interpreters/FullSortingMergeJoin.h index 3f1e0d59287..faa9114c618 100644 --- a/src/Interpreters/FullSortingMergeJoin.h +++ b/src/Interpreters/FullSortingMergeJoin.h @@ -36,7 +36,7 @@ public: bool isCloneSupported() const override { - return true; + return !getTotals(); } std::shared_ptr clone(const std::shared_ptr & table_join_, diff --git a/src/Interpreters/HashJoin/HashJoin.h b/src/Interpreters/HashJoin/HashJoin.h index d5abdc2ddb8..8a27961354a 100644 --- a/src/Interpreters/HashJoin/HashJoin.h +++ b/src/Interpreters/HashJoin/HashJoin.h @@ -127,7 +127,7 @@ public: bool isCloneSupported() const override { - return true; + return !getTotals() && getTotalRowCount() == 0; } std::shared_ptr clone(const std::shared_ptr & table_join_, diff --git a/src/Processors/QueryPlan/UnionStep.cpp b/src/Processors/QueryPlan/UnionStep.cpp index b7a87b27be5..d5c2469629b 100644 --- a/src/Processors/QueryPlan/UnionStep.cpp +++ b/src/Processors/QueryPlan/UnionStep.cpp @@ -34,8 +34,7 @@ UnionStep::UnionStep(Headers input_headers_, size_t max_threads_) void UnionStep::updateOutputHeader() { - if (input_headers.size() == 1 || !output_header) - output_header = checkHeaders(input_headers); + output_header = checkHeaders(input_headers); } QueryPipelineBuilderPtr UnionStep::updatePipeline(QueryPipelineBuilders pipelines, const BuildQueryPipelineSettings &) diff --git a/tests/clickhouse-test b/tests/clickhouse-test index 51496c924ac..fa565eb88a7 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -921,8 +921,7 @@ class SettingsRandomizer: "optimize_functions_to_subcolumns": lambda: random.randint(0, 1), "parallel_replicas_local_plan": lambda: random.randint(0, 1), "query_plan_join_inner_table_selection": lambda: random.choice( - ["left", "auto"] - # ["left", "auto", "right"] + ["left", "auto", "right"] ), "output_format_native_write_json_as_string": lambda: random.randint(0, 1), } From 772209e6c0bd0a124d6605a6fe6ef873df8ec161 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=9A=D0=B8=D1=80=D0=B8=D0=BB=D0=BB=20=D0=93=D0=B0=D1=80?= =?UTF-8?q?=D0=B1=D0=B0=D1=80?= Date: Tue, 29 Oct 2024 16:23:21 +0300 Subject: [PATCH 307/680] Test other replica and cleanup --- .../test_table_db_num_limit/config/config.xml | 5 ++- .../config/config1.xml | 4 ++ .../config/config2.xml | 4 ++ .../test_table_db_num_limit/test.py | 40 +++++++++++++++++-- 4 files changed, 48 insertions(+), 5 deletions(-) create mode 100644 tests/integration/test_table_db_num_limit/config/config1.xml create mode 100644 tests/integration/test_table_db_num_limit/config/config2.xml diff --git a/tests/integration/test_table_db_num_limit/config/config.xml b/tests/integration/test_table_db_num_limit/config/config.xml index bfe50325d3f..88438d51b94 100644 --- a/tests/integration/test_table_db_num_limit/config/config.xml +++ b/tests/integration/test_table_db_num_limit/config/config.xml @@ -6,13 +6,16 @@ node1 9000 + + node2 + 9000 + 10 10 - 5 10 diff --git a/tests/integration/test_table_db_num_limit/config/config1.xml b/tests/integration/test_table_db_num_limit/config/config1.xml new file mode 100644 index 00000000000..73b695f3cd6 --- /dev/null +++ b/tests/integration/test_table_db_num_limit/config/config1.xml @@ -0,0 +1,4 @@ + + 5 + + diff --git a/tests/integration/test_table_db_num_limit/config/config2.xml b/tests/integration/test_table_db_num_limit/config/config2.xml new file mode 100644 index 00000000000..e46ca03d70f --- /dev/null +++ b/tests/integration/test_table_db_num_limit/config/config2.xml @@ -0,0 +1,4 @@ + + 3 + + diff --git a/tests/integration/test_table_db_num_limit/test.py b/tests/integration/test_table_db_num_limit/test.py index bcfa60e48cd..53a644a262c 100644 --- a/tests/integration/test_table_db_num_limit/test.py +++ b/tests/integration/test_table_db_num_limit/test.py @@ -7,7 +7,15 @@ cluster = ClickHouseCluster(__file__) node = cluster.add_instance( "node1", with_zookeeper=True, - main_configs=["config/config.xml"], + macros={"replica": "r1"}, + main_configs=["config/config.xml", "config/config1.xml"], +) + +node2 = cluster.add_instance( + "node2", + with_zookeeper=True, + macros={"replica": "r2"}, + main_configs=["config/config.xml", "config/config2.xml"], ) @@ -64,15 +72,27 @@ def test_table_db_limit(started_cluster): for i in range(10): node.query("drop table t{}".format(i)) - for i in range(5): + for i in range(3): node.query( - "create table t{} (a Int32) Engine = ReplicatedMergeTree('/clickhouse/tables/t{}', 'r1') order by a".format( + "create table t{} on cluster 'cluster' (a Int32) Engine = ReplicatedMergeTree('/clickhouse/tables/t{}', '{{replica}}') order by a".format( + i, i + ) + ) + + # Test limit on other replica + assert "Too many replicated tables" in node2.query_and_get_error( + "create table tx (a Int32) Engine = ReplicatedMergeTree('/clickhouse/tables/tx', '{replica}') order by a" + ) + + for i in range(3, 5): + node.query( + "create table t{} (a Int32) Engine = ReplicatedMergeTree('/clickhouse/tables/t{}', '{{replica}}') order by a".format( i, i ) ) assert "Too many replicated tables" in node.query_and_get_error( - "create table tx (a Int32) Engine = ReplicatedMergeTree('/clickhouse/tables/tx', 'r1') order by a" + "create table tx (a Int32) Engine = ReplicatedMergeTree('/clickhouse/tables/tx', '{replica}') order by a" ) # Checks that replicated tables are also counted as regular tables @@ -82,3 +102,15 @@ def test_table_db_limit(started_cluster): assert "TOO_MANY_TABLES" in node.query_and_get_error( "create table tx (a Int32) Engine = Log" ) + + # Cleanup + for i in range(10): + node.query("drop table t{} sync".format(i)) + for i in range(3): + node2.query("drop table t{} sync".format(i)) + node.query("system drop replica 'r1' from ZKPATH '/clickhouse/tables/tx'") + node.query("system drop replica 'r2' from ZKPATH '/clickhouse/tables/tx'") + for i in range(9): + node.query("drop database db{}".format(i)) + for i in range(10): + node.query("drop dictionary d{}".format(i)) From 04f68594dcf3dccc5eaecd542e00073af39777d9 Mon Sep 17 00:00:00 2001 From: Amos Bird Date: Tue, 29 Oct 2024 21:36:43 +0800 Subject: [PATCH 308/680] Print method in clickhouse-compressor --stat. --- programs/compressor/Compressor.cpp | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/programs/compressor/Compressor.cpp b/programs/compressor/Compressor.cpp index 819f16cfd64..fc07a0adc66 100644 --- a/programs/compressor/Compressor.cpp +++ b/programs/compressor/Compressor.cpp @@ -33,12 +33,12 @@ namespace DB namespace { -/// Outputs sizes of uncompressed and compressed blocks for compressed file. +/// Outputs method, sizes of uncompressed and compressed blocks for compressed file. void checkAndWriteHeader(DB::ReadBuffer & in, DB::WriteBuffer & out) { while (!in.eof()) { - in.ignore(16); /// checksum + in.ignore(16); /// checksum char header[COMPRESSED_BLOCK_HEADER_SIZE]; in.readStrict(header, COMPRESSED_BLOCK_HEADER_SIZE); @@ -50,6 +50,13 @@ void checkAndWriteHeader(DB::ReadBuffer & in, DB::WriteBuffer & out) UInt32 size_decompressed = unalignedLoad(&header[5]); + auto method_byte = static_cast(header[0]); + auto method = magic_enum::enum_cast(method_byte); + if (method) + DB::writeText(magic_enum::enum_name(*method), out); + else + DB::writeText(fmt::format("UNKNOWN({})", method_byte), out); + DB::writeChar('\t', out); DB::writeText(size_decompressed, out); DB::writeChar('\t', out); DB::writeText(size_compressed, out); From b81e024c70cb27c41daacef6372846cd9478e654 Mon Sep 17 00:00:00 2001 From: divanik Date: Tue, 29 Oct 2024 13:54:22 +0000 Subject: [PATCH 309/680] Debug prints --- .../DataLakes/DataLakeConfiguration.h | 7 +++++-- .../ObjectStorage/StorageObjectStorage.cpp | 16 ++++++++-------- .../ObjectStorage/StorageObjectStorage.h | 2 +- 3 files changed, 14 insertions(+), 11 deletions(-) diff --git a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h index 8a4147308f3..9bb02436df1 100644 --- a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h +++ b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h @@ -34,9 +34,12 @@ public: std::string getEngineName() const override { return DataLakeMetadata::name; } - void update(ObjectStoragePtr object_storage, ContextPtr local_context) override + void update(ObjectStoragePtr object_storage, ContextPtr local_context, bool update_base) override { - // BaseStorageConfiguration::update(object_storage, local_context); + if (update_base) + { + BaseStorageConfiguratixon::update(object_storage, local_context); + } auto new_metadata = DataLakeMetadata::create(object_storage, weak_from_this(), local_context); if (current_metadata && *current_metadata == *new_metadata) return; diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index 6f4c0787e81..de5a4a08358 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -76,16 +76,16 @@ void printConfiguration(const Poco::Util::AbstractConfiguration & config, std::s for (const auto & key : keys) { - std::string fullKey = prefix.empty() ? key : (prefix + "." + key); + std::string full_key = prefix.empty() ? key : (prefix + "." + key); - if (config.hasProperty(fullKey)) + if (config.hasProperty(full_key)) { - std::string value = config.getString(fullKey); - LOG_DEBUG(&Poco::Logger::get(log_name), "{} = {}", fullKey, value); + std::string value = config.getString(full_key); + LOG_DEBUG(&Poco::Logger::get(log_name), "{} = {}", full_key, value); } // Recursively print sub-configurations - printConfiguration(config, fullKey, log_name); + printConfiguration(config, full_key, log_name); } } @@ -114,7 +114,7 @@ StorageObjectStorage::StorageObjectStorage( printConfiguration(context->getConfigRef(), "Storage create"); try { - // configuration->update(object_storage, context); + configuration->update(object_storage, context); } catch (...) { @@ -166,7 +166,7 @@ bool StorageObjectStorage::supportsSubsetOfColumns(const ContextPtr & context) c return FormatFactory::instance().checkIfFormatSupportsSubsetOfColumns(configuration->format, context, format_settings); } -void StorageObjectStorage::Configuration::update(ObjectStoragePtr object_storage_ptr, ContextPtr context) +void StorageObjectStorage::Configuration::update(ObjectStoragePtr object_storage_ptr, ContextPtr context, [[maybe_unused]] bool update_base) { IObjectStorage::ApplyNewSettingsOptions options{.allow_client_change = !isStaticConfiguration()}; object_storage_ptr->applyNewSettings(context->getConfigRef(), getTypeName() + ".", context, options); @@ -309,7 +309,7 @@ void StorageObjectStorage::read( size_t max_block_size, size_t num_streams) { - configuration->update(object_storage, local_context); + configuration->update(object_storage, local_context, true); printConfiguration(local_context->getConfigRef(), "Select query"); if (partition_by && configuration->withPartitionWildcard()) { diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.h b/src/Storages/ObjectStorage/StorageObjectStorage.h index 6ca1613e65c..3a85a2532f2 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.h +++ b/src/Storages/ObjectStorage/StorageObjectStorage.h @@ -216,7 +216,7 @@ public: String compression_method = "auto"; String structure = "auto"; - virtual void update(ObjectStoragePtr object_storage, ContextPtr local_context); + virtual void update(ObjectStoragePtr object_storage, ContextPtr local_context, [[maybe_unused]] bool update_base = false); protected: From a54df544050633074e9680049ffc315a1b143f72 Mon Sep 17 00:00:00 2001 From: divanik Date: Tue, 29 Oct 2024 15:04:30 +0000 Subject: [PATCH 310/680] Add changes --- src/Storages/ObjectStorage/StorageObjectStorage.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.h b/src/Storages/ObjectStorage/StorageObjectStorage.h index 3a85a2532f2..6ca1613e65c 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.h +++ b/src/Storages/ObjectStorage/StorageObjectStorage.h @@ -216,7 +216,7 @@ public: String compression_method = "auto"; String structure = "auto"; - virtual void update(ObjectStoragePtr object_storage, ContextPtr local_context, [[maybe_unused]] bool update_base = false); + virtual void update(ObjectStoragePtr object_storage, ContextPtr local_context); protected: From 886603d62541818f74d7e206209ef58f87c07e70 Mon Sep 17 00:00:00 2001 From: divanik Date: Tue, 29 Oct 2024 15:18:05 +0000 Subject: [PATCH 311/680] Fixed some bugs --- .../ObjectStorage/DataLakes/DataLakeConfiguration.h | 9 ++------- src/Storages/ObjectStorage/StorageObjectStorage.cpp | 4 ++-- tests/integration/test_storage_iceberg/test.py | 6 +++--- 3 files changed, 7 insertions(+), 12 deletions(-) diff --git a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h index 9bb02436df1..1a694a25dff 100644 --- a/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h +++ b/src/Storages/ObjectStorage/DataLakes/DataLakeConfiguration.h @@ -30,16 +30,11 @@ public: bool isDataLakeConfiguration() const override { return true; } - bool isStaticConfiguration() const override { return false; } - std::string getEngineName() const override { return DataLakeMetadata::name; } - void update(ObjectStoragePtr object_storage, ContextPtr local_context, bool update_base) override + void update(ObjectStoragePtr object_storage, ContextPtr local_context) override { - if (update_base) - { - BaseStorageConfiguratixon::update(object_storage, local_context); - } + BaseStorageConfiguration::update(object_storage, local_context); auto new_metadata = DataLakeMetadata::create(object_storage, weak_from_this(), local_context); if (current_metadata && *current_metadata == *new_metadata) return; diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index de5a4a08358..89a5bfe9469 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -166,7 +166,7 @@ bool StorageObjectStorage::supportsSubsetOfColumns(const ContextPtr & context) c return FormatFactory::instance().checkIfFormatSupportsSubsetOfColumns(configuration->format, context, format_settings); } -void StorageObjectStorage::Configuration::update(ObjectStoragePtr object_storage_ptr, ContextPtr context, [[maybe_unused]] bool update_base) +void StorageObjectStorage::Configuration::update(ObjectStoragePtr object_storage_ptr, ContextPtr context) { IObjectStorage::ApplyNewSettingsOptions options{.allow_client_change = !isStaticConfiguration()}; object_storage_ptr->applyNewSettings(context->getConfigRef(), getTypeName() + ".", context, options); @@ -309,7 +309,7 @@ void StorageObjectStorage::read( size_t max_block_size, size_t num_streams) { - configuration->update(object_storage, local_context, true); + configuration->update(object_storage, local_context); printConfiguration(local_context->getConfigRef(), "Select query"); if (partition_by && configuration->withPartitionWildcard()) { diff --git a/tests/integration/test_storage_iceberg/test.py b/tests/integration/test_storage_iceberg/test.py index ca78fbea667..3d93c1b163c 100644 --- a/tests/integration/test_storage_iceberg/test.py +++ b/tests/integration/test_storage_iceberg/test.py @@ -862,9 +862,9 @@ def test_restart_broken_s3(started_cluster): instance.restart_clickhouse() - # assert "NoSuchBucket" in instance.query_and_get_error( - # f"SELECT count() FROM {TABLE_NAME}" - # ) + assert "NoSuchBucket" in instance.query_and_get_error( + f"SELECT count() FROM {TABLE_NAME}" + ) time.sleep(10) From b81fadc6bfc3e69c8dc8c129de5ad6a2912db106 Mon Sep 17 00:00:00 2001 From: flynn Date: Tue, 29 Oct 2024 15:18:07 +0000 Subject: [PATCH 312/680] Remove test --- tests/integration/test_storage_mysql/test.py | 94 ------------------- .../test_storage_postgresql/test.py | 83 ---------------- 2 files changed, 177 deletions(-) diff --git a/tests/integration/test_storage_mysql/test.py b/tests/integration/test_storage_mysql/test.py index 2fc62d7f511..2d34a52c17b 100644 --- a/tests/integration/test_storage_mysql/test.py +++ b/tests/integration/test_storage_mysql/test.py @@ -386,100 +386,6 @@ CREATE TABLE {}(id UInt32, name String, age UInt32, money UInt32, source Enum8(' conn.close() -def test_mysql_distributed(started_cluster): - table_name = "test_replicas" - - conn1 = get_mysql_conn(started_cluster, started_cluster.mysql8_ip) - conn2 = get_mysql_conn(started_cluster, started_cluster.mysql2_ip) - conn3 = get_mysql_conn(started_cluster, started_cluster.mysql3_ip) - conn4 = get_mysql_conn(started_cluster, started_cluster.mysql4_ip) - - create_mysql_db(conn1, "clickhouse") - create_mysql_db(conn2, "clickhouse") - create_mysql_db(conn3, "clickhouse") - create_mysql_db(conn4, "clickhouse") - - create_mysql_table(conn1, table_name) - create_mysql_table(conn2, table_name) - create_mysql_table(conn3, table_name) - create_mysql_table(conn4, table_name) - - node2.query("DROP TABLE IF EXISTS test_replicas") - - # Storage with with 3 replicas - node2.query( - """ - CREATE TABLE test_replicas - (id UInt32, name String, age UInt32, money UInt32) - ENGINE = MySQL('mysql{2|3|4}:3306', 'clickhouse', 'test_replicas', 'root', 'clickhouse'); """ - ) - - # Fill remote tables with different data to be able to check - nodes = [node1, node2, node2, node2] - for i in range(1, 5): - nodes[i - 1].query("DROP TABLE IF EXISTS test_replica{}".format(i)) - nodes[i - 1].query( - """ - CREATE TABLE test_replica{} - (id UInt32, name String, age UInt32, money UInt32) - ENGINE = MySQL('mysql{}:3306', 'clickhouse', 'test_replicas', 'root', 'clickhouse');""".format( - i, 80 if i == 1 else i - ) - ) - nodes[i - 1].query( - "INSERT INTO test_replica{} (id, name) SELECT number, 'host{}' from numbers(10) ".format( - i, i - ) - ) - - # test multiple ports parsing - result = node2.query( - """SELECT DISTINCT(name) FROM mysql('mysql{80|2|3}:3306', 'clickhouse', 'test_replicas', 'root', 'clickhouse'); """ - ) - assert result == "host1\n" or result == "host2\n" or result == "host3\n" - result = node2.query( - """SELECT DISTINCT(name) FROM mysql('mysql80:3306|mysql2:3306|mysql3:3306', 'clickhouse', 'test_replicas', 'root', 'clickhouse'); """ - ) - assert result == "host1\n" or result == "host2\n" or result == "host3\n" - - # check all replicas are traversed - query = "SELECT * FROM (" - for i in range(3): - query += "SELECT name FROM test_replicas UNION DISTINCT " - query += "SELECT name FROM test_replicas) ORDER BY name" - - result = node2.query(query) - assert result == "host2\nhost3\nhost4\n" - - # Storage with with two shards, each has 2 replicas - node2.query("DROP TABLE IF EXISTS test_shards") - - node2.query( - """ - CREATE TABLE test_shards - (id UInt32, name String, age UInt32, money UInt32) - ENGINE = ExternalDistributed('MySQL', 'mysql{80|2}:3306,mysql{3|4}:3306', 'clickhouse', 'test_replicas', 'root', 'clickhouse'); """ - ) - - # Check only one replica in each shard is used - result = node2.query("SELECT DISTINCT(name) FROM test_shards ORDER BY name") - assert result == "host1\nhost3\n" - - # check all replicas are traversed - query = "SELECT name FROM (" - for i in range(3): - query += "SELECT name FROM test_shards UNION DISTINCT " - query += "SELECT name FROM test_shards) ORDER BY name" - result = node2.query(query) - assert result == "host1\nhost2\nhost3\nhost4\n" - - # disconnect mysql - started_cluster.pause_container("mysql80") - result = node2.query("SELECT DISTINCT(name) FROM test_shards ORDER BY name") - started_cluster.unpause_container("mysql80") - assert result == "host2\nhost4\n" or result == "host3\nhost4\n" - - def test_external_settings(started_cluster): table_name = "test_external_settings" node1.query(f"DROP TABLE IF EXISTS {table_name}") diff --git a/tests/integration/test_storage_postgresql/test.py b/tests/integration/test_storage_postgresql/test.py index aaecc7537cf..0cb551aecc5 100644 --- a/tests/integration/test_storage_postgresql/test.py +++ b/tests/integration/test_storage_postgresql/test.py @@ -449,89 +449,6 @@ def test_concurrent_queries(started_cluster): node1.query("DROP TABLE test.stat;") -def test_postgres_distributed(started_cluster): - cursor0 = started_cluster.postgres_conn.cursor() - cursor1 = started_cluster.postgres2_conn.cursor() - cursor2 = started_cluster.postgres3_conn.cursor() - cursor3 = started_cluster.postgres4_conn.cursor() - cursors = [cursor0, cursor1, cursor2, cursor3] - - for i in range(4): - cursors[i].execute("DROP TABLE IF EXISTS test_replicas") - cursors[i].execute("CREATE TABLE test_replicas (id Integer, name Text)") - cursors[i].execute( - f"""INSERT INTO test_replicas select i, 'host{i+1}' from generate_series(0, 99) as t(i);""" - ) - - # test multiple ports parsing - result = node2.query( - """SELECT DISTINCT(name) FROM postgresql('postgres{1|2|3}:5432', 'postgres', 'test_replicas', 'postgres', 'mysecretpassword'); """ - ) - assert result == "host1\n" or result == "host2\n" or result == "host3\n" - result = node2.query( - """SELECT DISTINCT(name) FROM postgresql('postgres2:5431|postgres3:5432', 'postgres', 'test_replicas', 'postgres', 'mysecretpassword'); """ - ) - assert result == "host3\n" or result == "host2\n" - - # Create storage with with 3 replicas - node2.query("DROP TABLE IF EXISTS test_replicas") - node2.query( - """ - CREATE TABLE test_replicas - (id UInt32, name String) - ENGINE = PostgreSQL('postgres{2|3|4}:5432', 'postgres', 'test_replicas', 'postgres', 'mysecretpassword'); """ - ) - - # Check all replicas are traversed - query = "SELECT name FROM (" - for i in range(3): - query += "SELECT name FROM test_replicas UNION DISTINCT " - query += "SELECT name FROM test_replicas) ORDER BY name" - result = node2.query(query) - assert result == "host2\nhost3\nhost4\n" - - # Create storage with with two two shards, each has 2 replicas - node2.query("DROP TABLE IF EXISTS test_shards") - - node2.query( - """ - CREATE TABLE test_shards - (id UInt32, name String, age UInt32, money UInt32) - ENGINE = ExternalDistributed('PostgreSQL', 'postgres{1|2}:5432,postgres{3|4}:5432', 'postgres', 'test_replicas', 'postgres', 'mysecretpassword'); """ - ) - - # Check only one replica in each shard is used - result = node2.query("SELECT DISTINCT(name) FROM test_shards ORDER BY name") - assert result == "host1\nhost3\n" - - node2.query( - """ - CREATE TABLE test_shards2 - (id UInt32, name String, age UInt32, money UInt32) - ENGINE = ExternalDistributed('PostgreSQL', postgres4, addresses_expr='postgres{1|2}:5432,postgres{3|4}:5432'); """ - ) - - result = node2.query("SELECT DISTINCT(name) FROM test_shards2 ORDER BY name") - assert result == "host1\nhost3\n" - - # Check all replicas are traversed - query = "SELECT name FROM (" - for i in range(3): - query += "SELECT name FROM test_shards UNION DISTINCT " - query += "SELECT name FROM test_shards) ORDER BY name" - result = node2.query(query) - assert result == "host1\nhost2\nhost3\nhost4\n" - - # Disconnect postgres1 - started_cluster.pause_container("postgres1") - result = node2.query("SELECT DISTINCT(name) FROM test_shards ORDER BY name") - started_cluster.unpause_container("postgres1") - assert result == "host2\nhost4\n" or result == "host3\nhost4\n" - node2.query("DROP TABLE test_shards2") - node2.query("DROP TABLE test_shards") - node2.query("DROP TABLE test_replicas") - - def test_datetime_with_timezone(started_cluster): cursor = started_cluster.postgres_conn.cursor() cursor.execute("DROP TABLE IF EXISTS test_timezone") From 9425b19f848ace4c7c183d2c36e1660986ce394d Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Tue, 29 Oct 2024 15:26:35 +0000 Subject: [PATCH 313/680] Automatic style fix --- tests/integration/test_storage_iceberg/test.py | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/integration/test_storage_iceberg/test.py b/tests/integration/test_storage_iceberg/test.py index 3d93c1b163c..690ebeeffbf 100644 --- a/tests/integration/test_storage_iceberg/test.py +++ b/tests/integration/test_storage_iceberg/test.py @@ -5,7 +5,6 @@ import os import time import uuid from datetime import datetime - from logging import log import pyspark From 7d2fc48b6d37c5120372349892f5382823cafa06 Mon Sep 17 00:00:00 2001 From: divanik Date: Tue, 29 Oct 2024 17:02:43 +0000 Subject: [PATCH 314/680] Fixed restart broken --- src/Storages/ObjectStorage/StorageObjectStorage.cpp | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index 89a5bfe9469..9fa7b669b79 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -118,14 +118,15 @@ StorageObjectStorage::StorageObjectStorage( } catch (...) { - if (mode <= LoadingStrictnessLevel::CREATE) + if (mode <= LoadingStrictnessLevel::CREATE || columns_.empty() + || (configuration->format + == "auto")) // If we don't have format or schema yet, we can't ignore failed configuration update, because relevant configuration is crucial for format and schema inference { throw; } else { tryLogCurrentException(__PRETTY_FUNCTION__); - return; } } From 9b435388deb183edc2dfee520107391e6b96a2f4 Mon Sep 17 00:00:00 2001 From: divanik Date: Tue, 29 Oct 2024 17:20:53 +0000 Subject: [PATCH 315/680] Remove useless stuff --- .../ObjectStorages/S3/S3ObjectStorage.cpp | 3 +- .../ObjectStorage/DataLakes/Common.cpp | 7 ----- .../ObjectStorage/StorageObjectStorage.cpp | 28 ++----------------- .../configs/config.d/filesystem_caches.xml | 1 - .../integration/test_storage_iceberg/test.py | 7 ----- 5 files changed, 4 insertions(+), 42 deletions(-) diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp index 44aeabc1c28..47ef97401f2 100644 --- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp +++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp @@ -501,7 +501,8 @@ void S3ObjectStorage::applyNewSettings( } auto current_settings = s3_settings.get(); - if (options.allow_client_change && (current_settings->auth_settings.hasUpdates(modified_settings->auth_settings) || for_disk_s3)) + if (options.allow_client_change + && (current_settings->auth_settings.hasUpdates(modified_settings->auth_settings) || for_disk_s3)) { auto new_client = getClient(uri, *modified_settings, context, for_disk_s3); client.set(std::move(new_client)); diff --git a/src/Storages/ObjectStorage/DataLakes/Common.cpp b/src/Storages/ObjectStorage/DataLakes/Common.cpp index c21c0486eca..4830cc52a90 100644 --- a/src/Storages/ObjectStorage/DataLakes/Common.cpp +++ b/src/Storages/ObjectStorage/DataLakes/Common.cpp @@ -1,9 +1,6 @@ #include "Common.h" #include #include -#include -#include -#include #include namespace DB @@ -16,10 +13,6 @@ std::vector listFiles( { auto key = std::filesystem::path(configuration.getPath()) / prefix; RelativePathsWithMetadata files_with_metadata; - // time_t now = time(nullptr); - Poco::DateTime now; - std::string formatted = Poco::DateTimeFormatter::format(now, Poco::DateTimeFormat::ISO8601_FORMAT); - LOG_ERROR(&Poco::Logger::get("Inside listFiles"), "Time of files listing: {}", formatted); object_storage.listObjects(key, files_with_metadata, 0); Strings res; for (const auto & file_with_metadata : files_with_metadata) diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index 9fa7b669b79..1ed6e137a31 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -69,27 +69,6 @@ String StorageObjectStorage::getPathSample(StorageInMemoryMetadata metadata, Con return ""; } -void printConfiguration(const Poco::Util::AbstractConfiguration & config, std::string log_name, const std::string & prefix = "") -{ - Poco::Util::AbstractConfiguration::Keys keys; - config.keys(prefix, keys); - - for (const auto & key : keys) - { - std::string full_key = prefix.empty() ? key : (prefix + "." + key); - - if (config.hasProperty(full_key)) - { - std::string value = config.getString(full_key); - LOG_DEBUG(&Poco::Logger::get(log_name), "{} = {}", full_key, value); - } - - // Recursively print sub-configurations - printConfiguration(config, full_key, log_name); - } -} - - StorageObjectStorage::StorageObjectStorage( ConfigurationPtr configuration_, ObjectStoragePtr object_storage_, @@ -110,17 +89,14 @@ StorageObjectStorage::StorageObjectStorage( , distributed_processing(distributed_processing_) , log(getLogger(fmt::format("Storage{}({})", configuration->getEngineName(), table_id_.getFullTableName()))) { - // LOG_DEBUG(&Poco::Logger::get("StorageObjectStorage Creation"), "Columns size {}", columns.size()); - printConfiguration(context->getConfigRef(), "Storage create"); try { configuration->update(object_storage, context); } catch (...) { - if (mode <= LoadingStrictnessLevel::CREATE || columns_.empty() - || (configuration->format - == "auto")) // If we don't have format or schema yet, we can't ignore failed configuration update, because relevant configuration is crucial for format and schema inference + // If we don't have format or schema yet, we can't ignore failed configuration update, because relevant configuration is crucial for format and schema inference + if (mode <= LoadingStrictnessLevel::CREATE || columns_.empty() || (configuration->format == "auto")) { throw; } diff --git a/tests/integration/test_storage_iceberg/configs/config.d/filesystem_caches.xml b/tests/integration/test_storage_iceberg/configs/config.d/filesystem_caches.xml index 3b1b2aeb37e..e91362640fe 100644 --- a/tests/integration/test_storage_iceberg/configs/config.d/filesystem_caches.xml +++ b/tests/integration/test_storage_iceberg/configs/config.d/filesystem_caches.xml @@ -5,5 +5,4 @@ cache1 - diff --git a/tests/integration/test_storage_iceberg/test.py b/tests/integration/test_storage_iceberg/test.py index 690ebeeffbf..36aba550dbd 100644 --- a/tests/integration/test_storage_iceberg/test.py +++ b/tests/integration/test_storage_iceberg/test.py @@ -5,7 +5,6 @@ import os import time import uuid from datetime import datetime -from logging import log import pyspark import pytest @@ -857,20 +856,14 @@ def test_restart_broken_s3(started_cluster): ) minio_client.remove_bucket(bucket) - print("Before restart: ", datetime.now()) - instance.restart_clickhouse() assert "NoSuchBucket" in instance.query_and_get_error( f"SELECT count() FROM {TABLE_NAME}" ) - time.sleep(10) - minio_client.make_bucket(bucket) - print("Before successful select: ", datetime.now()) - files = default_upload_directory( started_cluster, "s3", From 98c9afda2e48053877ec38a5dbe3eb48f0b5d8a4 Mon Sep 17 00:00:00 2001 From: divanik Date: Tue, 29 Oct 2024 17:24:30 +0000 Subject: [PATCH 316/680] Remove build ifdef issue --- src/Storages/registerStorages.cpp | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/src/Storages/registerStorages.cpp b/src/Storages/registerStorages.cpp index 4eb90955a6c..6f6d9c3148f 100644 --- a/src/Storages/registerStorages.cpp +++ b/src/Storages/registerStorages.cpp @@ -145,6 +145,10 @@ void registerStorages(bool use_legacy_mongodb_integration [[maybe_unused]]) registerStorageAzureQueue(factory); #endif +#if USE_AVRO + registerStorageIceberg(factory); +#endif + #if USE_AWS_S3 registerStorageHudi(factory); registerStorageS3Queue(factory); @@ -153,14 +157,10 @@ void registerStorages(bool use_legacy_mongodb_integration [[maybe_unused]]) registerStorageDeltaLake(factory); #endif - #if USE_AVRO - registerStorageIceberg(factory); - #endif +#endif - #endif - - #if USE_HDFS - #if USE_HIVE +#if USE_HDFS +# if USE_HIVE registerStorageHive(factory); #endif #endif From 33d986927036bcef001f220092523fd256baa350 Mon Sep 17 00:00:00 2001 From: Pavel Kruglov <48961922+Avogar@users.noreply.github.com> Date: Tue, 29 Oct 2024 19:42:43 +0100 Subject: [PATCH 317/680] Update settings.md --- docs/en/operations/settings/settings.md | 16 ---------------- 1 file changed, 16 deletions(-) diff --git a/docs/en/operations/settings/settings.md b/docs/en/operations/settings/settings.md index 821d08cad7b..e1af24a0b8e 100644 --- a/docs/en/operations/settings/settings.md +++ b/docs/en/operations/settings/settings.md @@ -717,22 +717,6 @@ Default value: 0 In CREATE TABLE statement allows specifying Variant type with similar variant types (for example, with different numeric or date types). Enabling this setting may introduce some ambiguity when working with values with similar types. -## allow_suspicious_types_in_group_by {#allow_suspicious_types_in_group_by} - -Type: Bool - -Default value: 0 - -Allows or restricts using [Variant](../../sql-reference/data-types/variant.md) and [Dynamic](../../sql-reference/data-types/dynamic.md) types in GROUP BY keys. - -## allow_suspicious_types_in_order_by {#allow_suspicious_types_in_order_by} - -Type: Bool - -Default value: 0 - -Allows or restricts using [Variant](../../sql-reference/data-types/variant.md) and [Dynamic](../../sql-reference/data-types/dynamic.md) types in ORDER BY keys. - ## allow_unrestricted_reads_from_keeper {#allow_unrestricted_reads_from_keeper} Type: Bool From 170a24a4187bda9a5bc25fa8263222e502963b10 Mon Sep 17 00:00:00 2001 From: Pavel Kruglov <48961922+Avogar@users.noreply.github.com> Date: Tue, 29 Oct 2024 19:43:13 +0100 Subject: [PATCH 318/680] Update SettingsChangesHistory.cpp --- src/Core/SettingsChangesHistory.cpp | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 169429d1c34..fc5066029e8 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -64,6 +64,8 @@ static std::initializer_list Date: Tue, 29 Oct 2024 19:44:00 +0100 Subject: [PATCH 319/680] Update settings.md --- docs/en/operations/settings/settings.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/docs/en/operations/settings/settings.md b/docs/en/operations/settings/settings.md index e1af24a0b8e..b9b81022d4f 100644 --- a/docs/en/operations/settings/settings.md +++ b/docs/en/operations/settings/settings.md @@ -9746,3 +9746,5 @@ Type: Int64 Default value: 0 Allows you to select the max window log of ZSTD (it will not be used for MergeTree family) + + From bb9355b3d3fd2748ed1877d839ff555580f1be70 Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Tue, 29 Oct 2024 22:52:36 +0100 Subject: [PATCH 320/680] stash --- src/Planner/findParallelReplicasQuery.cpp | 101 +++++++++++++++++++--- 1 file changed, 88 insertions(+), 13 deletions(-) diff --git a/src/Planner/findParallelReplicasQuery.cpp b/src/Planner/findParallelReplicasQuery.cpp index b97a9a36381..91cbc492fdc 100644 --- a/src/Planner/findParallelReplicasQuery.cpp +++ b/src/Planner/findParallelReplicasQuery.cpp @@ -17,10 +17,12 @@ #include #include #include +#include #include #include #include #include +#include "Processors/QueryPlan/SortingStep.h" namespace DB { @@ -52,22 +54,30 @@ std::stack getSupportingParallelReplicasQuery(const IQueryTre { case QueryTreeNodeType::TABLE: { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); const auto & table_node = query_tree_node->as(); const auto & storage = table_node.getStorage(); /// Here we check StorageDummy as well, to support a query tree with replaced storages. if (std::dynamic_pointer_cast(storage) || typeid_cast(storage.get())) { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); /// parallel replicas is not supported with FINAL if (table_node.getTableExpressionModifiers() && table_node.getTableExpressionModifiers()->hasFinal()) + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return {}; + } + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return res; } + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return {}; } case QueryTreeNodeType::TABLE_FUNCTION: { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return {}; } case QueryTreeNodeType::QUERY: @@ -75,6 +85,7 @@ std::stack getSupportingParallelReplicasQuery(const IQueryTre const auto & query_node_to_process = query_tree_node->as(); query_tree_node = query_node_to_process.getJoinTree().get(); res.push(&query_node_to_process); + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); break; } case QueryTreeNodeType::UNION: @@ -83,15 +94,20 @@ std::stack getSupportingParallelReplicasQuery(const IQueryTre const auto & union_queries = union_node.getQueries().getNodes(); if (union_queries.empty()) + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return {}; + } query_tree_node = union_queries.front().get(); + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); break; } case QueryTreeNodeType::ARRAY_JOIN: { const auto & array_join_node = query_tree_node->as(); query_tree_node = array_join_node.getTableExpression().get(); + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); break; } case QueryTreeNodeType::JOIN: @@ -105,9 +121,13 @@ std::stack getSupportingParallelReplicasQuery(const IQueryTre || (join_kind == JoinKind::Inner && join_strictness == JoinStrictness::All); if (!can_parallelize_join) + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return {}; + } query_tree_node = join_node.getLeftTableExpression().get(); + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); break; } default: @@ -173,75 +193,114 @@ const QueryNode * findQueryForParallelReplicas( const QueryPlan::Node * prev_checked_node = nullptr; const QueryNode * res = nullptr; + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); while (!stack.empty()) { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); const QueryNode * subquery_node = stack.top(); stack.pop(); auto it = mapping.find(subquery_node); /// This should not happen ideally. if (it == mapping.end()) - break; - - const QueryPlan::Node * curr_node = it->second; - const QueryPlan::Node * next_node_to_check = curr_node; - bool can_distribute_full_node = true; - - while (next_node_to_check && next_node_to_check != prev_checked_node) { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); + break; + } + + const QueryPlan::Node * const curr_node = it->second; + std::deque> nodes_to_check; + nodes_to_check.push_front(std::make_pair(curr_node, false)); + bool can_distribute_full_node = true; + bool in = false; + + while (!nodes_to_check.empty() /* && nodes_to_check.front() != prev_checked_node*/) + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); + const auto & [next_node_to_check, digging_into_rabbit_hole] = nodes_to_check.front(); + LOG_DEBUG( + &Poco::Logger::get("debug"), + "next_node_to_check->step->getName()={}, next_node_to_check->step->getStepDescription());={}", + next_node_to_check->step->getName(), + next_node_to_check->step->getStepDescription()); + nodes_to_check.pop_front(); const auto & children = next_node_to_check->children; auto * step = next_node_to_check->step.get(); if (children.empty()) { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); /// Found a source step. This should be possible only in the first iteration. if (prev_checked_node) - return nullptr; + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); + // return nullptr; + } - next_node_to_check = nullptr; + nodes_to_check = {}; } else if (children.size() == 1) { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); const auto * expression = typeid_cast(step); const auto * filter = typeid_cast(step); + const auto * sorting = typeid_cast(step); const auto * creating_sets = typeid_cast(step); bool allowed_creating_sets = settings[Setting::parallel_replicas_allow_in_with_subquery] && creating_sets; - if (!expression && !filter && !allowed_creating_sets) + if (!expression && !filter && !allowed_creating_sets && !(sorting && sorting->getStepDescription().contains("before JOIN"))) + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); can_distribute_full_node = false; + in = digging_into_rabbit_hole; + } - next_node_to_check = children.front(); + nodes_to_check.push_front(std::pair(children.front(), digging_into_rabbit_hole)); } else { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); const auto * join = typeid_cast(step); /// We've checked that JOIN is INNER/LEFT in query tree. /// Don't distribute UNION node. if (!join) + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return res; + } - next_node_to_check = children.front(); + for (const auto & child : children) + nodes_to_check.push_front(std::make_pair(child, true)); } } + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); + /// Current node contains steps like GROUP BY / DISTINCT /// Will try to execute query up to WithMergableStage if (!can_distribute_full_node) { /// Current query node does not contain subqueries. /// We can execute parallel replicas over storage::read. + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); if (!res) + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return nullptr; + } - return subquery_node; + return in ? res : subquery_node; } + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); + /// Query is simple enough to be fully distributed. res = subquery_node; prev_checked_node = curr_node; } + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return res; } @@ -261,16 +320,26 @@ const QueryNode * findQueryForParallelReplicas(const QueryTreeNodePtr & query_tr auto context = query_node ? query_node->getContext() : union_node->getContext(); if (!context->canUseParallelReplicasOnInitiator()) + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return nullptr; + } + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); auto stack = getSupportingParallelReplicasQuery(query_tree_node.get()); /// Empty stack means that storage does not support parallel replicas. if (stack.empty()) + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return nullptr; + } /// We don't have any subquery and storage can process parallel replicas by itself. if (stack.top() == query_tree_node.get()) + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return nullptr; + } /// This is needed to avoid infinite recursion. auto mutable_context = Context::createCopy(context); @@ -295,16 +364,22 @@ const QueryNode * findQueryForParallelReplicas(const QueryTreeNodePtr & query_tr /// Now, return a query from initial stack. if (res) { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); while (!new_stack.empty()) { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); if (res == new_stack.top()) + { + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return stack.top(); + } stack.pop(); new_stack.pop(); } } + LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return res; } From 41bd99510a3de0936ff6aab8c28f93a7f78107fb Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Tue, 29 Oct 2024 23:08:51 +0100 Subject: [PATCH 321/680] stash --- src/Planner/findParallelReplicasQuery.cpp | 74 +---------------------- 1 file changed, 2 insertions(+), 72 deletions(-) diff --git a/src/Planner/findParallelReplicasQuery.cpp b/src/Planner/findParallelReplicasQuery.cpp index 91cbc492fdc..a5d3e863521 100644 --- a/src/Planner/findParallelReplicasQuery.cpp +++ b/src/Planner/findParallelReplicasQuery.cpp @@ -17,12 +17,11 @@ #include #include #include -#include +#include #include #include #include #include -#include "Processors/QueryPlan/SortingStep.h" namespace DB { @@ -54,30 +53,22 @@ std::stack getSupportingParallelReplicasQuery(const IQueryTre { case QueryTreeNodeType::TABLE: { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); const auto & table_node = query_tree_node->as(); const auto & storage = table_node.getStorage(); /// Here we check StorageDummy as well, to support a query tree with replaced storages. if (std::dynamic_pointer_cast(storage) || typeid_cast(storage.get())) { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); /// parallel replicas is not supported with FINAL if (table_node.getTableExpressionModifiers() && table_node.getTableExpressionModifiers()->hasFinal()) - { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return {}; - } - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return res; } - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return {}; } case QueryTreeNodeType::TABLE_FUNCTION: { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return {}; } case QueryTreeNodeType::QUERY: @@ -85,7 +76,6 @@ std::stack getSupportingParallelReplicasQuery(const IQueryTre const auto & query_node_to_process = query_tree_node->as(); query_tree_node = query_node_to_process.getJoinTree().get(); res.push(&query_node_to_process); - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); break; } case QueryTreeNodeType::UNION: @@ -94,20 +84,15 @@ std::stack getSupportingParallelReplicasQuery(const IQueryTre const auto & union_queries = union_node.getQueries().getNodes(); if (union_queries.empty()) - { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return {}; - } query_tree_node = union_queries.front().get(); - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); break; } case QueryTreeNodeType::ARRAY_JOIN: { const auto & array_join_node = query_tree_node->as(); query_tree_node = array_join_node.getTableExpression().get(); - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); break; } case QueryTreeNodeType::JOIN: @@ -121,13 +106,9 @@ std::stack getSupportingParallelReplicasQuery(const IQueryTre || (join_kind == JoinKind::Inner && join_strictness == JoinStrictness::All); if (!can_parallelize_join) - { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return {}; - } query_tree_node = join_node.getLeftTableExpression().get(); - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); break; } default: @@ -190,23 +171,17 @@ const QueryNode * findQueryForParallelReplicas( const std::unordered_map & mapping, const Settings & settings) { - const QueryPlan::Node * prev_checked_node = nullptr; const QueryNode * res = nullptr; - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); while (!stack.empty()) { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); const QueryNode * subquery_node = stack.top(); stack.pop(); auto it = mapping.find(subquery_node); /// This should not happen ideally. if (it == mapping.end()) - { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); break; - } const QueryPlan::Node * const curr_node = it->second; std::deque> nodes_to_check; @@ -214,34 +189,20 @@ const QueryNode * findQueryForParallelReplicas( bool can_distribute_full_node = true; bool in = false; - while (!nodes_to_check.empty() /* && nodes_to_check.front() != prev_checked_node*/) + while (!nodes_to_check.empty()) { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); const auto & [next_node_to_check, digging_into_rabbit_hole] = nodes_to_check.front(); - LOG_DEBUG( - &Poco::Logger::get("debug"), - "next_node_to_check->step->getName()={}, next_node_to_check->step->getStepDescription());={}", - next_node_to_check->step->getName(), - next_node_to_check->step->getStepDescription()); nodes_to_check.pop_front(); const auto & children = next_node_to_check->children; auto * step = next_node_to_check->step.get(); if (children.empty()) { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); /// Found a source step. This should be possible only in the first iteration. - if (prev_checked_node) - { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); - // return nullptr; - } - nodes_to_check = {}; } else if (children.size() == 1) { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); const auto * expression = typeid_cast(step); const auto * filter = typeid_cast(step); const auto * sorting = typeid_cast(step); @@ -251,7 +212,6 @@ const QueryNode * findQueryForParallelReplicas( if (!expression && !filter && !allowed_creating_sets && !(sorting && sorting->getStepDescription().contains("before JOIN"))) { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); can_distribute_full_node = false; in = digging_into_rabbit_hole; } @@ -260,47 +220,33 @@ const QueryNode * findQueryForParallelReplicas( } else { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); const auto * join = typeid_cast(step); /// We've checked that JOIN is INNER/LEFT in query tree. /// Don't distribute UNION node. if (!join) - { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return res; - } for (const auto & child : children) nodes_to_check.push_front(std::make_pair(child, true)); } } - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); - /// Current node contains steps like GROUP BY / DISTINCT /// Will try to execute query up to WithMergableStage if (!can_distribute_full_node) { /// Current query node does not contain subqueries. /// We can execute parallel replicas over storage::read. - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); if (!res) - { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return nullptr; - } return in ? res : subquery_node; } - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); - /// Query is simple enough to be fully distributed. res = subquery_node; - prev_checked_node = curr_node; } - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return res; } @@ -320,26 +266,16 @@ const QueryNode * findQueryForParallelReplicas(const QueryTreeNodePtr & query_tr auto context = query_node ? query_node->getContext() : union_node->getContext(); if (!context->canUseParallelReplicasOnInitiator()) - { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return nullptr; - } - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); auto stack = getSupportingParallelReplicasQuery(query_tree_node.get()); /// Empty stack means that storage does not support parallel replicas. if (stack.empty()) - { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return nullptr; - } /// We don't have any subquery and storage can process parallel replicas by itself. if (stack.top() == query_tree_node.get()) - { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return nullptr; - } /// This is needed to avoid infinite recursion. auto mutable_context = Context::createCopy(context); @@ -364,22 +300,16 @@ const QueryNode * findQueryForParallelReplicas(const QueryTreeNodePtr & query_tr /// Now, return a query from initial stack. if (res) { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); while (!new_stack.empty()) { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); if (res == new_stack.top()) - { - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return stack.top(); - } stack.pop(); new_stack.pop(); } } - LOG_DEBUG(&Poco::Logger::get("debug"), "__PRETTY_FUNCTION__={}, __LINE__={}", __PRETTY_FUNCTION__, __LINE__); return res; } From 1ad1d372b2461101c1cf4d7180c1423b6424bdf0 Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Tue, 29 Oct 2024 23:08:56 +0100 Subject: [PATCH 322/680] stash --- src/Planner/findParallelReplicasQuery.cpp | 30 ++++++++++++++--------- 1 file changed, 18 insertions(+), 12 deletions(-) diff --git a/src/Planner/findParallelReplicasQuery.cpp b/src/Planner/findParallelReplicasQuery.cpp index a5d3e863521..fbcf5386620 100644 --- a/src/Planner/findParallelReplicasQuery.cpp +++ b/src/Planner/findParallelReplicasQuery.cpp @@ -171,11 +171,17 @@ const QueryNode * findQueryForParallelReplicas( const std::unordered_map & mapping, const Settings & settings) { + struct Frame + { + const QueryPlan::Node * node = nullptr; + bool inside_join = false; + }; + const QueryNode * res = nullptr; while (!stack.empty()) { - const QueryNode * subquery_node = stack.top(); + const QueryNode * const subquery_node = stack.top(); stack.pop(); auto it = mapping.find(subquery_node); @@ -183,23 +189,22 @@ const QueryNode * findQueryForParallelReplicas( if (it == mapping.end()) break; - const QueryPlan::Node * const curr_node = it->second; - std::deque> nodes_to_check; - nodes_to_check.push_front(std::make_pair(curr_node, false)); + std::stack nodes_to_check; + nodes_to_check.push({.node = it->second, .inside_join = false}); bool can_distribute_full_node = true; - bool in = false; + bool currently_inside_join = false; while (!nodes_to_check.empty()) { - const auto & [next_node_to_check, digging_into_rabbit_hole] = nodes_to_check.front(); - nodes_to_check.pop_front(); + const auto & [next_node_to_check, inside_join] = nodes_to_check.top(); + nodes_to_check.pop(); const auto & children = next_node_to_check->children; auto * step = next_node_to_check->step.get(); if (children.empty()) { /// Found a source step. This should be possible only in the first iteration. - nodes_to_check = {}; + break; } else if (children.size() == 1) { @@ -213,10 +218,10 @@ const QueryNode * findQueryForParallelReplicas( if (!expression && !filter && !allowed_creating_sets && !(sorting && sorting->getStepDescription().contains("before JOIN"))) { can_distribute_full_node = false; - in = digging_into_rabbit_hole; + currently_inside_join = inside_join; } - nodes_to_check.push_front(std::pair(children.front(), digging_into_rabbit_hole)); + nodes_to_check.push({.node = children.front(), .inside_join = inside_join}); } else { @@ -227,7 +232,7 @@ const QueryNode * findQueryForParallelReplicas( return res; for (const auto & child : children) - nodes_to_check.push_front(std::make_pair(child, true)); + nodes_to_check.push({.node = child, .inside_join = true}); } } @@ -240,7 +245,8 @@ const QueryNode * findQueryForParallelReplicas( if (!res) return nullptr; - return in ? res : subquery_node; + /// todo + return currently_inside_join ? res : subquery_node; } /// Query is simple enough to be fully distributed. From d9f427deba385b6ab708c8e57cb6caad14cfdfc4 Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Tue, 29 Oct 2024 23:33:45 +0100 Subject: [PATCH 323/680] stash --- src/Planner/PlannerJoinTree.cpp | 5 +- src/Planner/findParallelReplicasQuery.cpp | 2 +- src/Processors/QueryPlan/SortingStep.cpp | 6 +-- src/Processors/QueryPlan/SortingStep.h | 6 ++- ...rallel_replicas_join_with_totals.reference | 10 ++++ ...3254_parallel_replicas_join_with_totals.sh | 46 +++++++++++++++++++ 6 files changed, 65 insertions(+), 10 deletions(-) create mode 100644 tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.reference create mode 100755 tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.sh diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp index 39c1352c9cf..5c153f6db39 100644 --- a/src/Planner/PlannerJoinTree.cpp +++ b/src/Planner/PlannerJoinTree.cpp @@ -1555,10 +1555,7 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ SortingStep::Settings sort_settings(*query_context); auto sorting_step = std::make_unique( - plan.getCurrentHeader(), - std::move(sort_description), - 0 /*limit*/, - sort_settings); + plan.getCurrentHeader(), std::move(sort_description), 0 /*limit*/, sort_settings, true /*is_sorting_for_merge_join*/); sorting_step->setStepDescription(fmt::format("Sort {} before JOIN", join_table_side)); plan.addStep(std::move(sorting_step)); }; diff --git a/src/Planner/findParallelReplicasQuery.cpp b/src/Planner/findParallelReplicasQuery.cpp index fbcf5386620..66c7c6440c4 100644 --- a/src/Planner/findParallelReplicasQuery.cpp +++ b/src/Planner/findParallelReplicasQuery.cpp @@ -215,7 +215,7 @@ const QueryNode * findQueryForParallelReplicas( const auto * creating_sets = typeid_cast(step); bool allowed_creating_sets = settings[Setting::parallel_replicas_allow_in_with_subquery] && creating_sets; - if (!expression && !filter && !allowed_creating_sets && !(sorting && sorting->getStepDescription().contains("before JOIN"))) + if (!expression && !filter && !allowed_creating_sets && !(sorting && sorting->isSortingForMergeJoin())) { can_distribute_full_node = false; currently_inside_join = inside_join; diff --git a/src/Processors/QueryPlan/SortingStep.cpp b/src/Processors/QueryPlan/SortingStep.cpp index 5ad2f1f62d5..c15c45ee269 100644 --- a/src/Processors/QueryPlan/SortingStep.cpp +++ b/src/Processors/QueryPlan/SortingStep.cpp @@ -77,13 +77,11 @@ static ITransformingStep::Traits getTraits(size_t limit) } SortingStep::SortingStep( - const Header & input_header, - SortDescription description_, - UInt64 limit_, - const Settings & settings_) + const Header & input_header, SortDescription description_, UInt64 limit_, const Settings & settings_, bool is_sorting_for_merge_join_) : ITransformingStep(input_header, input_header, getTraits(limit_)) , type(Type::Full) , result_description(std::move(description_)) + , is_sorting_for_merge_join(is_sorting_for_merge_join_) , limit(limit_) , sort_settings(settings_) { diff --git a/src/Processors/QueryPlan/SortingStep.h b/src/Processors/QueryPlan/SortingStep.h index 6cdf626d4c8..9af591d603a 100644 --- a/src/Processors/QueryPlan/SortingStep.h +++ b/src/Processors/QueryPlan/SortingStep.h @@ -39,7 +39,8 @@ public: const Header & input_header, SortDescription description_, UInt64 limit_, - const Settings & settings_); + const Settings & settings_, + bool is_sorting_for_merge_join_ = false); /// Full with partitioning SortingStep( @@ -81,6 +82,8 @@ public: bool hasPartitions() const { return !partition_by_description.empty(); } + bool isSortingForMergeJoin() const { return is_sorting_for_merge_join; } + void convertToFinishSorting(SortDescription prefix_description, bool use_buffering_); Type getType() const { return type; } @@ -124,6 +127,7 @@ private: const SortDescription result_description; SortDescription partition_by_description; + bool is_sorting_for_merge_join = false; UInt64 limit; bool always_read_till_end = false; diff --git a/tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.reference b/tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.reference new file mode 100644 index 00000000000..f87bb786c46 --- /dev/null +++ b/tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.reference @@ -0,0 +1,10 @@ +1 1 +1 1 + +0 0 +----- +1 1 +1 1 + +0 0 +----- diff --git a/tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.sh b/tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.sh new file mode 100755 index 00000000000..d3780d12ae0 --- /dev/null +++ b/tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.sh @@ -0,0 +1,46 @@ +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + + +${CLICKHOUSE_CLIENT} --query=" +CREATE TABLE t +( + item_id UInt64, + price_sold Float32, + date Date +) +ENGINE = MergeTree +ORDER BY item_id; + +INSERT INTO t VALUES (1, 100, '1970-01-01'), (1, 200, '1970-01-02'); +" + +for enable_parallel_replicas in {0..1}; do + ${CLICKHOUSE_CLIENT} --query=" + set allow_experimental_parallel_reading_from_replicas=${enable_parallel_replicas}, cluster_for_parallel_replicas='parallel_replicas', max_parallel_replicas=100, parallel_replicas_for_non_replicated_merge_tree=1; + + SELECT * + FROM + ( + SELECT item_id + FROM t + ) AS l + LEFT JOIN + ( + SELECT item_id + FROM t + GROUP BY item_id + WITH TOTALS + ORDER BY item_id ASC + ) AS r ON l.item_id = r.item_id; + + SELECT '-----'; + " +done + +${CLICKHOUSE_CLIENT} --query=" +DROP TABLE t; +" From c5d6acf5e3ff24122518abb992e78c73954f8703 Mon Sep 17 00:00:00 2001 From: Amos Bird Date: Wed, 30 Oct 2024 09:00:18 +0800 Subject: [PATCH 324/680] Fix --- programs/compressor/Compressor.cpp | 3 +++ 1 file changed, 3 insertions(+) diff --git a/programs/compressor/Compressor.cpp b/programs/compressor/Compressor.cpp index fc07a0adc66..69936912d49 100644 --- a/programs/compressor/Compressor.cpp +++ b/programs/compressor/Compressor.cpp @@ -1,3 +1,6 @@ +/// For magic_enum to properly get enum name of DB::CompressionMethodByte +#define MAGIC_ENUM_RANGE_MAX 256 + #include #include #include From 10ee24d9a0c749624b86840f78b2bd3bbdb221d9 Mon Sep 17 00:00:00 2001 From: Amos Bird Date: Wed, 30 Oct 2024 09:41:18 +0800 Subject: [PATCH 325/680] Fix multiple codecs and add test --- programs/compressor/Compressor.cpp | 26 +++++----------- .../getCompressionCodecForFile.cpp | 31 ++++++++++++++----- src/Compression/getCompressionCodecForFile.h | 4 +++ .../03260_compressor_stat.reference | 1 + .../0_stateless/03260_compressor_stat.sh | 13 ++++++++ 5 files changed, 49 insertions(+), 26 deletions(-) create mode 100644 tests/queries/0_stateless/03260_compressor_stat.reference create mode 100755 tests/queries/0_stateless/03260_compressor_stat.sh diff --git a/programs/compressor/Compressor.cpp b/programs/compressor/Compressor.cpp index 69936912d49..7bb434d40a8 100644 --- a/programs/compressor/Compressor.cpp +++ b/programs/compressor/Compressor.cpp @@ -1,6 +1,3 @@ -/// For magic_enum to properly get enum name of DB::CompressionMethodByte -#define MAGIC_ENUM_RANGE_MAX 256 - #include #include #include @@ -14,9 +11,12 @@ #include #include #include +#include +#include #include #include #include +#include #include #include #include @@ -41,31 +41,19 @@ void checkAndWriteHeader(DB::ReadBuffer & in, DB::WriteBuffer & out) { while (!in.eof()) { - in.ignore(16); /// checksum - - char header[COMPRESSED_BLOCK_HEADER_SIZE]; - in.readStrict(header, COMPRESSED_BLOCK_HEADER_SIZE); - - UInt32 size_compressed = unalignedLoad(&header[1]); + UInt32 size_compressed; + UInt32 size_decompressed; + auto codec = DB::getCompressionCodecForFile(in, size_compressed, size_decompressed, true /* skip_to_next_block */); if (size_compressed > DBMS_MAX_COMPRESSED_SIZE) throw DB::Exception(DB::ErrorCodes::TOO_LARGE_SIZE_COMPRESSED, "Too large size_compressed. Most likely corrupted data."); - UInt32 size_decompressed = unalignedLoad(&header[5]); - - auto method_byte = static_cast(header[0]); - auto method = magic_enum::enum_cast(method_byte); - if (method) - DB::writeText(magic_enum::enum_name(*method), out); - else - DB::writeText(fmt::format("UNKNOWN({})", method_byte), out); + DB::writeText(queryToString(codec->getFullCodecDesc()), out); DB::writeChar('\t', out); DB::writeText(size_decompressed, out); DB::writeChar('\t', out); DB::writeText(size_compressed, out); DB::writeChar('\n', out); - - in.ignore(size_compressed - COMPRESSED_BLOCK_HEADER_SIZE); } } diff --git a/src/Compression/getCompressionCodecForFile.cpp b/src/Compression/getCompressionCodecForFile.cpp index 027ee0ac57a..b04e4b6371a 100644 --- a/src/Compression/getCompressionCodecForFile.cpp +++ b/src/Compression/getCompressionCodecForFile.cpp @@ -10,33 +10,50 @@ namespace DB { - using Checksum = CityHash_v1_0_2::uint128; -CompressionCodecPtr getCompressionCodecForFile(const IDataPartStorage & data_part_storage, const String & relative_path) +CompressionCodecPtr +getCompressionCodecForFile(ReadBuffer & read_buffer, UInt32 & size_compressed, UInt32 & size_decompressed, bool skip_to_next_block) { - auto read_buffer = data_part_storage.readFile(relative_path, {}, std::nullopt, std::nullopt); - read_buffer->ignore(sizeof(Checksum)); + read_buffer.ignore(sizeof(Checksum)); UInt8 header_size = ICompressionCodec::getHeaderSize(); + size_t starting_bytes = read_buffer.count(); PODArray compressed_buffer; compressed_buffer.resize(header_size); - read_buffer->readStrict(compressed_buffer.data(), header_size); + read_buffer.readStrict(compressed_buffer.data(), header_size); uint8_t method = ICompressionCodec::readMethod(compressed_buffer.data()); + size_compressed = unalignedLoad(&compressed_buffer[1]); + size_decompressed = unalignedLoad(&compressed_buffer[5]); if (method == static_cast(CompressionMethodByte::Multiple)) { compressed_buffer.resize(1); - read_buffer->readStrict(compressed_buffer.data(), 1); + read_buffer.readStrict(compressed_buffer.data(), 1); compressed_buffer.resize(1 + compressed_buffer[0]); - read_buffer->readStrict(compressed_buffer.data() + 1, compressed_buffer[0]); + read_buffer.readStrict(compressed_buffer.data() + 1, compressed_buffer[0]); auto codecs_bytes = CompressionCodecMultiple::getCodecsBytesFromData(compressed_buffer.data()); Codecs codecs; for (auto byte : codecs_bytes) codecs.push_back(CompressionCodecFactory::instance().get(byte)); + if (skip_to_next_block) + read_buffer.ignore(size_compressed - (read_buffer.count() - starting_bytes)); + return std::make_shared(codecs); } + + if (skip_to_next_block) + read_buffer.ignore(size_compressed - (read_buffer.count() - starting_bytes)); + return CompressionCodecFactory::instance().get(method); } +CompressionCodecPtr getCompressionCodecForFile(const IDataPartStorage & data_part_storage, const String & relative_path) +{ + auto read_buffer = data_part_storage.readFile(relative_path, {}, std::nullopt, std::nullopt); + UInt32 size_compressed; + UInt32 size_decompressed; + return getCompressionCodecForFile(*read_buffer, size_compressed, size_decompressed, false); +} + } diff --git a/src/Compression/getCompressionCodecForFile.h b/src/Compression/getCompressionCodecForFile.h index b6f22750e4d..535befa37e1 100644 --- a/src/Compression/getCompressionCodecForFile.h +++ b/src/Compression/getCompressionCodecForFile.h @@ -13,4 +13,8 @@ namespace DB /// from metadata. CompressionCodecPtr getCompressionCodecForFile(const IDataPartStorage & data_part_storage, const String & relative_path); +/// Same as above which is used by clickhouse-compressor to print compression statistics of each data block. +CompressionCodecPtr +getCompressionCodecForFile(ReadBuffer & read_buffer, UInt32 & size_compressed, UInt32 & size_decompressed, bool skip_to_next_block); + } diff --git a/tests/queries/0_stateless/03260_compressor_stat.reference b/tests/queries/0_stateless/03260_compressor_stat.reference new file mode 100644 index 00000000000..ba84b26cc48 --- /dev/null +++ b/tests/queries/0_stateless/03260_compressor_stat.reference @@ -0,0 +1 @@ +CODEC(Delta(1), LZ4) 14 48 diff --git a/tests/queries/0_stateless/03260_compressor_stat.sh b/tests/queries/0_stateless/03260_compressor_stat.sh new file mode 100755 index 00000000000..6efa7b6ee0a --- /dev/null +++ b/tests/queries/0_stateless/03260_compressor_stat.sh @@ -0,0 +1,13 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +echo "Hello, World!" > 03260_test_data + +$CLICKHOUSE_COMPRESSOR --codec 'Delta' --codec 'LZ4' --input '03260_test_data' --output '03260_test_out' + +$CLICKHOUSE_COMPRESSOR --stat '03260_test_out' + +rm -f 03260_test_data 03260_test_out From bd9cfaecea93dc3b6d469f3898fcde9506ae5f9b Mon Sep 17 00:00:00 2001 From: Amos Bird Date: Wed, 30 Oct 2024 14:35:06 +0800 Subject: [PATCH 326/680] No need to create tmp files --- tests/queries/0_stateless/03260_compressor_stat.sh | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/tests/queries/0_stateless/03260_compressor_stat.sh b/tests/queries/0_stateless/03260_compressor_stat.sh index 6efa7b6ee0a..8a03541763c 100755 --- a/tests/queries/0_stateless/03260_compressor_stat.sh +++ b/tests/queries/0_stateless/03260_compressor_stat.sh @@ -4,10 +4,4 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -echo "Hello, World!" > 03260_test_data - -$CLICKHOUSE_COMPRESSOR --codec 'Delta' --codec 'LZ4' --input '03260_test_data' --output '03260_test_out' - -$CLICKHOUSE_COMPRESSOR --stat '03260_test_out' - -rm -f 03260_test_data 03260_test_out +echo "Hello, World!" | $CLICKHOUSE_COMPRESSOR --codec 'Delta' --codec 'LZ4' | $CLICKHOUSE_COMPRESSOR --stat From ba9587c728d7af72f01618e44e58dfe9cc156e06 Mon Sep 17 00:00:00 2001 From: divanik Date: Wed, 30 Oct 2024 10:34:12 +0000 Subject: [PATCH 327/680] Removed trash --- src/Storages/ObjectStorage/StorageObjectStorage.cpp | 1 - 1 file changed, 1 deletion(-) diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index 1ed6e137a31..a72fd16abc2 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -287,7 +287,6 @@ void StorageObjectStorage::read( size_t num_streams) { configuration->update(object_storage, local_context); - printConfiguration(local_context->getConfigRef(), "Select query"); if (partition_by && configuration->withPartitionWildcard()) { throw Exception(ErrorCodes::NOT_IMPLEMENTED, From 623b2f11d30af6ff0d00caa56bffbd4590bc4fff Mon Sep 17 00:00:00 2001 From: flynn Date: Wed, 30 Oct 2024 02:40:51 +0000 Subject: [PATCH 328/680] Fix test --- tests/integration/test_storage_postgresql/test.py | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tests/integration/test_storage_postgresql/test.py b/tests/integration/test_storage_postgresql/test.py index 0cb551aecc5..78bb1167d79 100644 --- a/tests/integration/test_storage_postgresql/test.py +++ b/tests/integration/test_storage_postgresql/test.py @@ -767,6 +767,7 @@ def test_filter_pushdown(started_cluster): "INSERT INTO test_filter_pushdown.test_table VALUES (1, 10), (1, 110), (2, 0), (3, 33), (4, 0)" ) + node1.query("DROP TABLE IF EXISTS test_filter_pushdown_pg_table") node1.query( """ CREATE TABLE test_filter_pushdown_pg_table (id UInt32, value UInt32) @@ -774,12 +775,14 @@ def test_filter_pushdown(started_cluster): """ ) + node1.query("DROP TABLE IF EXISTS test_filter_pushdown_local_table") node1.query( """ CREATE TABLE test_filter_pushdown_local_table (id UInt32, value UInt32) ENGINE Memory AS SELECT * FROM test_filter_pushdown_pg_table """ ) + node1.query("DROP TABLE IF EXISTS ch_table") node1.query( "CREATE TABLE ch_table (id UInt32, pg_id UInt32) ENGINE MergeTree ORDER BY id" ) From e3890a9de103a560a9804dc4b1fb63c0eb68a569 Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Wed, 30 Oct 2024 11:12:21 +0000 Subject: [PATCH 329/680] Disable virtual row better. --- .../QueryPlan/Optimizations/applyOrder.cpp | 2 +- .../Optimizations/optimizeReadInOrder.cpp | 16 ++++++++++------ src/Processors/QueryPlan/SortingStep.cpp | 8 ++++++-- src/Processors/QueryPlan/SortingStep.h | 3 ++- 4 files changed, 19 insertions(+), 10 deletions(-) diff --git a/src/Processors/QueryPlan/Optimizations/applyOrder.cpp b/src/Processors/QueryPlan/Optimizations/applyOrder.cpp index 8695f29c26b..51a5aa099ac 100644 --- a/src/Processors/QueryPlan/Optimizations/applyOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/applyOrder.cpp @@ -124,7 +124,7 @@ SortingProperty applyOrder(QueryPlan::Node * parent, SortingProperty * propertie auto common_prefix = commonPrefix(properties->sort_description, sorting_step->getSortDescription()); if (!common_prefix.empty()) /// Buffering is useful for reading from MergeTree, and it is applied in optimizeReadInOrder only. - sorting_step->convertToFinishSorting(common_prefix, /*use_buffering*/ false); + sorting_step->convertToFinishSorting(common_prefix, /*use_buffering*/ false, false); } auto scope = sorting_step->hasPartitions() ? SortingProperty::SortScope::Stream : SortingProperty::SortScope::Global; diff --git a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp index 7d9e1a7c5f7..9cb9db8eebe 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeReadInOrder.cpp @@ -899,7 +899,7 @@ InputOrder buildInputOrderFromUnorderedKeys( return order_info; } -InputOrderInfoPtr buildInputOrderInfo(SortingStep & sorting, QueryPlan::Node & node) +InputOrderInfoPtr buildInputOrderInfo(SortingStep & sorting, bool & apply_virtual_row, QueryPlan::Node & node) { QueryPlan::Node * reading_node = findReadingStep(node, /*allow_existing_order=*/ false); if (!reading_node) @@ -925,6 +925,8 @@ InputOrderInfoPtr buildInputOrderInfo(SortingStep & sorting, QueryPlan::Node & n if (order_info.input_order) { + apply_virtual_row = order_info.virtual_row_conversion != std::nullopt; + bool can_read = reading->requestReadingInOrder( order_info.input_order->used_prefix_of_sorting_key_size, order_info.input_order->direction, @@ -1128,6 +1130,8 @@ void optimizeReadInOrder(QueryPlan::Node & node, QueryPlan::Nodes & nodes) if (sorting->getType() != SortingStep::Type::Full) return; + bool apply_virtual_row = false; + if (typeid_cast(node.children.front()->step.get())) { auto & union_node = node.children.front(); @@ -1150,7 +1154,7 @@ void optimizeReadInOrder(QueryPlan::Node & node, QueryPlan::Nodes & nodes) for (auto * child : union_node->children) { - infos.push_back(buildInputOrderInfo(*sorting, *child)); + infos.push_back(buildInputOrderInfo(*sorting, apply_virtual_row, *child)); if (infos.back()) { @@ -1202,13 +1206,13 @@ void optimizeReadInOrder(QueryPlan::Node & node, QueryPlan::Nodes & nodes) } } - sorting->convertToFinishSorting(*max_sort_descr, use_buffering); + sorting->convertToFinishSorting(*max_sort_descr, use_buffering, false); } - else if (auto order_info = buildInputOrderInfo(*sorting, *node.children.front())) + else if (auto order_info = buildInputOrderInfo(*sorting, apply_virtual_row, *node.children.front())) { /// Use buffering only if have filter or don't have limit. bool use_buffering = order_info->limit == 0; - sorting->convertToFinishSorting(order_info->sort_description_for_merging, use_buffering); + sorting->convertToFinishSorting(order_info->sort_description_for_merging, use_buffering, apply_virtual_row); } } @@ -1350,7 +1354,7 @@ size_t tryReuseStorageOrderingForWindowFunctions(QueryPlan::Node * parent_node, bool can_read = read_from_merge_tree->requestReadingInOrder(order_info->used_prefix_of_sorting_key_size, order_info->direction, order_info->limit, {}); if (!can_read) return 0; - sorting->convertToFinishSorting(order_info->sort_description_for_merging, false); + sorting->convertToFinishSorting(order_info->sort_description_for_merging, false, false); } return 0; diff --git a/src/Processors/QueryPlan/SortingStep.cpp b/src/Processors/QueryPlan/SortingStep.cpp index 5ad2f1f62d5..5f0e54faf18 100644 --- a/src/Processors/QueryPlan/SortingStep.cpp +++ b/src/Processors/QueryPlan/SortingStep.cpp @@ -147,11 +147,12 @@ void SortingStep::updateLimit(size_t limit_) } } -void SortingStep::convertToFinishSorting(SortDescription prefix_description_, bool use_buffering_) +void SortingStep::convertToFinishSorting(SortDescription prefix_description_, bool use_buffering_, bool apply_virtual_row_conversions_) { type = Type::FinishSorting; prefix_description = std::move(prefix_description_); use_buffering = use_buffering_; + apply_virtual_row_conversions = apply_virtual_row_conversions_; } void SortingStep::scatterByPartitionIfNeeded(QueryPipelineBuilder& pipeline) @@ -255,7 +256,10 @@ void SortingStep::mergingSorted(QueryPipelineBuilder & pipeline, const SortDescr /*max_block_size_bytes=*/0, SortingQueueStrategy::Batch, limit_, - always_read_till_end); + always_read_till_end, + nullptr, + false, + apply_virtual_row_conversions); pipeline.addTransform(std::move(transform)); } diff --git a/src/Processors/QueryPlan/SortingStep.h b/src/Processors/QueryPlan/SortingStep.h index 6cdf626d4c8..9366630f0fb 100644 --- a/src/Processors/QueryPlan/SortingStep.h +++ b/src/Processors/QueryPlan/SortingStep.h @@ -81,7 +81,7 @@ public: bool hasPartitions() const { return !partition_by_description.empty(); } - void convertToFinishSorting(SortDescription prefix_description, bool use_buffering_); + void convertToFinishSorting(SortDescription prefix_description, bool use_buffering_, bool apply_virtual_row_conversions_); Type getType() const { return type; } const Settings & getSettings() const { return sort_settings; } @@ -128,6 +128,7 @@ private: UInt64 limit; bool always_read_till_end = false; bool use_buffering = false; + bool apply_virtual_row_conversions = false; Settings sort_settings; }; From e7fe8fed22db3c8772f9b6fe1bd9eb233e50c36c Mon Sep 17 00:00:00 2001 From: divanik Date: Wed, 30 Oct 2024 11:13:03 +0000 Subject: [PATCH 330/680] Added flag for parquet files --- .../registerStorageObjectStorage.cpp | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp index a0393ea3e6a..e94f1860176 100644 --- a/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/registerStorageObjectStorage.cpp @@ -16,10 +16,14 @@ namespace ErrorCodes extern const int BAD_ARGUMENTS; } -static std::shared_ptr createStorageObjectStorage( - const StorageFactory::Arguments & args, - StorageObjectStorage::ConfigurationPtr configuration, - ContextPtr context) +namespace +{ + +// LocalObjectStorage is only supported for Iceberg Datalake operations where Avro format is required. For regular file access, use FileStorage instead. +#if USE_AWS_S3 || USE_AZURE_BLOB_STORAGE || USE_HDFS || USE_AVRO + +std::shared_ptr +createStorageObjectStorage(const StorageFactory::Arguments & args, StorageObjectStorage::ConfigurationPtr configuration, ContextPtr context) { auto & engine_args = args.engine_args; if (engine_args.empty()) @@ -63,6 +67,9 @@ static std::shared_ptr createStorageObjectStorage( partition_by); } +#endif +} + #if USE_AZURE_BLOB_STORAGE void registerStorageAzure(StorageFactory & factory) { From 5e2355b1231774c7f3525c296df0e56ecb3d9c9f Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Wed, 30 Oct 2024 13:01:20 +0100 Subject: [PATCH 331/680] better --- src/Planner/findParallelReplicasQuery.cpp | 23 ++++++++++++------- src/Processors/QueryPlan/SortingStep.h | 2 ++ ...3254_parallel_replicas_join_with_totals.sh | 2 ++ 3 files changed, 19 insertions(+), 8 deletions(-) diff --git a/src/Planner/findParallelReplicasQuery.cpp b/src/Planner/findParallelReplicasQuery.cpp index 66c7c6440c4..8a806045111 100644 --- a/src/Planner/findParallelReplicasQuery.cpp +++ b/src/Planner/findParallelReplicasQuery.cpp @@ -174,6 +174,14 @@ const QueryNode * findQueryForParallelReplicas( struct Frame { const QueryPlan::Node * node = nullptr; + /// Below we will check subqueries from `stack` to find outtermost subquery that could be executed remotely. + /// Currently traversal algorithm considers only steps with 0 or 1 children and JOIN specifically. + /// When we found some step that requires finalization on the initiator (e.g. GROUP BY) there are two options: + /// 1. If plan looks like a single path (e.g. AggregatingStep -> ExpressionStep -> Reading) we can execute + /// current subquery as a whole with replicas. + /// 2. If we were inside JOIN we cannot offload the whole subquery to replicas because at least one side + /// of the JOIN needs to be finalized on the initiator. + /// So this flag is used to track what subquery to return once we hit a step that needs finalization. bool inside_join = false; }; @@ -203,19 +211,21 @@ const QueryNode * findQueryForParallelReplicas( if (children.empty()) { - /// Found a source step. This should be possible only in the first iteration. - break; + /// Found a source step. } else if (children.size() == 1) { const auto * expression = typeid_cast(step); const auto * filter = typeid_cast(step); - const auto * sorting = typeid_cast(step); const auto * creating_sets = typeid_cast(step); - bool allowed_creating_sets = settings[Setting::parallel_replicas_allow_in_with_subquery] && creating_sets; + const bool allowed_creating_sets = settings[Setting::parallel_replicas_allow_in_with_subquery] && creating_sets; - if (!expression && !filter && !allowed_creating_sets && !(sorting && sorting->isSortingForMergeJoin())) + const auto * sorting = typeid_cast(step); + /// Sorting for merge join is supposed to be done locally before join itself, so it doesn't need finalization. + const bool allowed_sorting = sorting && sorting->isSortingForMergeJoin(); + + if (!expression && !filter && !allowed_creating_sets && !allowed_sorting) { can_distribute_full_node = false; currently_inside_join = inside_join; @@ -236,8 +246,6 @@ const QueryNode * findQueryForParallelReplicas( } } - /// Current node contains steps like GROUP BY / DISTINCT - /// Will try to execute query up to WithMergableStage if (!can_distribute_full_node) { /// Current query node does not contain subqueries. @@ -245,7 +253,6 @@ const QueryNode * findQueryForParallelReplicas( if (!res) return nullptr; - /// todo return currently_inside_join ? res : subquery_node; } diff --git a/src/Processors/QueryPlan/SortingStep.h b/src/Processors/QueryPlan/SortingStep.h index 9af591d603a..be2e4b0149c 100644 --- a/src/Processors/QueryPlan/SortingStep.h +++ b/src/Processors/QueryPlan/SortingStep.h @@ -127,6 +127,8 @@ private: const SortDescription result_description; SortDescription partition_by_description; + + /// See `findQueryForParallelReplicas` bool is_sorting_for_merge_join = false; UInt64 limit; diff --git a/tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.sh b/tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.sh index d3780d12ae0..365d7abed7a 100755 --- a/tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.sh +++ b/tests/queries/0_stateless/03254_parallel_replicas_join_with_totals.sh @@ -20,6 +20,8 @@ INSERT INTO t VALUES (1, 100, '1970-01-01'), (1, 200, '1970-01-02'); for enable_parallel_replicas in {0..1}; do ${CLICKHOUSE_CLIENT} --query=" + --- Old analyzer uses different code path and it produces wrong result in this case. + set enable_analyzer=1; set allow_experimental_parallel_reading_from_replicas=${enable_parallel_replicas}, cluster_for_parallel_replicas='parallel_replicas', max_parallel_replicas=100, parallel_replicas_for_non_replicated_merge_tree=1; SELECT * From 0dcb2b9c2c61674be298b706498763e8fcae7018 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 12:24:39 +0000 Subject: [PATCH 332/680] try another approach --- src/Interpreters/FillingRow.cpp | 315 +++++++++++++++--- src/Interpreters/FillingRow.h | 18 +- .../Transforms/FillingTransform.cpp | 92 +++-- 3 files changed, 348 insertions(+), 77 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index 8c5f102bcd6..caf6ad9e3ba 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -1,4 +1,7 @@ +#include #include +#include "Common/Logger.h" +#include "Common/logger_useful.h" #include #include @@ -95,108 +98,326 @@ std::optional FillingRow::doLongJump(const FillColumnDescription & descr, Field next_value = shifted_value; descr.step_func(next_value, step_len); - if (less(next_value, to, getDirection(0))) + // if (less(next_value, to, getDirection(0))) + // { + // shifted_value = std::move(next_value); + // step_len *= 2; + // } + // else + // { + // step_len /= 2; + // } + + if (less(to, next_value, getDirection(0))) { - shifted_value = std::move(next_value); - step_len *= 2; + step_len /= 2; } else { - step_len /= 2; + shifted_value = std::move(next_value); + step_len *= 2; } } return shifted_value; } -std::pair FillingRow::next(const FillingRow & to_row, bool long_jump) +Field findMin(Field a, Field b, Field c, int dir) { + auto logger = getLogger("FillingRow"); + LOG_DEBUG(logger, "a: {} b: {} c: {}", a.dump(), b.dump(), c.dump()); + + if (a.isNull() || (!b.isNull() && less(b, a, dir))) + a = b; + + if (a.isNull() || (!c.isNull() && less(c, a, dir))) + a = c; + + return a; +} + +std::pair FillingRow::next(const FillingRow & next_original_row) +{ + auto logger = getLogger("FillingRow"); + const size_t row_size = size(); size_t pos = 0; /// Find position we need to increment for generating next row. for (; pos < row_size; ++pos) - if (!row[pos].isNull() && !to_row.row[pos].isNull() && !equals(row[pos], to_row.row[pos])) - break; + { + if (row[pos].isNull()) + continue; - if (pos == row_size || less(to_row.row[pos], row[pos], getDirection(pos))) + const auto & descr = getFillDescription(pos); + auto min_constr = findMin(next_original_row[pos], staleness_border[pos], descr.fill_to, getDirection(pos)); + LOG_DEBUG(logger, "min_constr: {}", min_constr); + + if (!min_constr.isNull() && !equals(row[pos], min_constr)) + break; + } + + LOG_DEBUG(logger, "pos: {}", pos); + + if (pos == row_size) return {false, false}; - /// If we have any 'fill_to' value at position greater than 'pos', - /// we need to generate rows up to 'fill_to' value. + const auto & pos_descr = getFillDescription(pos); + + if (!next_original_row[pos].isNull() && less(next_original_row[pos], row[pos], getDirection(pos))) + return {false, false}; + + if (!staleness_border[pos].isNull() && !less(row[pos], staleness_border[pos], getDirection(pos))) + return {false, false}; + + if (!pos_descr.fill_to.isNull() && !less(row[pos], pos_descr.fill_to, getDirection(pos))) + return {false, false}; + + /// If we have any 'fill_to' value at position greater than 'pos' or configured staleness, + /// we need to generate rows up to one of this borders. for (size_t i = row_size - 1; i > pos; --i) { auto & fill_column_desc = getFillDescription(i); - if (fill_column_desc.fill_to.isNull() || row[i].isNull()) + if (row[i].isNull()) continue; - auto next_value = doJump(fill_column_desc, i); - if (next_value.has_value() && !equals(next_value.value(), fill_column_desc.fill_to)) - { - row[i] = std::move(next_value.value()); - initFromDefaults(i + 1); - return {true, true}; - } + if (fill_column_desc.fill_to.isNull() && staleness_border[i].isNull()) + continue; + + Field next_value = row[i]; + fill_column_desc.step_func(next_value, 1); + + if (!staleness_border[i].isNull() && !less(next_value, staleness_border[i], getDirection(i))) + continue; + + if (!fill_column_desc.fill_to.isNull() && !less(next_value, fill_column_desc.fill_to, getDirection(i))) + continue; + + row[i] = next_value; + initWithFrom(i + 1); + return {true, true}; } - auto & fill_column_desc = getFillDescription(pos); - std::optional next_value; + auto next_value = row[pos]; + getFillDescription(pos).step_func(next_value, 1); - if (long_jump) - { - next_value = doLongJump(fill_column_desc, pos, to_row[pos]); - - if (!next_value.has_value()) - return {false, false}; - - /// We need value >= to_row[pos] - fill_column_desc.step_func(next_value.value(), 1); - } - else - { - next_value = doJump(fill_column_desc, pos); - } - - if (!next_value.has_value() || less(to_row.row[pos], next_value.value(), getDirection(pos)) || equals(next_value.value(), getFillDescription(pos).fill_to)) + if (!next_original_row[pos].isNull() && less(next_original_row[pos], next_value, getDirection(pos))) return {false, false}; - row[pos] = std::move(next_value.value()); - if (equals(row[pos], to_row.row[pos])) + if (!staleness_border[pos].isNull() && !less(next_value, staleness_border[pos], getDirection(pos))) + return {false, false}; + + if (!pos_descr.fill_to.isNull() && !less(next_value, pos_descr.fill_to, getDirection(pos))) + return {false, false}; + + row[pos] = next_value; + if (equals(row[pos], next_original_row[pos])) { bool is_less = false; for (size_t i = pos + 1; i < row_size; ++i) { - const auto & fill_from = getFillDescription(i).fill_from; - if (!fill_from.isNull()) - row[i] = fill_from; + const auto & descr = getFillDescription(i); + if (!descr.fill_from.isNull()) + row[i] = descr.fill_from; else - row[i] = to_row.row[i]; - is_less |= less(row[i], to_row.row[i], getDirection(i)); + row[i] = next_original_row[i]; + + is_less |= ( + (next_original_row[i].isNull() || less(row[i], next_original_row[i], getDirection(i))) && + (staleness_border[i].isNull() || less(row[i], staleness_border[i], getDirection(i))) && + (descr.fill_to.isNull() || less(row[i], descr.fill_to, getDirection(i))) + ); } return {is_less, true}; } - initFromDefaults(pos + 1); + initWithFrom(pos + 1); return {true, true}; } -void FillingRow::initFromDefaults(size_t from_pos) +bool FillingRow::shift(const FillingRow & next_original_row, bool& value_changed) +{ + auto logger = getLogger("FillingRow::shift"); + LOG_DEBUG(logger, "next_original_row: {}, current: {}", next_original_row.dump(), dump()); + + for (size_t pos = 0; pos < size(); ++pos) + { + if (row[pos].isNull() || next_original_row[pos].isNull() || equals(row[pos], next_original_row[pos])) + continue; + + if (less(next_original_row[pos], row[pos], getDirection(pos))) + return false; + + std::optional next_value = doLongJump(getFillDescription(pos), pos, next_original_row[pos]); + + if (!next_value.has_value()) + { + LOG_DEBUG(logger, "next value: {}", "None"); + continue; + } + else + { + LOG_DEBUG(logger, "next value: {}", next_value->dump()); + } + + row[pos] = std::move(next_value.value()); + + if (equals(row[pos], next_original_row[pos])) + { + bool is_less = false; + for (size_t i = pos + 1; i < size(); ++i) + { + const auto & descr = getFillDescription(i); + if (!descr.fill_from.isNull()) + row[i] = descr.fill_from; + else + row[i] = next_original_row[i]; + + is_less |= ( + (next_original_row[i].isNull() || less(row[i], next_original_row[i], getDirection(i))) && + (staleness_border[i].isNull() || less(row[i], staleness_border[i], getDirection(i))) && + (descr.fill_to.isNull() || less(row[i], descr.fill_to, getDirection(i))) + ); + } + + LOG_DEBUG(logger, "is less: {}", is_less); + + value_changed = true; + return is_less; + } + else + { + // getFillDescription(pos).step_func(row[pos], 1); + initWithTo(/*from_pos=*/pos + 1); + + value_changed = false; + return false; + } + } + + return false; +} + +bool FillingRow::isConstraintComplete(size_t pos) const +{ + auto logger = getLogger("FillingRow::isConstraintComplete"); + + if (row[pos].isNull()) + { + LOG_DEBUG(logger, "disabled"); + return true; /// disabled + } + + const auto & descr = getFillDescription(pos); + int direction = getDirection(pos); + + if (!descr.fill_to.isNull() && !less(row[pos], descr.fill_to, direction)) + { + LOG_DEBUG(logger, "fill to: {}, row: {}, direction: {}", descr.fill_to.dump(), row[pos].dump(), direction); + return false; + } + + if (!staleness_border[pos].isNull() && !less(row[pos], staleness_border[pos], direction)) + { + LOG_DEBUG(logger, "staleness border: {}, row: {}, direction: {}", staleness_border[pos].dump(), row[pos].dump(), direction); + return false; + } + + return true; +} + +bool FillingRow::isConstraintsComplete() const +{ + for (size_t pos = 0; pos < size(); ++pos) + { + if (isConstraintComplete(pos)) + return true; + } + + return false; +} + +bool FillingRow::isLessStaleness() const +{ + auto logger = getLogger("FillingRow::isLessStaleness"); + + for (size_t pos = 0; pos < size(); ++pos) + { + LOG_DEBUG(logger, "staleness border: {}, row: {}", staleness_border[pos].dump(), row[pos].dump()); + + if (row[pos].isNull() || staleness_border[pos].isNull()) + continue; + + if (less(row[pos], staleness_border[pos], getDirection(pos))) + return true; + } + + return false; +} + +bool FillingRow::isStalenessConfigured() const +{ + for (size_t pos = 0; pos < size(); ++pos) + if (!getFillDescription(pos).fill_staleness.isNull()) + return true; + + return false; +} + +bool FillingRow::isLessFillTo() const +{ + auto logger = getLogger("FillingRow::isLessFillTo"); + + for (size_t pos = 0; pos < size(); ++pos) + { + const auto & descr = getFillDescription(pos); + + LOG_DEBUG(logger, "fill to: {}, row: {}", descr.fill_to.dump(), row[pos].dump()); + + if (row[pos].isNull() || descr.fill_to.isNull()) + continue; + + if (less(row[pos], descr.fill_to, getDirection(pos))) + return true; + } + + return false; +} + +bool FillingRow::isFillToConfigured() const +{ + for (size_t pos = 0; pos < size(); ++pos) + if (!getFillDescription(pos).fill_to.isNull()) + return true; + + return false; +} + + +void FillingRow::initWithFrom(size_t from_pos) { for (size_t i = from_pos; i < sort_description.size(); ++i) row[i] = getFillDescription(i).fill_from; } +void FillingRow::initWithTo(size_t from_pos) +{ + for (size_t i = from_pos; i < sort_description.size(); ++i) + row[i] = getFillDescription(i).fill_to; +} + void FillingRow::initStalenessRow(const Columns& base_row, size_t row_ind) { for (size_t i = 0; i < size(); ++i) { - staleness_border[i] = (*base_row[i])[row_ind]; - const auto& descr = getFillDescription(i); if (!descr.fill_staleness.isNull()) + { + staleness_border[i] = (*base_row[i])[row_ind]; descr.staleness_step_func(staleness_border[i], 1); + } } } diff --git a/src/Interpreters/FillingRow.h b/src/Interpreters/FillingRow.h index dc787173191..a5e622e4c6e 100644 --- a/src/Interpreters/FillingRow.h +++ b/src/Interpreters/FillingRow.h @@ -25,9 +25,22 @@ public: /// Return pair of boolean /// apply - true if filling values should be inserted into result set /// value_changed - true if filling row value was changed - std::pair next(const FillingRow & to_row, bool long_jump); + std::pair next(const FillingRow & next_original_row); - void initFromDefaults(size_t from_pos = 0); + /// Returns true if need to generate some prefix for to_row + bool shift(const FillingRow & next_original_row, bool& value_changed); + + bool isConstraintComplete(size_t pos) const; + bool isConstraintsComplete() const; + + bool isLessStaleness() const; + bool isStalenessConfigured() const; + + bool isLessFillTo() const; + bool isFillToConfigured() const; + + void initWithFrom(size_t from_pos = 0); + void initWithTo(size_t from_pos = 0); void initStalenessRow(const Columns& base_row, size_t row_ind); Field & operator[](size_t index) { return row[index]; } @@ -39,6 +52,7 @@ public: bool isNull() const; int getDirection(size_t index) const { return sort_description[index].direction; } + Field getStalenessBorder(size_t index) const { return staleness_border[index]; } FillColumnDescription & getFillDescription(size_t index) { return sort_description[index].fill_description; } const FillColumnDescription & getFillDescription(size_t index) const { return sort_description[index].fill_description; } diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index 46a670394a5..a3a185929dc 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -11,13 +11,14 @@ #include #include #include +#include "Interpreters/FillingRow.h" #include namespace DB { -constexpr bool debug_logging_enabled = false; +constexpr bool debug_logging_enabled = true; template void logDebug(String key, const T & value, const char * separator = " : ") @@ -507,18 +508,39 @@ bool FillingTransform::generateSuffixIfNeeded( logDebug("should_insert_first", should_insert_first); for (size_t i = 0, size = filling_row.size(); i < size; ++i) - next_row[i] = filling_row.getFillDescription(i).fill_to; + next_row[i] = Field{}; logDebug("generateSuffixIfNeeded next_row updated", next_row); - if (filling_row >= next_row) + // if (!filling_row.isFillToConfigured() && !filling_row.isStalenessConfigured()) + // { + // logDebug("generateSuffixIfNeeded", "no other constraints, will not generate suffix"); + // return false; + // } + + // logDebug("filling_row.isLessFillTo()", filling_row.isLessFillTo()); + // logDebug("filling_row.isLessStaleness()", filling_row.isLessStaleness()); + + // if (filling_row.isFillToConfigured() && !filling_row.isLessFillTo()) + // { + // logDebug("generateSuffixIfNeeded", "not less than fill to, will not generate suffix"); + // return false; + // } + + // if (filling_row.isStalenessConfigured() && !filling_row.isLessStaleness()) + // { + // logDebug("generateSuffixIfNeeded", "not less than staleness border, will not generate suffix"); + // return false; + // } + + if (!filling_row.isConstraintsComplete()) { - logDebug("generateSuffixIfNeeded", "no need to generate suffix"); + logDebug("generateSuffixIfNeeded", "will not generate suffix"); return false; } Block interpolate_block; - if (should_insert_first && filling_row < next_row) + if (should_insert_first) { interpolate(result_columns, interpolate_block); insertFromFillingRow(res_fill_columns, res_interpolate_columns, res_other_columns, interpolate_block); @@ -533,7 +555,7 @@ bool FillingTransform::generateSuffixIfNeeded( bool filling_row_changed = false; while (true) { - const auto [apply, changed] = filling_row.next(next_row, /*long_jump=*/false); + const auto [apply, changed] = filling_row.next(next_row); filling_row_changed = changed; if (!apply) break; @@ -615,7 +637,7 @@ void FillingTransform::transformRange( if (!fill_from.isNull() && !equals(current_value, fill_from)) { - filling_row.initFromDefaults(i); + filling_row.initWithFrom(i); filling_row_inserted = false; if (less(fill_from, current_value, filling_row.getDirection(i))) { @@ -642,24 +664,14 @@ void FillingTransform::transformRange( logDebug("should_insert_first", should_insert_first); for (size_t i = 0, size = filling_row.size(); i < size; ++i) - { - const auto current_value = (*input_fill_columns[i])[row_ind]; - const auto & fill_to = filling_row.getFillDescription(i).fill_to; + next_row[i] = (*input_fill_columns[i])[row_ind]; - logDebug("current value", current_value.dump()); - logDebug("fill to", fill_to.dump()); - - if (fill_to.isNull() || less(current_value, fill_to, filling_row.getDirection(i))) - next_row[i] = current_value; - else - next_row[i] = fill_to; - } logDebug("next_row updated", next_row); /// The condition is true when filling row is initialized by value(s) in FILL FROM, /// and there are row(s) in current range with value(s) < then in the filling row. /// It can happen only once for a range. - if (should_insert_first && filling_row < next_row) + if (should_insert_first && filling_row < next_row && filling_row.isConstraintsComplete()) { interpolate(result_columns, interpolate_block); insertFromFillingRow(res_fill_columns, res_interpolate_columns, res_other_columns, interpolate_block); @@ -669,7 +681,7 @@ void FillingTransform::transformRange( bool filling_row_changed = false; while (true) { - const auto [apply, changed] = filling_row.next(next_row, /*long_jump=*/false); + const auto [apply, changed] = filling_row.next(next_row); filling_row_changed = changed; if (!apply) break; @@ -679,12 +691,36 @@ void FillingTransform::transformRange( copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); } - const auto [apply, changed] = filling_row.next(next_row, /*long_jump=*/true); - logDebug("long jump apply", apply); - logDebug("long jump changed", changed); + { + filling_row.initStalenessRow(input_fill_columns, row_ind); - if (changed) - filling_row_changed = true; + bool shift_apply = filling_row.shift(next_row, filling_row_changed); + logDebug("shift_apply", shift_apply); + logDebug("filling_row_changed", filling_row_changed); + + while (shift_apply) + { + logDebug("after shift", filling_row); + + while (true) + { + logDebug("filling_row in prefix", filling_row); + + interpolate(result_columns, interpolate_block); + insertFromFillingRow(res_fill_columns, res_interpolate_columns, res_other_columns, interpolate_block); + copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); + + const auto [apply, changed] = filling_row.next(next_row); + logDebug("filling_row in prefix", filling_row); + + filling_row_changed = changed; + if (!apply) + break; + } + + shift_apply = filling_row.shift(next_row, filling_row_changed); + } + } /// new valid filling row was generated but not inserted, will use it during suffix generation if (filling_row_changed) @@ -697,8 +733,8 @@ void FillingTransform::transformRange( copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); copyRowFromColumns(res_other_columns, input_other_columns, row_ind); - /// Init next staleness interval with current row, because we have already made the long jump to it - filling_row.initStalenessRow(input_fill_columns, row_ind); + // /// Init next staleness interval with current row, because we have already made the long jump to it + // filling_row.initStalenessRow(input_fill_columns, row_ind); } /// save sort prefix of last row in the range, it's used to generate suffix @@ -744,7 +780,7 @@ void FillingTransform::transform(Chunk & chunk) /// if no data was processed, then need to initialize filling_row if (last_row.empty()) { - filling_row.initFromDefaults(); + filling_row.initWithFrom(); filling_row_inserted = false; } From 98f358baa3cac9813ed071067686af56653792c5 Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Wed, 30 Oct 2024 13:42:27 +0100 Subject: [PATCH 333/680] add test --- ...eplicas_join_algo_and_analyzer_4.reference | 29 ++++++ ...allel_replicas_join_algo_and_analyzer_4.sh | 93 +++++++++++++++++++ 2 files changed, 122 insertions(+) create mode 100644 tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference create mode 100755 tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh diff --git a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference new file mode 100644 index 00000000000..9fc156b5fb0 --- /dev/null +++ b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference @@ -0,0 +1,29 @@ +4999950000 +4999950000 +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t` AS `__table1` GROUP BY `__table1`.`item_id` +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` +4999950000 +4999950000 +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t` AS `__table1` +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` GROUP BY `__table1`.`item_id` +499950000 +499960000 +499970000 +499980000 +499990000 +500000000 +500010000 +500020000 +500030000 +500040000 +499950000 +499960000 +499970000 +499980000 +499990000 +500000000 +500010000 +500020000 +500030000 +500040000 +SELECT sum(`__table1`.`item_id`) AS `sum(item_id)` FROM (SELECT `__table2`.`item_id` AS `item_id`, `__table2`.`price_sold` AS `price_sold` FROM `default`.`t` AS `__table2`) AS `__table1` ALL LEFT JOIN (SELECT `__table4`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table4`) AS `__table3` ON `__table1`.`item_id` = `__table3`.`item_id` GROUP BY `__table1`.`price_sold` ORDER BY `__table1`.`price_sold` ASC diff --git a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh new file mode 100755 index 00000000000..a588fa47c2d --- /dev/null +++ b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh @@ -0,0 +1,93 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + + +${CLICKHOUSE_CLIENT} --query=" +CREATE TABLE t +( + item_id UInt64, + price_sold Float32, + date Date +) +ENGINE = MergeTree +ORDER BY item_id; + +CREATE TABLE t1 +( + item_id UInt64, + price_sold Float32, + date Date +) +ENGINE = MergeTree +ORDER BY item_id; + +INSERT INTO t SELECT number, number % 10, toDate(number) FROM numbers(100000); +INSERT INTO t1 SELECT number, number % 10, toDate(number) FROM numbers(100000); +" + +query1=" + SELECT sum(item_id) + FROM + ( + SELECT item_id + FROM t + GROUP BY item_id + ) AS l + LEFT JOIN + ( + SELECT item_id + FROM t1 + ) AS r ON l.item_id = r.item_id +" + +query2=" + SELECT sum(item_id) + FROM + ( + SELECT item_id + FROM t + ) AS l + LEFT JOIN + ( + SELECT item_id + FROM t1 + GROUP BY item_id + ) AS r ON l.item_id = r.item_id +" + +query3=" + SELECT sum(item_id) + FROM + ( + SELECT item_id, price_sold + FROM t + ) AS l + LEFT JOIN + ( + SELECT item_id + FROM t1 + ) AS r ON l.item_id = r.item_id + GROUP BY price_sold + ORDER BY price_sold +" + +for query in "${query1}" "${query2}" "${query3}"; do + for enable_parallel_replicas in {0..1}; do + ${CLICKHOUSE_CLIENT} --query=" + set enable_analyzer=1; + set allow_experimental_parallel_reading_from_replicas=${enable_parallel_replicas}, cluster_for_parallel_replicas='parallel_replicas', max_parallel_replicas=100, parallel_replicas_for_non_replicated_merge_tree=1; + + ${query}; + + SELECT replaceRegexpAll(explain, '.*Query: (.*) Replicas:.*', '\\1') + FROM + ( + EXPLAIN actions=1 ${query} + ) + WHERE explain LIKE '%ParallelReplicas%'; + " + done +done From e76f66d865540f86e32ac415974cfcd9b35c6b65 Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Wed, 30 Oct 2024 13:58:33 +0100 Subject: [PATCH 334/680] fix typo --- 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 8a806045111..fce86a6cda0 100644 --- a/src/Planner/findParallelReplicasQuery.cpp +++ b/src/Planner/findParallelReplicasQuery.cpp @@ -174,7 +174,7 @@ const QueryNode * findQueryForParallelReplicas( struct Frame { const QueryPlan::Node * node = nullptr; - /// Below we will check subqueries from `stack` to find outtermost subquery that could be executed remotely. + /// Below we will check subqueries from `stack` to find outermost subquery that could be executed remotely. /// Currently traversal algorithm considers only steps with 0 or 1 children and JOIN specifically. /// When we found some step that requires finalization on the initiator (e.g. GROUP BY) there are two options: /// 1. If plan looks like a single path (e.g. AggregatingStep -> ExpressionStep -> Reading) we can execute From 0840f7854c9ff286623d2165b79cec72254cdc67 Mon Sep 17 00:00:00 2001 From: divanik Date: Wed, 30 Oct 2024 13:40:27 +0000 Subject: [PATCH 335/680] Fix ifdefs in ObjectStorageObject table --- src/TableFunctions/TableFunctionObjectStorage.cpp | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/src/TableFunctions/TableFunctionObjectStorage.cpp b/src/TableFunctions/TableFunctionObjectStorage.cpp index 6d81269f2d7..12de08afad0 100644 --- a/src/TableFunctions/TableFunctionObjectStorage.cpp +++ b/src/TableFunctions/TableFunctionObjectStorage.cpp @@ -269,41 +269,43 @@ void registerTableFunctionIceberg(TableFunctionFactory & factory) } #endif + +#if USE_AWS_S3 #if USE_PARQUET void registerTableFunctionDeltaLake(TableFunctionFactory & factory) { -#if USE_AWS_S3 factory.registerFunction( {.documentation = {.description = R"(The table function can be used to read the DeltaLake table stored on object store.)", .examples{{"deltaLake", "SELECT * FROM deltaLake(url, access_key_id, secret_access_key)", ""}}, .categories{"DataLake"}}, .allow_readonly = false}); -#endif } #endif void registerTableFunctionHudi(TableFunctionFactory & factory) { -#if USE_AWS_S3 factory.registerFunction( {.documentation = {.description = R"(The table function can be used to read the Hudi table stored on object store.)", .examples{{"hudi", "SELECT * FROM hudi(url, access_key_id, secret_access_key)", ""}}, .categories{"DataLake"}}, .allow_readonly = false}); -#endif } +#endif + void registerDataLakeTableFunctions(TableFunctionFactory & factory) { UNUSED(factory); #if USE_AVRO registerTableFunctionIceberg(factory); #endif +#if USE_AWS_S3 #if USE_PARQUET registerTableFunctionDeltaLake(factory); #endif registerTableFunctionHudi(factory); +#endif } } From b9829c703fd4ceae38b5d195ae195c2321e17444 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 13:44:59 +0000 Subject: [PATCH 336/680] change constraints check --- src/Interpreters/FillingRow.cpp | 75 ++++++++++++------- src/Interpreters/FillingRow.h | 6 +- .../Transforms/FillingTransform.cpp | 2 +- 3 files changed, 53 insertions(+), 30 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index caf6ad9e3ba..825b0b1488a 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -3,6 +3,7 @@ #include "Common/Logger.h" #include "Common/logger_useful.h" #include +#include "base/defines.h" #include @@ -122,6 +123,43 @@ std::optional FillingRow::doLongJump(const FillColumnDescription & descr, return shifted_value; } +bool FillingRow::hasSomeConstraints(size_t pos) const +{ + const auto & descr = getFillDescription(pos); + + if (!descr.fill_to.isNull()) + return true; + + if (!descr.fill_staleness.isNull()) + return true; + + return false; +} + +bool FillingRow::isConstraintsComplete(size_t pos) const +{ + auto logger = getLogger("FillingRow::isConstraintComplete"); + chassert(!row[pos].isNull()); + chassert(hasSomeConstraints(pos)); + + const auto & descr = getFillDescription(pos); + int direction = getDirection(pos); + + if (!descr.fill_to.isNull() && !less(row[pos], descr.fill_to, direction)) + { + LOG_DEBUG(logger, "fill to: {}, row: {}, direction: {}", descr.fill_to.dump(), row[pos].dump(), direction); + return false; + } + + if (!descr.fill_staleness.isNull() && !less(row[pos], staleness_border[pos], direction)) + { + LOG_DEBUG(logger, "staleness border: {}, row: {}, direction: {}", staleness_border[pos].dump(), row[pos].dump(), direction); + return false; + } + + return true; +} + Field findMin(Field a, Field b, Field c, int dir) { auto logger = getLogger("FillingRow"); @@ -300,43 +338,26 @@ bool FillingRow::shift(const FillingRow & next_original_row, bool& value_changed return false; } -bool FillingRow::isConstraintComplete(size_t pos) const +bool FillingRow::hasSomeConstraints() const { - auto logger = getLogger("FillingRow::isConstraintComplete"); + for (size_t pos = 0; pos < size(); ++pos) + if (hasSomeConstraints(pos)) + return true; - if (row[pos].isNull()) - { - LOG_DEBUG(logger, "disabled"); - return true; /// disabled - } - - const auto & descr = getFillDescription(pos); - int direction = getDirection(pos); - - if (!descr.fill_to.isNull() && !less(row[pos], descr.fill_to, direction)) - { - LOG_DEBUG(logger, "fill to: {}, row: {}, direction: {}", descr.fill_to.dump(), row[pos].dump(), direction); - return false; - } - - if (!staleness_border[pos].isNull() && !less(row[pos], staleness_border[pos], direction)) - { - LOG_DEBUG(logger, "staleness border: {}, row: {}, direction: {}", staleness_border[pos].dump(), row[pos].dump(), direction); - return false; - } - - return true; + return false; } bool FillingRow::isConstraintsComplete() const { for (size_t pos = 0; pos < size(); ++pos) { - if (isConstraintComplete(pos)) - return true; + if (row[pos].isNull() || !hasSomeConstraints(pos)) + continue; + + return isConstraintsComplete(pos); } - return false; + return true; } bool FillingRow::isLessStaleness() const diff --git a/src/Interpreters/FillingRow.h b/src/Interpreters/FillingRow.h index a5e622e4c6e..bd5a1b877a5 100644 --- a/src/Interpreters/FillingRow.h +++ b/src/Interpreters/FillingRow.h @@ -18,6 +18,9 @@ class FillingRow std::optional doJump(const FillColumnDescription & descr, size_t column_ind); std::optional doLongJump(const FillColumnDescription & descr, size_t column_ind, const Field & to); + bool hasSomeConstraints(size_t pos) const; + bool isConstraintsComplete(size_t pos) const; + public: explicit FillingRow(const SortDescription & sort_description); @@ -30,7 +33,7 @@ public: /// Returns true if need to generate some prefix for to_row bool shift(const FillingRow & next_original_row, bool& value_changed); - bool isConstraintComplete(size_t pos) const; + bool hasSomeConstraints() const; bool isConstraintsComplete() const; bool isLessStaleness() const; @@ -52,7 +55,6 @@ public: bool isNull() const; int getDirection(size_t index) const { return sort_description[index].direction; } - Field getStalenessBorder(size_t index) const { return staleness_border[index]; } FillColumnDescription & getFillDescription(size_t index) { return sort_description[index].fill_description; } const FillColumnDescription & getFillDescription(size_t index) const { return sort_description[index].fill_description; } diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index a3a185929dc..ce804c94d8e 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -533,7 +533,7 @@ bool FillingTransform::generateSuffixIfNeeded( // return false; // } - if (!filling_row.isConstraintsComplete()) + if (!filling_row.hasSomeConstraints() || !filling_row.isConstraintsComplete()) { logDebug("generateSuffixIfNeeded", "will not generate suffix"); return false; From 433523c6f29a55d28930ec86fe268edffc16738e Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 13:49:42 +0000 Subject: [PATCH 337/680] update test --- .../03266_with_fill_staleness.reference | 32 +++++++++++++------ 1 file changed, 22 insertions(+), 10 deletions(-) diff --git a/tests/queries/0_stateless/03266_with_fill_staleness.reference b/tests/queries/0_stateless/03266_with_fill_staleness.reference index 6b090443359..25d7b7c3f24 100644 --- a/tests/queries/0_stateless/03266_with_fill_staleness.reference +++ b/tests/queries/0_stateless/03266_with_fill_staleness.reference @@ -50,6 +50,8 @@ staleness 3 seconds 2016-06-15 23:00:21 20 2016-06-15 23:00:22 20 2016-06-15 23:00:25 25 original +2016-06-15 23:00:26 25 +2016-06-15 23:00:27 25 descending order 2016-06-15 23:00:25 25 original 2016-06-15 23:00:24 25 @@ -62,6 +64,7 @@ descending order 2016-06-15 23:00:05 5 original 2016-06-15 23:00:04 5 2016-06-15 23:00:00 0 original +2016-06-15 22:59:59 0 staleness with to and step 2016-06-15 23:00:00 0 original 2016-06-15 23:00:03 0 @@ -86,33 +89,41 @@ staleness with another regular with fill 2016-06-15 23:00:01 1970-01-01 01:00:00 0 2016-06-15 23:00:01 1970-01-01 01:00:01 0 2016-06-15 23:00:01 1970-01-01 01:00:02 0 +2016-06-15 23:00:05 1970-01-01 01:00:00 0 +2016-06-15 23:00:05 1970-01-01 01:00:01 0 +2016-06-15 23:00:05 1970-01-01 01:00:02 0 2016-06-15 23:00:05 2016-06-15 23:00:05 5 original -2016-06-15 23:00:05 1970-01-01 01:00:01 5 -2016-06-15 23:00:05 1970-01-01 01:00:02 5 2016-06-15 23:00:06 1970-01-01 01:00:00 5 2016-06-15 23:00:06 1970-01-01 01:00:01 5 2016-06-15 23:00:06 1970-01-01 01:00:02 5 +2016-06-15 23:00:10 1970-01-01 01:00:00 5 +2016-06-15 23:00:10 1970-01-01 01:00:01 5 +2016-06-15 23:00:10 1970-01-01 01:00:02 5 2016-06-15 23:00:10 2016-06-15 23:00:10 10 original -2016-06-15 23:00:10 1970-01-01 01:00:01 10 -2016-06-15 23:00:10 1970-01-01 01:00:02 10 2016-06-15 23:00:11 1970-01-01 01:00:00 10 2016-06-15 23:00:11 1970-01-01 01:00:01 10 2016-06-15 23:00:11 1970-01-01 01:00:02 10 +2016-06-15 23:00:15 1970-01-01 01:00:00 10 +2016-06-15 23:00:15 1970-01-01 01:00:01 10 +2016-06-15 23:00:15 1970-01-01 01:00:02 10 2016-06-15 23:00:15 2016-06-15 23:00:15 15 original -2016-06-15 23:00:15 1970-01-01 01:00:01 15 -2016-06-15 23:00:15 1970-01-01 01:00:02 15 2016-06-15 23:00:16 1970-01-01 01:00:00 15 2016-06-15 23:00:16 1970-01-01 01:00:01 15 2016-06-15 23:00:16 1970-01-01 01:00:02 15 +2016-06-15 23:00:20 1970-01-01 01:00:00 15 +2016-06-15 23:00:20 1970-01-01 01:00:01 15 +2016-06-15 23:00:20 1970-01-01 01:00:02 15 2016-06-15 23:00:20 2016-06-15 23:00:20 20 original -2016-06-15 23:00:20 1970-01-01 01:00:01 20 -2016-06-15 23:00:20 1970-01-01 01:00:02 20 2016-06-15 23:00:21 1970-01-01 01:00:00 20 2016-06-15 23:00:21 1970-01-01 01:00:01 20 2016-06-15 23:00:21 1970-01-01 01:00:02 20 +2016-06-15 23:00:25 1970-01-01 01:00:00 20 +2016-06-15 23:00:25 1970-01-01 01:00:01 20 +2016-06-15 23:00:25 1970-01-01 01:00:02 20 2016-06-15 23:00:25 2016-06-15 23:00:25 25 original -2016-06-15 23:00:25 1970-01-01 01:00:01 25 -2016-06-15 23:00:25 1970-01-01 01:00:02 25 +2016-06-15 23:00:26 1970-01-01 01:00:00 25 +2016-06-15 23:00:26 1970-01-01 01:00:01 25 +2016-06-15 23:00:26 1970-01-01 01:00:02 25 double staleness 2016-06-15 23:00:00 2016-06-15 23:00:00 0 original 2016-06-15 23:00:00 2016-06-15 23:00:02 0 @@ -137,3 +148,4 @@ double staleness 2016-06-15 23:00:25 2016-06-15 23:00:25 25 original 2016-06-15 23:00:25 2016-06-15 23:00:27 25 2016-06-15 23:00:25 2016-06-15 23:00:29 25 +2016-06-15 23:00:26 1970-01-01 01:00:00 25 From 60840cb05fc2f948745e92d00b0e15cbd7a8923a Mon Sep 17 00:00:00 2001 From: kssenii Date: Wed, 30 Oct 2024 14:55:15 +0100 Subject: [PATCH 338/680] Fix memory usage in remote read when enable_filesystem_cache=1, but cached disk absent --- src/Disks/IO/AsynchronousBoundedReadBuffer.cpp | 6 ++++-- src/Disks/IO/AsynchronousBoundedReadBuffer.h | 2 ++ src/Disks/IO/CachedOnDiskReadBufferFromFile.h | 2 ++ src/Disks/IO/ReadBufferFromRemoteFSGather.cpp | 16 +++------------- src/Disks/IO/ReadBufferFromRemoteFSGather.h | 5 ++--- src/Disks/ObjectStorages/DiskObjectStorage.cpp | 18 ++++++++++++++++-- src/IO/ReadBufferFromFileBase.h | 2 ++ .../StorageObjectStorageSource.cpp | 10 +++++++++- 8 files changed, 40 insertions(+), 21 deletions(-) diff --git a/src/Disks/IO/AsynchronousBoundedReadBuffer.cpp b/src/Disks/IO/AsynchronousBoundedReadBuffer.cpp index b24b95af85c..77b03cdd1f7 100644 --- a/src/Disks/IO/AsynchronousBoundedReadBuffer.cpp +++ b/src/Disks/IO/AsynchronousBoundedReadBuffer.cpp @@ -46,11 +46,13 @@ AsynchronousBoundedReadBuffer::AsynchronousBoundedReadBuffer( ImplPtr impl_, IAsynchronousReader & reader_, const ReadSettings & settings_, + size_t buffer_size_, AsyncReadCountersPtr async_read_counters_, FilesystemReadPrefetchesLogPtr prefetches_log_) : ReadBufferFromFileBase(0, nullptr, 0) , impl(std::move(impl_)) , read_settings(settings_) + , buffer_size(buffer_size_) , reader(reader_) , query_id(CurrentThread::isInitialized() && CurrentThread::get().getQueryContext() != nullptr ? CurrentThread::getQueryId() : "") , current_reader_id(getRandomASCIIString(8)) @@ -112,7 +114,7 @@ void AsynchronousBoundedReadBuffer::prefetch(Priority priority) last_prefetch_info.submit_time = std::chrono::system_clock::now(); last_prefetch_info.priority = priority; - prefetch_buffer.resize(chooseBufferSizeForRemoteReading(read_settings, impl->getFileSize())); + prefetch_buffer.resize(buffer_size); prefetch_future = readAsync(prefetch_buffer.data(), prefetch_buffer.size(), priority); ProfileEvents::increment(ProfileEvents::RemoteFSPrefetches); } @@ -211,7 +213,7 @@ bool AsynchronousBoundedReadBuffer::nextImpl() } else { - memory.resize(chooseBufferSizeForRemoteReading(read_settings, impl->getFileSize())); + memory.resize(buffer_size); { ProfileEventTimeIncrement watch(ProfileEvents::SynchronousRemoteReadWaitMicroseconds); diff --git a/src/Disks/IO/AsynchronousBoundedReadBuffer.h b/src/Disks/IO/AsynchronousBoundedReadBuffer.h index 3dc8fcc39cb..7664cc4d386 100644 --- a/src/Disks/IO/AsynchronousBoundedReadBuffer.h +++ b/src/Disks/IO/AsynchronousBoundedReadBuffer.h @@ -27,6 +27,7 @@ public: ImplPtr impl_, IAsynchronousReader & reader_, const ReadSettings & settings_, + size_t buffer_size_, AsyncReadCountersPtr async_read_counters_ = nullptr, FilesystemReadPrefetchesLogPtr prefetches_log_ = nullptr); @@ -53,6 +54,7 @@ public: private: const ImplPtr impl; const ReadSettings read_settings; + const size_t buffer_size; IAsynchronousReader & reader; size_t file_offset_of_buffer_end = 0; diff --git a/src/Disks/IO/CachedOnDiskReadBufferFromFile.h b/src/Disks/IO/CachedOnDiskReadBufferFromFile.h index 119fa166214..4881b6a309d 100644 --- a/src/Disks/IO/CachedOnDiskReadBufferFromFile.h +++ b/src/Disks/IO/CachedOnDiskReadBufferFromFile.h @@ -41,6 +41,8 @@ public: ~CachedOnDiskReadBufferFromFile() override; + bool isCached() const override { return true; } + bool nextImpl() override; off_t seek(off_t off, int whence) override; diff --git a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp index 8e4ec6f3dfb..8d3b9366261 100644 --- a/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp +++ b/src/Disks/IO/ReadBufferFromRemoteFSGather.cpp @@ -18,24 +18,14 @@ namespace ErrorCodes extern const int CANNOT_SEEK_THROUGH_FILE; } -size_t chooseBufferSizeForRemoteReading(const DB::ReadSettings & settings, size_t file_size) -{ - /// Only when cache is used we could download bigger portions of FileSegments than what we actually gonna read within particular task. - if (!settings.enable_filesystem_cache && !settings.read_through_distributed_cache) - return settings.remote_fs_buffer_size; - - /// Buffers used for prefetch and pre-download better to have enough size, but not bigger than the whole file. - return std::min(std::max(settings.remote_fs_buffer_size, DBMS_DEFAULT_BUFFER_SIZE), file_size); -} - ReadBufferFromRemoteFSGather::ReadBufferFromRemoteFSGather( ReadBufferCreator && read_buffer_creator_, const StoredObjects & blobs_to_read_, const ReadSettings & settings_, std::shared_ptr cache_log_, - bool use_external_buffer_) - : ReadBufferFromFileBase(use_external_buffer_ ? 0 : chooseBufferSizeForRemoteReading( - settings_, getTotalSize(blobs_to_read_)), nullptr, 0) + bool use_external_buffer_, + size_t buffer_size) + : ReadBufferFromFileBase(use_external_buffer_ ? 0 : buffer_size, nullptr, 0) , settings(settings_) , blobs_to_read(blobs_to_read_) , read_buffer_creator(std::move(read_buffer_creator_)) diff --git a/src/Disks/IO/ReadBufferFromRemoteFSGather.h b/src/Disks/IO/ReadBufferFromRemoteFSGather.h index 27f94a3e552..c5f1966dc38 100644 --- a/src/Disks/IO/ReadBufferFromRemoteFSGather.h +++ b/src/Disks/IO/ReadBufferFromRemoteFSGather.h @@ -28,7 +28,8 @@ public: const StoredObjects & blobs_to_read_, const ReadSettings & settings_, std::shared_ptr cache_log_, - bool use_external_buffer_); + bool use_external_buffer_, + size_t buffer_size); ~ReadBufferFromRemoteFSGather() override; @@ -84,6 +85,4 @@ private: LoggerPtr log; }; - -size_t chooseBufferSizeForRemoteReading(const DB::ReadSettings & settings, size_t file_size); } diff --git a/src/Disks/ObjectStorages/DiskObjectStorage.cpp b/src/Disks/ObjectStorages/DiskObjectStorage.cpp index fbab25490c1..bd7ffeb5a00 100644 --- a/src/Disks/ObjectStorages/DiskObjectStorage.cpp +++ b/src/Disks/ObjectStorages/DiskObjectStorage.cpp @@ -532,19 +532,33 @@ std::unique_ptr DiskObjectStorage::readFile( return impl; }; + /// Avoid cache fragmentation by choosing bigger buffer size. + bool prefer_bigger_buffer_size = object_storage->supportsCache() && read_settings.enable_filesystem_cache; + size_t buffer_size = prefer_bigger_buffer_size + ? std::max(settings.remote_fs_buffer_size, DBMS_DEFAULT_BUFFER_SIZE) + : settings.remote_fs_buffer_size; + + size_t total_objects_size = getTotalSize(storage_objects); + if (total_objects_size) + buffer_size = std::min(buffer_size, total_objects_size); + const bool use_async_buffer = read_settings.remote_fs_method == RemoteFSReadMethod::threadpool; auto impl = std::make_unique( std::move(read_buffer_creator), storage_objects, read_settings, global_context->getFilesystemCacheLog(), - /* use_external_buffer */use_async_buffer); + /* use_external_buffer */use_async_buffer, + /* buffer_size */use_async_buffer ? 0 : buffer_size); if (use_async_buffer) { auto & reader = global_context->getThreadPoolReader(FilesystemReaderType::ASYNCHRONOUS_REMOTE_FS_READER); return std::make_unique( - std::move(impl), reader, read_settings, + std::move(impl), + reader, + read_settings, + buffer_size, global_context->getAsyncReadCounters(), global_context->getFilesystemReadPrefetchesLog()); diff --git a/src/IO/ReadBufferFromFileBase.h b/src/IO/ReadBufferFromFileBase.h index c98dcd5a93e..c59a5c152b6 100644 --- a/src/IO/ReadBufferFromFileBase.h +++ b/src/IO/ReadBufferFromFileBase.h @@ -60,6 +60,8 @@ public: /// file offset and what getPosition() returns. virtual bool isRegularLocalFile(size_t * /*out_view_offsee*/) { return false; } + virtual bool isCached() const { return false; } + protected: std::optional file_size; ProfileCallback profile_callback; diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp index 52b0f00f71a..90871b8c0ad 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp @@ -517,9 +517,17 @@ std::unique_ptr StorageObjectStorageSource::createReadBu LOG_TRACE(log, "Downloading object of size {} with initial prefetch", object_size); + bool prefer_bigger_buffer_size = impl->isCached(); + size_t buffer_size = prefer_bigger_buffer_size + ? std::max(read_settings.remote_fs_buffer_size, DBMS_DEFAULT_BUFFER_SIZE) + : read_settings.remote_fs_buffer_size; + auto & reader = context_->getThreadPoolReader(FilesystemReaderType::ASYNCHRONOUS_REMOTE_FS_READER); impl = std::make_unique( - std::move(impl), reader, modified_read_settings, + std::move(impl), + reader, + modified_read_settings, + buffer_size, context_->getAsyncReadCounters(), context_->getFilesystemReadPrefetchesLog()); From 81c58d9406a0194f83bc800f5f7c0cc502f13b10 Mon Sep 17 00:00:00 2001 From: kssenii Date: Wed, 30 Oct 2024 15:15:41 +0100 Subject: [PATCH 339/680] Better check --- src/Disks/ObjectStorages/DiskObjectStorage.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Disks/ObjectStorages/DiskObjectStorage.cpp b/src/Disks/ObjectStorages/DiskObjectStorage.cpp index bd7ffeb5a00..d677623ab57 100644 --- a/src/Disks/ObjectStorages/DiskObjectStorage.cpp +++ b/src/Disks/ObjectStorages/DiskObjectStorage.cpp @@ -538,7 +538,7 @@ std::unique_ptr DiskObjectStorage::readFile( ? std::max(settings.remote_fs_buffer_size, DBMS_DEFAULT_BUFFER_SIZE) : settings.remote_fs_buffer_size; - size_t total_objects_size = getTotalSize(storage_objects); + size_t total_objects_size = file_size ? *file_size : getTotalSize(storage_objects); if (total_objects_size) buffer_size = std::min(buffer_size, total_objects_size); From e5fe7a0f52625d3460ca04a21982a1af24e0adcd Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 14:35:29 +0000 Subject: [PATCH 340/680] add more tests --- .../0_stateless/03266_with_fill_staleness.sql | 1 + .../03266_with_fill_staleness_cases.reference | 67 +++++++++++++++++++ .../03266_with_fill_staleness_cases.sql | 25 +++++++ 3 files changed, 93 insertions(+) create mode 100644 tests/queries/0_stateless/03266_with_fill_staleness_cases.reference create mode 100644 tests/queries/0_stateless/03266_with_fill_staleness_cases.sql diff --git a/tests/queries/0_stateless/03266_with_fill_staleness.sql b/tests/queries/0_stateless/03266_with_fill_staleness.sql index fff702ffd83..de47d8287ad 100644 --- a/tests/queries/0_stateless/03266_with_fill_staleness.sql +++ b/tests/queries/0_stateless/03266_with_fill_staleness.sql @@ -1,4 +1,5 @@ SET session_timezone='Europe/Amsterdam'; +SET enable_analyzer=1; DROP TABLE IF EXISTS with_fill_staleness; CREATE TABLE with_fill_staleness (a DateTime, b DateTime, c UInt64) ENGINE = MergeTree ORDER BY a; diff --git a/tests/queries/0_stateless/03266_with_fill_staleness_cases.reference b/tests/queries/0_stateless/03266_with_fill_staleness_cases.reference new file mode 100644 index 00000000000..bf8e5bbe331 --- /dev/null +++ b/tests/queries/0_stateless/03266_with_fill_staleness_cases.reference @@ -0,0 +1,67 @@ +test-1 +0 5 10 original +0 5 13 +0 5 16 +0 5 19 +0 5 22 +0 7 0 +7 8 15 original +7 8 18 +7 8 21 +7 8 24 +7 10 0 +14 10 20 original +14 10 23 +14 12 0 +test-2-1 +1 0 original +1 1 +1 2 +1 3 +1 4 original +1 5 +1 6 +1 7 +1 8 original +1 9 +1 10 +1 11 +1 12 original +test-2-2 +1 0 original +1 1 +1 2 +1 3 +1 4 original +1 5 +1 6 +1 7 +1 8 original +1 9 +1 10 +1 11 +1 12 original +1 13 +1 14 +2 0 +3 0 +4 0 +test-3-1 +25 -10 +25 -8 +25 -6 +25 -4 +25 -2 +25 0 +25 2 +25 4 +25 6 +25 8 +25 10 +25 12 +25 14 +25 16 +25 17 original +28 -10 +30 18 original +31 -10 diff --git a/tests/queries/0_stateless/03266_with_fill_staleness_cases.sql b/tests/queries/0_stateless/03266_with_fill_staleness_cases.sql new file mode 100644 index 00000000000..9e28041c9a1 --- /dev/null +++ b/tests/queries/0_stateless/03266_with_fill_staleness_cases.sql @@ -0,0 +1,25 @@ +SET enable_analyzer=1; + +DROP TABLE IF EXISTS test; +CREATE TABLE test (a Int64, b Int64, c Int64) Engine=MergeTree ORDER BY a; +INSERT INTO test(a, b, c) VALUES (0, 5, 10), (7, 8, 15), (14, 10, 20); + +SELECT 'test-1'; +SELECT *, 'original' AS orig FROM test ORDER BY a, b WITH FILL TO 20 STEP 2 STALENESS 3, c WITH FILL TO 25 step 3; + +DROP TABLE IF EXISTS test2; +CREATE TABLE test2 (a Int64, b Int64) Engine=MergeTree ORDER BY a; +INSERT INTO test2(a, b) values (1, 0), (1, 4), (1, 8), (1, 12); + +SELECT 'test-2-1'; +SELECT *, 'original' AS orig FROM test2 ORDER BY a, b WITH FILL; + +SELECT 'test-2-2'; +SELECT *, 'original' AS orig FROM test2 ORDER BY a WITH FILL to 20 STALENESS 4, b WITH FILL TO 15 STALENESS 7; + +DROP TABLE IF EXISTS test2; +CREATE TABLE test3 (a Int64, b Int64) Engine=MergeTree ORDER BY a; +INSERT INTO test3(a, b) VALUES (25, 17), (30, 18); + +SELECT 'test-3-1'; +SELECT a, b, 'original' AS orig FROM test3 ORDER BY a WITH FILL TO 33 STEP 3, b WITH FILL FROM -10 STEP 2; From 2cda4dd9012059b6c287df7c615cef8e310b2d8e Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 14:46:56 +0000 Subject: [PATCH 341/680] cleanup --- src/Interpreters/FillingRow.cpp | 97 +------------------ src/Interpreters/FillingRow.h | 12 +-- .../Transforms/FillingTransform.cpp | 30 +----- 3 files changed, 11 insertions(+), 128 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index 825b0b1488a..a87ca418b7b 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -68,25 +68,6 @@ bool FillingRow::isNull() const return true; } -std::optional FillingRow::doJump(const FillColumnDescription& descr, size_t column_ind) -{ - Field next_value = row[column_ind]; - descr.step_func(next_value, 1); - - if (!descr.fill_to.isNull() && less(descr.fill_to, next_value, getDirection(column_ind))) - return std::nullopt; - - if (!descr.fill_staleness.isNull()) - { - if (less(next_value, staleness_border[column_ind], getDirection(column_ind))) - return next_value; - else - return std::nullopt; - } - - return next_value; -} - std::optional FillingRow::doLongJump(const FillColumnDescription & descr, size_t column_ind, const Field & to) { Field shifted_value = row[column_ind]; @@ -99,16 +80,6 @@ std::optional FillingRow::doLongJump(const FillColumnDescription & descr, Field next_value = shifted_value; descr.step_func(next_value, step_len); - // if (less(next_value, to, getDirection(0))) - // { - // shifted_value = std::move(next_value); - // step_len *= 2; - // } - // else - // { - // step_len /= 2; - // } - if (less(to, next_value, getDirection(0))) { step_len /= 2; @@ -233,7 +204,7 @@ std::pair FillingRow::next(const FillingRow & next_original_row) continue; row[i] = next_value; - initWithFrom(i + 1); + initUsingFrom(i + 1); return {true, true}; } @@ -271,7 +242,7 @@ std::pair FillingRow::next(const FillingRow & next_original_row) return {is_less, true}; } - initWithFrom(pos + 1); + initUsingFrom(pos + 1); return {true, true}; } @@ -327,8 +298,7 @@ bool FillingRow::shift(const FillingRow & next_original_row, bool& value_changed } else { - // getFillDescription(pos).step_func(row[pos], 1); - initWithTo(/*from_pos=*/pos + 1); + initUsingTo(/*from_pos=*/pos + 1); value_changed = false; return false; @@ -360,70 +330,13 @@ bool FillingRow::isConstraintsComplete() const return true; } -bool FillingRow::isLessStaleness() const -{ - auto logger = getLogger("FillingRow::isLessStaleness"); - - for (size_t pos = 0; pos < size(); ++pos) - { - LOG_DEBUG(logger, "staleness border: {}, row: {}", staleness_border[pos].dump(), row[pos].dump()); - - if (row[pos].isNull() || staleness_border[pos].isNull()) - continue; - - if (less(row[pos], staleness_border[pos], getDirection(pos))) - return true; - } - - return false; -} - -bool FillingRow::isStalenessConfigured() const -{ - for (size_t pos = 0; pos < size(); ++pos) - if (!getFillDescription(pos).fill_staleness.isNull()) - return true; - - return false; -} - -bool FillingRow::isLessFillTo() const -{ - auto logger = getLogger("FillingRow::isLessFillTo"); - - for (size_t pos = 0; pos < size(); ++pos) - { - const auto & descr = getFillDescription(pos); - - LOG_DEBUG(logger, "fill to: {}, row: {}", descr.fill_to.dump(), row[pos].dump()); - - if (row[pos].isNull() || descr.fill_to.isNull()) - continue; - - if (less(row[pos], descr.fill_to, getDirection(pos))) - return true; - } - - return false; -} - -bool FillingRow::isFillToConfigured() const -{ - for (size_t pos = 0; pos < size(); ++pos) - if (!getFillDescription(pos).fill_to.isNull()) - return true; - - return false; -} - - -void FillingRow::initWithFrom(size_t from_pos) +void FillingRow::initUsingFrom(size_t from_pos) { for (size_t i = from_pos; i < sort_description.size(); ++i) row[i] = getFillDescription(i).fill_from; } -void FillingRow::initWithTo(size_t from_pos) +void FillingRow::initUsingTo(size_t from_pos) { for (size_t i = from_pos; i < sort_description.size(); ++i) row[i] = getFillDescription(i).fill_to; diff --git a/src/Interpreters/FillingRow.h b/src/Interpreters/FillingRow.h index bd5a1b877a5..d33e3f95541 100644 --- a/src/Interpreters/FillingRow.h +++ b/src/Interpreters/FillingRow.h @@ -15,7 +15,7 @@ bool equals(const Field & lhs, const Field & rhs); */ class FillingRow { - std::optional doJump(const FillColumnDescription & descr, size_t column_ind); + /// finds last value <= to std::optional doLongJump(const FillColumnDescription & descr, size_t column_ind, const Field & to); bool hasSomeConstraints(size_t pos) const; @@ -36,14 +36,8 @@ public: bool hasSomeConstraints() const; bool isConstraintsComplete() const; - bool isLessStaleness() const; - bool isStalenessConfigured() const; - - bool isLessFillTo() const; - bool isFillToConfigured() const; - - void initWithFrom(size_t from_pos = 0); - void initWithTo(size_t from_pos = 0); + void initUsingFrom(size_t from_pos = 0); + void initUsingTo(size_t from_pos = 0); void initStalenessRow(const Columns& base_row, size_t row_ind); Field & operator[](size_t index) { return row[index]; } diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index ce804c94d8e..40650b485f8 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -21,7 +21,7 @@ namespace DB constexpr bool debug_logging_enabled = true; template -void logDebug(String key, const T & value, const char * separator = " : ") +static void logDebug(String key, const T & value, const char * separator = " : ") { if constexpr (debug_logging_enabled) { @@ -512,27 +512,6 @@ bool FillingTransform::generateSuffixIfNeeded( logDebug("generateSuffixIfNeeded next_row updated", next_row); - // if (!filling_row.isFillToConfigured() && !filling_row.isStalenessConfigured()) - // { - // logDebug("generateSuffixIfNeeded", "no other constraints, will not generate suffix"); - // return false; - // } - - // logDebug("filling_row.isLessFillTo()", filling_row.isLessFillTo()); - // logDebug("filling_row.isLessStaleness()", filling_row.isLessStaleness()); - - // if (filling_row.isFillToConfigured() && !filling_row.isLessFillTo()) - // { - // logDebug("generateSuffixIfNeeded", "not less than fill to, will not generate suffix"); - // return false; - // } - - // if (filling_row.isStalenessConfigured() && !filling_row.isLessStaleness()) - // { - // logDebug("generateSuffixIfNeeded", "not less than staleness border, will not generate suffix"); - // return false; - // } - if (!filling_row.hasSomeConstraints() || !filling_row.isConstraintsComplete()) { logDebug("generateSuffixIfNeeded", "will not generate suffix"); @@ -637,7 +616,7 @@ void FillingTransform::transformRange( if (!fill_from.isNull() && !equals(current_value, fill_from)) { - filling_row.initWithFrom(i); + filling_row.initUsingFrom(i); filling_row_inserted = false; if (less(fill_from, current_value, filling_row.getDirection(i))) { @@ -732,9 +711,6 @@ void FillingTransform::transformRange( copyRowFromColumns(res_interpolate_columns, input_interpolate_columns, row_ind); copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); copyRowFromColumns(res_other_columns, input_other_columns, row_ind); - - // /// Init next staleness interval with current row, because we have already made the long jump to it - // filling_row.initStalenessRow(input_fill_columns, row_ind); } /// save sort prefix of last row in the range, it's used to generate suffix @@ -780,7 +756,7 @@ void FillingTransform::transform(Chunk & chunk) /// if no data was processed, then need to initialize filling_row if (last_row.empty()) { - filling_row.initWithFrom(); + filling_row.initUsingFrom(); filling_row_inserted = false; } From 3099eae4794a4ec669306e5abc790b01f1fd18bf Mon Sep 17 00:00:00 2001 From: kssenii Date: Wed, 30 Oct 2024 15:56:20 +0100 Subject: [PATCH 342/680] Fix build --- src/Disks/tests/gtest_asynchronous_bounded_read_buffer.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Disks/tests/gtest_asynchronous_bounded_read_buffer.cpp b/src/Disks/tests/gtest_asynchronous_bounded_read_buffer.cpp index 63a39fe39c7..11b4fc3118d 100644 --- a/src/Disks/tests/gtest_asynchronous_bounded_read_buffer.cpp +++ b/src/Disks/tests/gtest_asynchronous_bounded_read_buffer.cpp @@ -51,7 +51,7 @@ TEST_F(AsynchronousBoundedReadBufferTest, setReadUntilPosition) for (bool with_prefetch : {false, true}) { - AsynchronousBoundedReadBuffer read_buffer(createReadBufferFromFileBase(file_path, {}), remote_fs_reader, {}); + AsynchronousBoundedReadBuffer read_buffer(createReadBufferFromFileBase(file_path, {}), remote_fs_reader, {}, DBMS_DEFAULT_BUFFER_SIZE); read_buffer.setReadUntilPosition(20); auto try_read = [&](size_t count) From 7af2e822e7eb486ae95319a09364ea36498bb49b Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 15:22:45 +0000 Subject: [PATCH 343/680] cleanup --- src/Interpreters/FillingRow.cpp | 37 +++++++++------- src/Interpreters/FillingRow.h | 6 +-- .../Transforms/FillingTransform.cpp | 44 ++++++------------- 3 files changed, 36 insertions(+), 51 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index a87ca418b7b..df93ece2af4 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -1,10 +1,10 @@ #include -#include -#include "Common/Logger.h" -#include "Common/logger_useful.h" -#include -#include "base/defines.h" + #include +#include +#include +#include +#include namespace DB @@ -145,7 +145,7 @@ Field findMin(Field a, Field b, Field c, int dir) return a; } -std::pair FillingRow::next(const FillingRow & next_original_row) +bool FillingRow::next(const FillingRow & next_original_row, bool& value_changed) { auto logger = getLogger("FillingRow"); @@ -169,18 +169,18 @@ std::pair FillingRow::next(const FillingRow & next_original_row) LOG_DEBUG(logger, "pos: {}", pos); if (pos == row_size) - return {false, false}; + return false; const auto & pos_descr = getFillDescription(pos); if (!next_original_row[pos].isNull() && less(next_original_row[pos], row[pos], getDirection(pos))) - return {false, false}; + return false; if (!staleness_border[pos].isNull() && !less(row[pos], staleness_border[pos], getDirection(pos))) - return {false, false}; + return false; if (!pos_descr.fill_to.isNull() && !less(row[pos], pos_descr.fill_to, getDirection(pos))) - return {false, false}; + return false; /// If we have any 'fill_to' value at position greater than 'pos' or configured staleness, /// we need to generate rows up to one of this borders. @@ -205,20 +205,22 @@ std::pair FillingRow::next(const FillingRow & next_original_row) row[i] = next_value; initUsingFrom(i + 1); - return {true, true}; + + value_changed = true; + return true; } auto next_value = row[pos]; getFillDescription(pos).step_func(next_value, 1); if (!next_original_row[pos].isNull() && less(next_original_row[pos], next_value, getDirection(pos))) - return {false, false}; + return false; if (!staleness_border[pos].isNull() && !less(next_value, staleness_border[pos], getDirection(pos))) - return {false, false}; + return false; if (!pos_descr.fill_to.isNull() && !less(next_value, pos_descr.fill_to, getDirection(pos))) - return {false, false}; + return false; row[pos] = next_value; if (equals(row[pos], next_original_row[pos])) @@ -239,11 +241,14 @@ std::pair FillingRow::next(const FillingRow & next_original_row) ); } - return {is_less, true}; + value_changed = true; + return is_less; } initUsingFrom(pos + 1); - return {true, true}; + + value_changed = true; + return true; } bool FillingRow::shift(const FillingRow & next_original_row, bool& value_changed) diff --git a/src/Interpreters/FillingRow.h b/src/Interpreters/FillingRow.h index d33e3f95541..d4590d7b81c 100644 --- a/src/Interpreters/FillingRow.h +++ b/src/Interpreters/FillingRow.h @@ -25,10 +25,8 @@ public: explicit FillingRow(const SortDescription & sort_description); /// Generates next row according to fill 'from', 'to' and 'step' values. - /// Return pair of boolean - /// apply - true if filling values should be inserted into result set - /// value_changed - true if filling row value was changed - std::pair next(const FillingRow & next_original_row); + /// Returns true if filling values should be inserted into result set + bool next(const FillingRow & next_original_row, bool& value_changed); /// Returns true if need to generate some prefix for to_row bool shift(const FillingRow & next_original_row, bool& value_changed); diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index 40650b485f8..f23ffec43de 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -11,7 +11,6 @@ #include #include #include -#include "Interpreters/FillingRow.h" #include @@ -534,9 +533,7 @@ bool FillingTransform::generateSuffixIfNeeded( bool filling_row_changed = false; while (true) { - const auto [apply, changed] = filling_row.next(next_row); - filling_row_changed = changed; - if (!apply) + if (!filling_row.next(next_row, filling_row_changed)) break; interpolate(result_columns, interpolate_block); @@ -660,9 +657,7 @@ void FillingTransform::transformRange( bool filling_row_changed = false; while (true) { - const auto [apply, changed] = filling_row.next(next_row); - filling_row_changed = changed; - if (!apply) + if (!filling_row.next(next_row, filling_row_changed)) break; interpolate(result_columns, interpolate_block); @@ -670,35 +665,22 @@ void FillingTransform::transformRange( copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); } + /// Initialize staleness border for current row to generate it's prefix + filling_row.initStalenessRow(input_fill_columns, row_ind); + + while (filling_row.shift(next_row, filling_row_changed)) { - filling_row.initStalenessRow(input_fill_columns, row_ind); + logDebug("filling_row after shift", filling_row); - bool shift_apply = filling_row.shift(next_row, filling_row_changed); - logDebug("shift_apply", shift_apply); - logDebug("filling_row_changed", filling_row_changed); - - while (shift_apply) + do { - logDebug("after shift", filling_row); + logDebug("inserting prefix filling_row", filling_row); - while (true) - { - logDebug("filling_row in prefix", filling_row); + interpolate(result_columns, interpolate_block); + insertFromFillingRow(res_fill_columns, res_interpolate_columns, res_other_columns, interpolate_block); + copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); - interpolate(result_columns, interpolate_block); - insertFromFillingRow(res_fill_columns, res_interpolate_columns, res_other_columns, interpolate_block); - copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); - - const auto [apply, changed] = filling_row.next(next_row); - logDebug("filling_row in prefix", filling_row); - - filling_row_changed = changed; - if (!apply) - break; - } - - shift_apply = filling_row.shift(next_row, filling_row_changed); - } + } while (filling_row.next(next_row, filling_row_changed)); } /// new valid filling row was generated but not inserted, will use it during suffix generation From ab5738b9f1e87cf8b49b3d74a3bbd05e53c39850 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 16:11:40 +0000 Subject: [PATCH 344/680] merge constraints --- src/Interpreters/FillingRow.cpp | 92 +++++++------------ src/Interpreters/FillingRow.h | 4 +- .../Transforms/FillingTransform.cpp | 4 +- 3 files changed, 37 insertions(+), 63 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index df93ece2af4..67827567e04 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -32,7 +32,10 @@ FillingRow::FillingRow(const SortDescription & sort_description_) : sort_description(sort_description_) { row.resize(sort_description.size()); - staleness_border.resize(sort_description.size()); + + constraints.reserve(sort_description.size()); + for (size_t i = 0; i < size(); ++i) + constraints.push_back(getFillDescription(i).fill_to); } bool FillingRow::operator<(const FillingRow & other) const @@ -96,53 +99,33 @@ std::optional FillingRow::doLongJump(const FillColumnDescription & descr, bool FillingRow::hasSomeConstraints(size_t pos) const { - const auto & descr = getFillDescription(pos); - - if (!descr.fill_to.isNull()) - return true; - - if (!descr.fill_staleness.isNull()) - return true; - - return false; + return !constraints[pos].isNull(); } bool FillingRow::isConstraintsComplete(size_t pos) const { - auto logger = getLogger("FillingRow::isConstraintComplete"); + auto logger = getLogger("FillingRow::isConstraintsComplete"); chassert(!row[pos].isNull()); chassert(hasSomeConstraints(pos)); - const auto & descr = getFillDescription(pos); int direction = getDirection(pos); + LOG_DEBUG(logger, "constraint: {}, row: {}, direction: {}", constraints[pos].dump(), row[pos].dump(), direction); - if (!descr.fill_to.isNull() && !less(row[pos], descr.fill_to, direction)) - { - LOG_DEBUG(logger, "fill to: {}, row: {}, direction: {}", descr.fill_to.dump(), row[pos].dump(), direction); - return false; - } - - if (!descr.fill_staleness.isNull() && !less(row[pos], staleness_border[pos], direction)) - { - LOG_DEBUG(logger, "staleness border: {}, row: {}, direction: {}", staleness_border[pos].dump(), row[pos].dump(), direction); - return false; - } - - return true; + return less(row[pos], constraints[pos], direction); } -Field findMin(Field a, Field b, Field c, int dir) +static const Field & findBorder(const Field & constraint, const Field & next_original, int direction) { - auto logger = getLogger("FillingRow"); - LOG_DEBUG(logger, "a: {} b: {} c: {}", a.dump(), b.dump(), c.dump()); + if (constraint.isNull()) + return next_original; - if (a.isNull() || (!b.isNull() && less(b, a, dir))) - a = b; + if (next_original.isNull()) + return constraint; - if (a.isNull() || (!c.isNull() && less(c, a, dir))) - a = c; + if (less(constraint, next_original, direction)) + return constraint; - return a; + return next_original; } bool FillingRow::next(const FillingRow & next_original_row, bool& value_changed) @@ -158,11 +141,10 @@ bool FillingRow::next(const FillingRow & next_original_row, bool& value_changed) if (row[pos].isNull()) continue; - const auto & descr = getFillDescription(pos); - auto min_constr = findMin(next_original_row[pos], staleness_border[pos], descr.fill_to, getDirection(pos)); - LOG_DEBUG(logger, "min_constr: {}", min_constr); + const Field & border = findBorder(constraints[pos], next_original_row[pos], getDirection(pos)); + LOG_DEBUG(logger, "border: {}", border); - if (!min_constr.isNull() && !equals(row[pos], min_constr)) + if (!border.isNull() && !equals(row[pos], border)) break; } @@ -171,15 +153,10 @@ bool FillingRow::next(const FillingRow & next_original_row, bool& value_changed) if (pos == row_size) return false; - const auto & pos_descr = getFillDescription(pos); - if (!next_original_row[pos].isNull() && less(next_original_row[pos], row[pos], getDirection(pos))) return false; - if (!staleness_border[pos].isNull() && !less(row[pos], staleness_border[pos], getDirection(pos))) - return false; - - if (!pos_descr.fill_to.isNull() && !less(row[pos], pos_descr.fill_to, getDirection(pos))) + if (!constraints[pos].isNull() && !less(row[pos], constraints[pos], getDirection(pos))) return false; /// If we have any 'fill_to' value at position greater than 'pos' or configured staleness, @@ -191,16 +168,13 @@ bool FillingRow::next(const FillingRow & next_original_row, bool& value_changed) if (row[i].isNull()) continue; - if (fill_column_desc.fill_to.isNull() && staleness_border[i].isNull()) + if (constraints[i].isNull()) continue; Field next_value = row[i]; fill_column_desc.step_func(next_value, 1); - if (!staleness_border[i].isNull() && !less(next_value, staleness_border[i], getDirection(i))) - continue; - - if (!fill_column_desc.fill_to.isNull() && !less(next_value, fill_column_desc.fill_to, getDirection(i))) + if (!less(next_value, constraints[i], getDirection(i))) continue; row[i] = next_value; @@ -216,10 +190,7 @@ bool FillingRow::next(const FillingRow & next_original_row, bool& value_changed) if (!next_original_row[pos].isNull() && less(next_original_row[pos], next_value, getDirection(pos))) return false; - if (!staleness_border[pos].isNull() && !less(next_value, staleness_border[pos], getDirection(pos))) - return false; - - if (!pos_descr.fill_to.isNull() && !less(next_value, pos_descr.fill_to, getDirection(pos))) + if (!constraints[pos].isNull() && !less(next_value, constraints[pos], getDirection(pos))) return false; row[pos] = next_value; @@ -236,8 +207,7 @@ bool FillingRow::next(const FillingRow & next_original_row, bool& value_changed) is_less |= ( (next_original_row[i].isNull() || less(row[i], next_original_row[i], getDirection(i))) && - (staleness_border[i].isNull() || less(row[i], staleness_border[i], getDirection(i))) && - (descr.fill_to.isNull() || less(row[i], descr.fill_to, getDirection(i))) + (constraints[i].isNull() || less(row[i], constraints[i], getDirection(i))) ); } @@ -291,8 +261,7 @@ bool FillingRow::shift(const FillingRow & next_original_row, bool& value_changed is_less |= ( (next_original_row[i].isNull() || less(row[i], next_original_row[i], getDirection(i))) && - (staleness_border[i].isNull() || less(row[i], staleness_border[i], getDirection(i))) && - (descr.fill_to.isNull() || less(row[i], descr.fill_to, getDirection(i))) + (constraints[i].isNull() || less(row[i], constraints[i], getDirection(i))) ); } @@ -347,15 +316,20 @@ void FillingRow::initUsingTo(size_t from_pos) row[i] = getFillDescription(i).fill_to; } -void FillingRow::initStalenessRow(const Columns& base_row, size_t row_ind) +void FillingRow::updateConstraintsWithStalenessRow(const Columns& base_row, size_t row_ind) { for (size_t i = 0; i < size(); ++i) { const auto& descr = getFillDescription(i); + constraints[i] = descr.fill_to; + if (!descr.fill_staleness.isNull()) { - staleness_border[i] = (*base_row[i])[row_ind]; - descr.staleness_step_func(staleness_border[i], 1); + Field staleness_border = (*base_row[i])[row_ind]; + descr.staleness_step_func(staleness_border, 1); + + if (constraints[i].isNull() || less(staleness_border, constraints[i], getDirection(i))) + constraints[i] = std::move(staleness_border); } } } diff --git a/src/Interpreters/FillingRow.h b/src/Interpreters/FillingRow.h index d4590d7b81c..edcaba02aa7 100644 --- a/src/Interpreters/FillingRow.h +++ b/src/Interpreters/FillingRow.h @@ -36,7 +36,7 @@ public: void initUsingFrom(size_t from_pos = 0); void initUsingTo(size_t from_pos = 0); - void initStalenessRow(const Columns& base_row, size_t row_ind); + void updateConstraintsWithStalenessRow(const Columns& base_row, size_t row_ind); Field & operator[](size_t index) { return row[index]; } const Field & operator[](size_t index) const { return row[index]; } @@ -54,7 +54,7 @@ public: private: Row row; - Row staleness_border; + Row constraints; SortDescription sort_description; }; diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index f23ffec43de..407a79efb93 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -628,7 +628,7 @@ void FillingTransform::transformRange( } /// Init staleness first interval - filling_row.initStalenessRow(input_fill_columns, range_begin); + filling_row.updateConstraintsWithStalenessRow(input_fill_columns, range_begin); for (size_t row_ind = range_begin; row_ind < range_end; ++row_ind) { @@ -666,7 +666,7 @@ void FillingTransform::transformRange( } /// Initialize staleness border for current row to generate it's prefix - filling_row.initStalenessRow(input_fill_columns, row_ind); + filling_row.updateConstraintsWithStalenessRow(input_fill_columns, row_ind); while (filling_row.shift(next_row, filling_row_changed)) { From 5b4d55dd3f0ff4393e81a7a36ad092eee46be2c6 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 16:41:02 +0000 Subject: [PATCH 345/680] move logs under flag --- src/Interpreters/FillingRow.cpp | 33 +++++++++---------- .../Transforms/FillingTransform.cpp | 2 +- 2 files changed, 16 insertions(+), 19 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index 67827567e04..deb4c765d31 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -10,6 +10,15 @@ namespace DB { +constexpr static bool debug_logging_enabled = true; + +template +static void logDebug(String fmt_str, Args&&... args) +{ + if constexpr (debug_logging_enabled) + LOG_DEBUG(getLogger("FillingRow"), "{}", fmt::format(fmt::runtime(fmt_str), std::forward(args)...)); +} + bool less(const Field & lhs, const Field & rhs, int direction) { if (direction == -1) @@ -104,12 +113,11 @@ bool FillingRow::hasSomeConstraints(size_t pos) const bool FillingRow::isConstraintsComplete(size_t pos) const { - auto logger = getLogger("FillingRow::isConstraintsComplete"); chassert(!row[pos].isNull()); chassert(hasSomeConstraints(pos)); int direction = getDirection(pos); - LOG_DEBUG(logger, "constraint: {}, row: {}, direction: {}", constraints[pos].dump(), row[pos].dump(), direction); + logDebug("constraint: {}, row: {}, direction: {}", constraints[pos].dump(), row[pos].dump(), direction); return less(row[pos], constraints[pos], direction); } @@ -130,7 +138,6 @@ static const Field & findBorder(const Field & constraint, const Field & next_ori bool FillingRow::next(const FillingRow & next_original_row, bool& value_changed) { - auto logger = getLogger("FillingRow"); const size_t row_size = size(); size_t pos = 0; @@ -142,13 +149,13 @@ bool FillingRow::next(const FillingRow & next_original_row, bool& value_changed) continue; const Field & border = findBorder(constraints[pos], next_original_row[pos], getDirection(pos)); - LOG_DEBUG(logger, "border: {}", border); + logDebug("border: {}", border); if (!border.isNull() && !equals(row[pos], border)) break; } - LOG_DEBUG(logger, "pos: {}", pos); + logDebug("pos: {}", pos); if (pos == row_size) return false; @@ -223,8 +230,7 @@ bool FillingRow::next(const FillingRow & next_original_row, bool& value_changed) bool FillingRow::shift(const FillingRow & next_original_row, bool& value_changed) { - auto logger = getLogger("FillingRow::shift"); - LOG_DEBUG(logger, "next_original_row: {}, current: {}", next_original_row.dump(), dump()); + logDebug("next_original_row: {}, current: {}", next_original_row.dump(), dump()); for (size_t pos = 0; pos < size(); ++pos) { @@ -235,16 +241,7 @@ bool FillingRow::shift(const FillingRow & next_original_row, bool& value_changed return false; std::optional next_value = doLongJump(getFillDescription(pos), pos, next_original_row[pos]); - - if (!next_value.has_value()) - { - LOG_DEBUG(logger, "next value: {}", "None"); - continue; - } - else - { - LOG_DEBUG(logger, "next value: {}", next_value->dump()); - } + logDebug("jumped to next value: {}", next_value.value_or("Did not complete")); row[pos] = std::move(next_value.value()); @@ -265,7 +262,7 @@ bool FillingRow::shift(const FillingRow & next_original_row, bool& value_changed ); } - LOG_DEBUG(logger, "is less: {}", is_less); + logDebug("is less: {}", is_less); value_changed = true; return is_less; diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index 407a79efb93..81d93a6eadb 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -17,7 +17,7 @@ namespace DB { -constexpr bool debug_logging_enabled = true; +constexpr static bool debug_logging_enabled = true; template static void logDebug(String key, const T & value, const char * separator = " : ") From 82783fe020b83425590ab14949d5b5face7c9fd6 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 16:41:38 +0000 Subject: [PATCH 346/680] disable logs --- src/Interpreters/FillingRow.cpp | 2 +- src/Processors/Transforms/FillingTransform.cpp | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index deb4c765d31..3b40c2b6cdd 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -10,7 +10,7 @@ namespace DB { -constexpr static bool debug_logging_enabled = true; +constexpr static bool debug_logging_enabled = false; template static void logDebug(String fmt_str, Args&&... args) diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index 81d93a6eadb..dc0bafba3e3 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -17,7 +17,7 @@ namespace DB { -constexpr static bool debug_logging_enabled = true; +constexpr static bool debug_logging_enabled = false; template static void logDebug(String key, const T & value, const char * separator = " : ") From b6bd776355171896abb3ef95d2dfdb204799a4b1 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 17:09:35 +0000 Subject: [PATCH 347/680] cleanup --- src/Interpreters/FillingRow.cpp | 8 ++++---- src/Interpreters/FillingRow.h | 4 ++-- src/Processors/Transforms/FillingTransform.cpp | 6 +++--- 3 files changed, 9 insertions(+), 9 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index 3b40c2b6cdd..98c18e9b2ae 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -13,7 +13,7 @@ namespace DB constexpr static bool debug_logging_enabled = false; template -static void logDebug(String fmt_str, Args&&... args) +inline static void logDebug(String fmt_str, Args&&... args) { if constexpr (debug_logging_enabled) LOG_DEBUG(getLogger("FillingRow"), "{}", fmt::format(fmt::runtime(fmt_str), std::forward(args)...)); @@ -111,7 +111,7 @@ bool FillingRow::hasSomeConstraints(size_t pos) const return !constraints[pos].isNull(); } -bool FillingRow::isConstraintsComplete(size_t pos) const +bool FillingRow::isConstraintsSatisfied(size_t pos) const { chassert(!row[pos].isNull()); chassert(hasSomeConstraints(pos)); @@ -288,14 +288,14 @@ bool FillingRow::hasSomeConstraints() const return false; } -bool FillingRow::isConstraintsComplete() const +bool FillingRow::isConstraintsSatisfied() const { for (size_t pos = 0; pos < size(); ++pos) { if (row[pos].isNull() || !hasSomeConstraints(pos)) continue; - return isConstraintsComplete(pos); + return isConstraintsSatisfied(pos); } return true; diff --git a/src/Interpreters/FillingRow.h b/src/Interpreters/FillingRow.h index edcaba02aa7..08d624a2405 100644 --- a/src/Interpreters/FillingRow.h +++ b/src/Interpreters/FillingRow.h @@ -19,7 +19,7 @@ class FillingRow std::optional doLongJump(const FillColumnDescription & descr, size_t column_ind, const Field & to); bool hasSomeConstraints(size_t pos) const; - bool isConstraintsComplete(size_t pos) const; + bool isConstraintsSatisfied(size_t pos) const; public: explicit FillingRow(const SortDescription & sort_description); @@ -32,7 +32,7 @@ public: bool shift(const FillingRow & next_original_row, bool& value_changed); bool hasSomeConstraints() const; - bool isConstraintsComplete() const; + bool isConstraintsSatisfied() const; void initUsingFrom(size_t from_pos = 0); void initUsingTo(size_t from_pos = 0); diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index dc0bafba3e3..a5c6460db0a 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -20,7 +20,7 @@ namespace DB constexpr static bool debug_logging_enabled = false; template -static void logDebug(String key, const T & value, const char * separator = " : ") +inline static void logDebug(String key, const T & value, const char * separator = " : ") { if constexpr (debug_logging_enabled) { @@ -511,7 +511,7 @@ bool FillingTransform::generateSuffixIfNeeded( logDebug("generateSuffixIfNeeded next_row updated", next_row); - if (!filling_row.hasSomeConstraints() || !filling_row.isConstraintsComplete()) + if (!filling_row.hasSomeConstraints() || !filling_row.isConstraintsSatisfied()) { logDebug("generateSuffixIfNeeded", "will not generate suffix"); return false; @@ -647,7 +647,7 @@ void FillingTransform::transformRange( /// The condition is true when filling row is initialized by value(s) in FILL FROM, /// and there are row(s) in current range with value(s) < then in the filling row. /// It can happen only once for a range. - if (should_insert_first && filling_row < next_row && filling_row.isConstraintsComplete()) + if (should_insert_first && filling_row < next_row && filling_row.isConstraintsSatisfied()) { interpolate(result_columns, interpolate_block); insertFromFillingRow(res_fill_columns, res_interpolate_columns, res_other_columns, interpolate_block); From c8b94a3c61330fb0649ee92ec69ffe6e6059860b Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 17:21:29 +0000 Subject: [PATCH 348/680] fix empty stream filling --- src/Processors/Transforms/FillingTransform.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index a5c6460db0a..4a8965dcfaa 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -503,7 +503,7 @@ bool FillingTransform::generateSuffixIfNeeded( logDebug("generateSuffixIfNeeded next_row", next_row); /// Determines if we should insert filling row before start generating next rows - bool should_insert_first = (next_row < filling_row && !filling_row_inserted) || next_row.isNull(); + bool should_insert_first = (next_row < filling_row && !filling_row_inserted) || (next_row.isNull() && !filling_row.isNull()); logDebug("should_insert_first", should_insert_first); for (size_t i = 0, size = filling_row.size(); i < size; ++i) From a99428fcd9d10da6b6f6fea10d033b485e558b1c Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 17:25:06 +0000 Subject: [PATCH 349/680] add errors test --- .../0_stateless/03266_with_fill_staleness_errors.reference | 0 .../queries/0_stateless/03266_with_fill_staleness_errors.sql | 5 +++++ 2 files changed, 5 insertions(+) create mode 100644 tests/queries/0_stateless/03266_with_fill_staleness_errors.reference create mode 100644 tests/queries/0_stateless/03266_with_fill_staleness_errors.sql diff --git a/tests/queries/0_stateless/03266_with_fill_staleness_errors.reference b/tests/queries/0_stateless/03266_with_fill_staleness_errors.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03266_with_fill_staleness_errors.sql b/tests/queries/0_stateless/03266_with_fill_staleness_errors.sql new file mode 100644 index 00000000000..339747e4343 --- /dev/null +++ b/tests/queries/0_stateless/03266_with_fill_staleness_errors.sql @@ -0,0 +1,5 @@ +SET enable_analyzer=1; + +SELECT 1 AS a, 2 AS b ORDER BY a, b WITH FILL FROM 0 TO 10 STALENESS 3; -- { serverError INVALID_WITH_FILL_EXPRESSION } +SELECT 1 AS a, 2 AS b ORDER BY a, b DESC WITH FILL FROM 0 TO 10 STALENESS 3; -- { serverError INVALID_WITH_FILL_EXPRESSION } +SELECT 1 AS a, 2 AS b ORDER BY a, b ASC WITH FILL FROM 0 TO 10 STALENESS -3; -- { serverError INVALID_WITH_FILL_EXPRESSION } From 10088a0947aaf16a3ce1664c422d66daea3324d2 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 17:26:31 +0000 Subject: [PATCH 350/680] extend fuzzer dict with staleness --- tests/fuzz/dictionaries/keywords.dict | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/fuzz/dictionaries/keywords.dict b/tests/fuzz/dictionaries/keywords.dict index abaaf9e53b5..a37675ebcad 100644 --- a/tests/fuzz/dictionaries/keywords.dict +++ b/tests/fuzz/dictionaries/keywords.dict @@ -538,6 +538,7 @@ "WITH ADMIN OPTION" "WITH CHECK" "WITH FILL" +"STALENESS" "WITH GRANT OPTION" "WITH NAME" "WITH REPLACE OPTION" From e50176c62f18a95648c6b65627b17a095bdccbe5 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Wed, 30 Oct 2024 17:29:08 +0000 Subject: [PATCH 351/680] improve test --- .../queries/0_stateless/03266_with_fill_staleness_errors.sql | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/03266_with_fill_staleness_errors.sql b/tests/queries/0_stateless/03266_with_fill_staleness_errors.sql index 339747e4343..fbfaf3743ca 100644 --- a/tests/queries/0_stateless/03266_with_fill_staleness_errors.sql +++ b/tests/queries/0_stateless/03266_with_fill_staleness_errors.sql @@ -1,5 +1,5 @@ SET enable_analyzer=1; SELECT 1 AS a, 2 AS b ORDER BY a, b WITH FILL FROM 0 TO 10 STALENESS 3; -- { serverError INVALID_WITH_FILL_EXPRESSION } -SELECT 1 AS a, 2 AS b ORDER BY a, b DESC WITH FILL FROM 0 TO 10 STALENESS 3; -- { serverError INVALID_WITH_FILL_EXPRESSION } -SELECT 1 AS a, 2 AS b ORDER BY a, b ASC WITH FILL FROM 0 TO 10 STALENESS -3; -- { serverError INVALID_WITH_FILL_EXPRESSION } +SELECT 1 AS a, 2 AS b ORDER BY a, b DESC WITH FILL TO 10 STALENESS 3; -- { serverError INVALID_WITH_FILL_EXPRESSION } +SELECT 1 AS a, 2 AS b ORDER BY a, b ASC WITH FILL TO 10 STALENESS -3; -- { serverError INVALID_WITH_FILL_EXPRESSION } From 0cfbe95ca69d0bb52578c83570b34f4f40de92df Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Wed, 30 Oct 2024 21:20:11 +0100 Subject: [PATCH 352/680] Update 03258_multiple_array_joins.sql --- tests/queries/0_stateless/03258_multiple_array_joins.sql | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/03258_multiple_array_joins.sql b/tests/queries/0_stateless/03258_multiple_array_joins.sql index 5afe7725d3f..ddfac1da080 100644 --- a/tests/queries/0_stateless/03258_multiple_array_joins.sql +++ b/tests/queries/0_stateless/03258_multiple_array_joins.sql @@ -1,3 +1,4 @@ +SET enable_analyzer = 1; DROP TABLE IF EXISTS test_multiple_array_join; CREATE TABLE test_multiple_array_join ( From fc1fd46686722c5bb13c95edf7051c4e21be7b68 Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Wed, 30 Oct 2024 23:36:15 +0100 Subject: [PATCH 353/680] fix test --- ...eplicas_join_algo_and_analyzer_4.reference | 29 ++++++++++++++++++ ...allel_replicas_join_algo_and_analyzer_4.sh | 30 +++++++++++-------- 2 files changed, 46 insertions(+), 13 deletions(-) diff --git a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference index 9fc156b5fb0..8464317f7e6 100644 --- a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference +++ b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference @@ -27,3 +27,32 @@ SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` GROUP 500030000 500040000 SELECT sum(`__table1`.`item_id`) AS `sum(item_id)` FROM (SELECT `__table2`.`item_id` AS `item_id`, `__table2`.`price_sold` AS `price_sold` FROM `default`.`t` AS `__table2`) AS `__table1` ALL LEFT JOIN (SELECT `__table4`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table4`) AS `__table3` ON `__table1`.`item_id` = `__table3`.`item_id` GROUP BY `__table1`.`price_sold` ORDER BY `__table1`.`price_sold` ASC +4999950000 +4999950000 +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t` AS `__table1` GROUP BY `__table1`.`item_id` +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` +4999950000 +4999950000 +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t` AS `__table1` +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` GROUP BY `__table1`.`item_id` +499950000 +499960000 +499970000 +499980000 +499990000 +500000000 +500010000 +500020000 +500030000 +500040000 +499950000 +499960000 +499970000 +499980000 +499990000 +500000000 +500010000 +500020000 +500030000 +500040000 +SELECT sum(`__table1`.`item_id`) AS `sum(item_id)` FROM (SELECT `__table2`.`item_id` AS `item_id`, `__table2`.`price_sold` AS `price_sold` FROM `default`.`t` AS `__table2`) AS `__table1` ALL LEFT JOIN (SELECT `__table4`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table4`) AS `__table3` ON `__table1`.`item_id` = `__table3`.`item_id` GROUP BY `__table1`.`price_sold` ORDER BY `__table1`.`price_sold` ASC diff --git a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh index a588fa47c2d..0e1f07b6ac5 100755 --- a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh +++ b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh @@ -1,4 +1,5 @@ #!/usr/bin/env bash +# Tags: long, no-random-settings, no-random-merge-tree-settings CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh @@ -74,20 +75,23 @@ query3=" ORDER BY price_sold " -for query in "${query1}" "${query2}" "${query3}"; do - for enable_parallel_replicas in {0..1}; do - ${CLICKHOUSE_CLIENT} --query=" - set enable_analyzer=1; - set allow_experimental_parallel_reading_from_replicas=${enable_parallel_replicas}, cluster_for_parallel_replicas='parallel_replicas', max_parallel_replicas=100, parallel_replicas_for_non_replicated_merge_tree=1; +for prefer_local_plan in {0..1}; do + for query in "${query1}" "${query2}" "${query3}"; do + for enable_parallel_replicas in {0..1}; do + ${CLICKHOUSE_CLIENT} --query=" + set enable_analyzer=1; + set parallel_replicas_local_plan=${prefer_local_plan}; + set allow_experimental_parallel_reading_from_replicas=${enable_parallel_replicas}, cluster_for_parallel_replicas='parallel_replicas', max_parallel_replicas=100, parallel_replicas_for_non_replicated_merge_tree=1; - ${query}; + ${query}; - SELECT replaceRegexpAll(explain, '.*Query: (.*) Replicas:.*', '\\1') - FROM - ( - EXPLAIN actions=1 ${query} - ) - WHERE explain LIKE '%ParallelReplicas%'; - " + SELECT replaceRegexpAll(explain, '.*Query: (.*) Replicas:.*', '\\1') + FROM + ( + EXPLAIN actions=1 ${query} + ) + WHERE explain LIKE '%ParallelReplicas%'; + " + done done done From 4e2693bb466a07ab06d4155a091e6782a495ed45 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Thu, 31 Oct 2024 02:01:23 +0000 Subject: [PATCH 354/680] add test --- ...in_order_optimization_with_virtual_row.sql | 5 ++--- ...ization_with_virtual_row_special.reference | 2 ++ ..._optimization_with_virtual_row_special.sql | 21 +++++++++++++++++++ 3 files changed, 25 insertions(+), 3 deletions(-) create mode 100644 tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.reference create mode 100644 tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index f66b4be2c69..8826f2c27cf 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -156,7 +156,6 @@ read_in_order_two_level_merge_threshold = 5; --avoid preliminary merge DROP TABLE fixed_prefix; SELECT '========'; --- currently don't support virtual row in this case DROP TABLE IF EXISTS function_pk; CREATE TABLE function_pk @@ -179,7 +178,7 @@ ORDER BY (A,-B) ASC limit 3 SETTINGS max_threads = 1, optimize_read_in_order = 1, -read_in_order_two_level_merge_threshold = 0; --force preliminary merge +read_in_order_two_level_merge_threshold = 5; --avoid preliminary merge DROP TABLE function_pk; @@ -214,4 +213,4 @@ SETTINGS read_in_order_two_level_merge_threshold = 0, optimize_read_in_order = 1, max_threads = 2; -DROP TABLE distinct_in_order; +DROP TABLE distinct_in_order; \ No newline at end of file diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.reference b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.reference new file mode 100644 index 00000000000..b03759364cf --- /dev/null +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.reference @@ -0,0 +1,2 @@ +dist +src diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql new file mode 100644 index 00000000000..ee7336bdf02 --- /dev/null +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql @@ -0,0 +1,21 @@ +-- Tags: no-parallel + +-- modified from test_01155_ordinary +DROP DATABASE IF EXISTS test_01155_ordinary; + +SET allow_deprecated_database_ordinary = 1; + +CREATE DATABASE test_01155_ordinary ENGINE = Ordinary; + +USE test_01155_ordinary; + +CREATE TABLE src (s String) ENGINE = MergeTree() ORDER BY s; +INSERT INTO src(s) VALUES ('before moving tables'); +CREATE TABLE dist (s String) ENGINE = Distributed(test_shard_localhost, test_01155_ordinary, src); + +SET enable_analyzer=0; +SELECT _table FROM merge('test_01155_ordinary', '') ORDER BY _table, s; + +DROP TABLE src; +DROP TABLE dist; +DROP DATABASE test_01155_ordinary; \ No newline at end of file From b229fb1664c8ed5b2c19ff569bb94c51e2f8cbec Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Thu, 31 Oct 2024 12:04:24 +0000 Subject: [PATCH 355/680] Check if the mutation query is valid. --- src/Interpreters/MutationsInterpreter.cpp | 3 +++ .../03256_invalid_mutation_query.reference | 0 .../03256_invalid_mutation_query.sql | 19 +++++++++++++++++++ 3 files changed, 22 insertions(+) create mode 100644 tests/queries/0_stateless/03256_invalid_mutation_query.reference create mode 100644 tests/queries/0_stateless/03256_invalid_mutation_query.sql diff --git a/src/Interpreters/MutationsInterpreter.cpp b/src/Interpreters/MutationsInterpreter.cpp index 0f25d5ac21c..da99b217341 100644 --- a/src/Interpreters/MutationsInterpreter.cpp +++ b/src/Interpreters/MutationsInterpreter.cpp @@ -1386,6 +1386,9 @@ void MutationsInterpreter::validate() } } + // Make sure the mutations query is valid + prepareQueryAffectedQueryTree(commands, source.getStorage(), context); + QueryPlan plan; initQueryPlan(stages.front(), plan); diff --git a/tests/queries/0_stateless/03256_invalid_mutation_query.reference b/tests/queries/0_stateless/03256_invalid_mutation_query.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03256_invalid_mutation_query.sql b/tests/queries/0_stateless/03256_invalid_mutation_query.sql new file mode 100644 index 00000000000..010f96414d4 --- /dev/null +++ b/tests/queries/0_stateless/03256_invalid_mutation_query.sql @@ -0,0 +1,19 @@ +DROP TABLE IF EXISTS t; +DROP TABLE IF EXISTS t2; + +CREATE TABLE t (x int) ENGINE = MergeTree() ORDER BY (); + +DELETE FROM t WHERE y in (SELECT y FROM t); -- { serverError 47 } +DELETE FROM t WHERE x in (SELECT y FROM t); -- { serverError 47 } +DELETE FROM t WHERE x IN (SELECT * FROM t2); -- { serverError 60 } +ALTER TABLE t DELETE WHERE x in (SELECT y FROM t); -- { serverError 47 } +ALTER TABLE t UPDATE x = 1 WHERE x IN (SELECT y FROM t); -- { serverError 47 } + +ALTER TABLE t ADD COLUMN y int; +DELETE FROM t WHERE y in (SELECT y FROM t); + +CREATE TABLE t2 (x int) ENGINE = MergeTree() ORDER BY (); +DELETE FROM t WHERE x IN (SELECT * FROM t2); + +DROP TABLE t; +DROP TABLE t2; From 41e4076c5c0b7207327e7a9eff143a8346a936cd Mon Sep 17 00:00:00 2001 From: kssenii Date: Thu, 31 Oct 2024 13:18:30 +0100 Subject: [PATCH 356/680] Fix test --- src/Storages/ObjectStorage/StorageObjectStorageSource.cpp | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp index 90871b8c0ad..a1737c55c26 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp @@ -521,6 +521,8 @@ std::unique_ptr StorageObjectStorageSource::createReadBu size_t buffer_size = prefer_bigger_buffer_size ? std::max(read_settings.remote_fs_buffer_size, DBMS_DEFAULT_BUFFER_SIZE) : read_settings.remote_fs_buffer_size; + if (object_size) + buffer_size = std::min(object_size, buffer_size); auto & reader = context_->getThreadPoolReader(FilesystemReaderType::ASYNCHRONOUS_REMOTE_FS_READER); impl = std::make_unique( From 1563689c034992866c2de6ede7776c41888395ac Mon Sep 17 00:00:00 2001 From: kssenii Date: Thu, 31 Oct 2024 13:31:54 +0100 Subject: [PATCH 357/680] Transfer changes from sync --- src/Core/Settings.cpp | 6 +++++ src/Core/SettingsChangesHistory.cpp | 4 +++- .../IO/CachedOnDiskReadBufferFromFile.cpp | 6 +++++ src/IO/ReadSettings.h | 2 ++ src/Interpreters/Cache/FileSegment.cpp | 9 +++++++- src/Interpreters/Context.cpp | 5 +++++ src/Storages/MergeTree/DataPartsExchange.cpp | 2 +- src/Storages/MergeTree/IMergeTreeDataPart.cpp | 2 +- src/Storages/MergeTree/MergeTask.cpp | 4 ++-- src/Storages/MergeTree/MergeTreeData.cpp | 22 +++++++++---------- src/Storages/MergeTree/MergeTreeData.h | 2 +- .../MergeTree/MergeTreeDataPartBuilder.cpp | 18 +++++++++------ .../MergeTree/MergeTreeDataPartBuilder.h | 12 ++++++---- .../MergeTree/MergeTreeDataWriter.cpp | 2 +- .../MergeTree/MergeTreePartsMover.cpp | 2 +- src/Storages/MergeTree/MutateTask.cpp | 2 +- src/Storages/StorageReplicatedMergeTree.cpp | 2 +- 17 files changed, 69 insertions(+), 33 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index cdaa305e804..6b16cc132bc 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -4842,6 +4842,12 @@ Limit on size of a single batch of file segments that a read buffer can request )", 0) \ M(UInt64, filesystem_cache_reserve_space_wait_lock_timeout_milliseconds, 1000, R"( Wait time to lock cache for space reservation in filesystem cache +)", 0) \ + M(Bool, filesystem_cache_enable_background_download_for_metadata_files_in_packed_storage, true, R"( +Wait time to lock cache for space reservation in filesystem cache +)", 0) \ + M(Bool, filesystem_cache_enable_background_download_during_fetch, true, R"( +Wait time to lock cache for space reservation in filesystem cache )", 0) \ M(UInt64, temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds, (10 * 60 * 1000), R"( Wait time to lock cache for space reservation for temporary data in filesystem cache diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index ad9499c6d86..c36add485bb 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -103,7 +103,9 @@ static std::initializer_listfront(), read_type); } + + if (file_segments && !file_segments->empty() && !file_segments->front().isCompleted()) + { + file_segments->completeAndPopFront(settings.filesystem_cache_allow_background_download); + file_segments = {}; + } } void CachedOnDiskReadBufferFromFile::predownload(FileSegment & file_segment) diff --git a/src/IO/ReadSettings.h b/src/IO/ReadSettings.h index ac3d7fc9faf..24392891e72 100644 --- a/src/IO/ReadSettings.h +++ b/src/IO/ReadSettings.h @@ -107,6 +107,8 @@ struct ReadSettings size_t filesystem_cache_segments_batch_size = 20; size_t filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = 1000; bool filesystem_cache_allow_background_download = true; + bool filesystem_cache_allow_background_download_for_metadata_files_in_packed_storage = true; + bool filesystem_cache_allow_background_download_during_fetch = true; bool use_page_cache_for_disks_without_file_cache = false; bool read_from_page_cache_if_exists_otherwise_bypass_cache = false; diff --git a/src/Interpreters/Cache/FileSegment.cpp b/src/Interpreters/Cache/FileSegment.cpp index 7081ac81ae4..5e42bf0113a 100644 --- a/src/Interpreters/Cache/FileSegment.cpp +++ b/src/Interpreters/Cache/FileSegment.cpp @@ -1003,7 +1003,14 @@ void FileSegmentsHolder::reset() ProfileEvents::increment(ProfileEvents::FilesystemCacheUnusedHoldFileSegments, file_segments.size()); for (auto file_segment_it = file_segments.begin(); file_segment_it != file_segments.end();) - file_segment_it = completeAndPopFrontImpl(false); + { + /// One might think it would have been more correct to do `false` here, + /// not to allow background download for file segments that we actually did not start reading. + /// But actually we would only do that, if those file segments were already read partially by some other thread/query + /// but they were not put to the download queue, because current thread was holding them in Holder. + /// So as a culprit, we need to allow to happen what would have happened if we did not exist. + file_segment_it = completeAndPopFrontImpl(true); + } file_segments.clear(); } diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp index 8962be59f86..9b775b9eb61 100644 --- a/src/Interpreters/Context.cpp +++ b/src/Interpreters/Context.cpp @@ -193,6 +193,8 @@ namespace Setting extern const SettingsUInt64 filesystem_cache_max_download_size; extern const SettingsUInt64 filesystem_cache_reserve_space_wait_lock_timeout_milliseconds; extern const SettingsUInt64 filesystem_cache_segments_batch_size; + extern const SettingsBool filesystem_cache_enable_background_download_for_metadata_files_in_packed_storage; + extern const SettingsBool filesystem_cache_enable_background_download_during_fetch; extern const SettingsBool http_make_head_request; extern const SettingsUInt64 http_max_fields; extern const SettingsUInt64 http_max_field_name_size; @@ -5687,6 +5689,9 @@ ReadSettings Context::getReadSettings() const res.filesystem_cache_segments_batch_size = settings_ref[Setting::filesystem_cache_segments_batch_size]; res.filesystem_cache_reserve_space_wait_lock_timeout_milliseconds = settings_ref[Setting::filesystem_cache_reserve_space_wait_lock_timeout_milliseconds]; + res.filesystem_cache_allow_background_download_for_metadata_files_in_packed_storage + = settings_ref[Setting::filesystem_cache_enable_background_download_for_metadata_files_in_packed_storage]; + res.filesystem_cache_allow_background_download_during_fetch = settings_ref[Setting::filesystem_cache_enable_background_download_during_fetch]; res.filesystem_cache_max_download_size = settings_ref[Setting::filesystem_cache_max_download_size]; res.skip_download_if_exceeds_query_cache = settings_ref[Setting::skip_download_if_exceeds_query_cache]; diff --git a/src/Storages/MergeTree/DataPartsExchange.cpp b/src/Storages/MergeTree/DataPartsExchange.cpp index e13ec5a7515..1d79ae5aacb 100644 --- a/src/Storages/MergeTree/DataPartsExchange.cpp +++ b/src/Storages/MergeTree/DataPartsExchange.cpp @@ -908,7 +908,7 @@ MergeTreeData::MutableDataPartPtr Fetcher::downloadPartToDisk( { part_storage_for_loading->commitTransaction(); - MergeTreeDataPartBuilder builder(data, part_name, volume, part_relative_path, part_dir); + MergeTreeDataPartBuilder builder(data, part_name, volume, part_relative_path, part_dir, getReadSettings()); new_data_part = builder.withPartFormatFromDisk().build(); new_data_part->version.setCreationTID(Tx::PrehistoricTID, nullptr); diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp index 20d7528d38a..41783ffddb0 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp @@ -833,7 +833,7 @@ MergeTreeDataPartBuilder IMergeTreeDataPart::getProjectionPartBuilder(const Stri { const char * projection_extension = is_temp_projection ? ".tmp_proj" : ".proj"; auto projection_storage = getDataPartStorage().getProjection(projection_name + projection_extension, !is_temp_projection); - MergeTreeDataPartBuilder builder(storage, projection_name, projection_storage); + MergeTreeDataPartBuilder builder(storage, projection_name, projection_storage, getReadSettings()); return builder.withPartInfo(MergeListElement::FAKE_RESULT_PART_FOR_PROJECTION).withParentPart(this); } diff --git a/src/Storages/MergeTree/MergeTask.cpp b/src/Storages/MergeTree/MergeTask.cpp index 74d6d60ba1b..06471bbe2ba 100644 --- a/src/Storages/MergeTree/MergeTask.cpp +++ b/src/Storages/MergeTree/MergeTask.cpp @@ -342,13 +342,13 @@ bool MergeTask::ExecuteAndFinalizeHorizontalPart::prepare() const if (global_ctx->parent_part) { auto data_part_storage = global_ctx->parent_part->getDataPartStorage().getProjection(local_tmp_part_basename, /* use parent transaction */ false); - builder.emplace(*global_ctx->data, global_ctx->future_part->name, data_part_storage); + builder.emplace(*global_ctx->data, global_ctx->future_part->name, data_part_storage, getReadSettings()); builder->withParentPart(global_ctx->parent_part); } else { auto local_single_disk_volume = std::make_shared("volume_" + global_ctx->future_part->name, global_ctx->disk, 0); - builder.emplace(global_ctx->data->getDataPartBuilder(global_ctx->future_part->name, local_single_disk_volume, local_tmp_part_basename)); + builder.emplace(global_ctx->data->getDataPartBuilder(global_ctx->future_part->name, local_single_disk_volume, local_tmp_part_basename, getReadSettings())); builder->withPartStorageType(global_ctx->future_part->part_format.storage_type); } diff --git a/src/Storages/MergeTree/MergeTreeData.cpp b/src/Storages/MergeTree/MergeTreeData.cpp index 0ebb082f399..1ed70f7dd4e 100644 --- a/src/Storages/MergeTree/MergeTreeData.cpp +++ b/src/Storages/MergeTree/MergeTreeData.cpp @@ -1423,7 +1423,7 @@ void MergeTreeData::loadUnexpectedDataPart(UnexpectedPartLoadState & state) try { - state.part = getDataPartBuilder(part_name, single_disk_volume, part_name) + state.part = getDataPartBuilder(part_name, single_disk_volume, part_name, getReadSettings()) .withPartInfo(part_info) .withPartFormatFromDisk() .build(); @@ -1438,7 +1438,7 @@ void MergeTreeData::loadUnexpectedDataPart(UnexpectedPartLoadState & state) /// Build a fake part and mark it as broken in case of filesystem error. /// If the error impacts part directory instead of single files, /// an exception will be thrown during detach and silently ignored. - state.part = getDataPartBuilder(part_name, single_disk_volume, part_name) + state.part = getDataPartBuilder(part_name, single_disk_volume, part_name, getReadSettings()) .withPartStorageType(MergeTreeDataPartStorageType::Full) .withPartType(MergeTreeDataPartType::Wide) .build(); @@ -1472,7 +1472,7 @@ MergeTreeData::LoadPartResult MergeTreeData::loadDataPart( /// Build a fake part and mark it as broken in case of filesystem error. /// If the error impacts part directory instead of single files, /// an exception will be thrown during detach and silently ignored. - res.part = getDataPartBuilder(part_name, single_disk_volume, part_name) + res.part = getDataPartBuilder(part_name, single_disk_volume, part_name, getReadSettings()) .withPartStorageType(MergeTreeDataPartStorageType::Full) .withPartType(MergeTreeDataPartType::Wide) .build(); @@ -1493,7 +1493,7 @@ MergeTreeData::LoadPartResult MergeTreeData::loadDataPart( try { - res.part = getDataPartBuilder(part_name, single_disk_volume, part_name) + res.part = getDataPartBuilder(part_name, single_disk_volume, part_name, getReadSettings()) .withPartInfo(part_info) .withPartFormatFromDisk() .build(); @@ -3732,9 +3732,9 @@ MergeTreeDataPartFormat MergeTreeData::choosePartFormatOnDisk(size_t bytes_uncom } MergeTreeDataPartBuilder MergeTreeData::getDataPartBuilder( - const String & name, const VolumePtr & volume, const String & part_dir) const + const String & name, const VolumePtr & volume, const String & part_dir, const ReadSettings & read_settings_) const { - return MergeTreeDataPartBuilder(*this, name, volume, relative_data_path, part_dir); + return MergeTreeDataPartBuilder(*this, name, volume, relative_data_path, part_dir, read_settings_); } void MergeTreeData::changeSettings( @@ -5812,7 +5812,7 @@ MergeTreeData::MutableDataPartPtr MergeTreeData::loadPartRestoredFromBackup(cons /// Load this part from the directory `temp_part_dir`. auto load_part = [&] { - MergeTreeDataPartBuilder builder(*this, part_name, single_disk_volume, parent_part_dir, part_dir_name); + MergeTreeDataPartBuilder builder(*this, part_name, single_disk_volume, parent_part_dir, part_dir_name, getReadSettings()); builder.withPartFormatFromDisk(); part = std::move(builder).build(); part->version.setCreationTID(Tx::PrehistoricTID, nullptr); @@ -5827,7 +5827,7 @@ MergeTreeData::MutableDataPartPtr MergeTreeData::loadPartRestoredFromBackup(cons if (!part) { /// Make a fake data part only to copy its files to /detached/. - part = MergeTreeDataPartBuilder{*this, part_name, single_disk_volume, parent_part_dir, part_dir_name} + part = MergeTreeDataPartBuilder{*this, part_name, single_disk_volume, parent_part_dir, part_dir_name, getReadSettings()} .withPartStorageType(MergeTreeDataPartStorageType::Full) .withPartType(MergeTreeDataPartType::Wide) .build(); @@ -6473,7 +6473,7 @@ MergeTreeData::MutableDataPartsVector MergeTreeData::tryLoadPartsToAttach(const LOG_DEBUG(log, "Checking part {}", new_name); auto single_disk_volume = std::make_shared("volume_" + old_name, disk); - auto part = getDataPartBuilder(old_name, single_disk_volume, source_dir / new_name) + auto part = getDataPartBuilder(old_name, single_disk_volume, source_dir / new_name, getReadSettings()) .withPartFormatFromDisk() .build(); @@ -7528,7 +7528,7 @@ std::pair MergeTreeData::cloneAn std::string(fs::path(dst_part_storage->getFullRootPath()) / tmp_dst_part_name), with_copy); - auto dst_data_part = MergeTreeDataPartBuilder(*this, dst_part_name, dst_part_storage) + auto dst_data_part = MergeTreeDataPartBuilder(*this, dst_part_name, dst_part_storage, getReadSettings()) .withPartFormatFromDisk() .build(); @@ -8786,7 +8786,7 @@ std::pair MergeTreeData::createE VolumePtr data_part_volume = createVolumeFromReservation(reservation, volume); auto tmp_dir_holder = getTemporaryPartDirectoryHolder(EMPTY_PART_TMP_PREFIX + new_part_name); - auto new_data_part = getDataPartBuilder(new_part_name, data_part_volume, EMPTY_PART_TMP_PREFIX + new_part_name) + auto new_data_part = getDataPartBuilder(new_part_name, data_part_volume, EMPTY_PART_TMP_PREFIX + new_part_name, getReadSettings()) .withBytesAndRowsOnDisk(0, 0) .withPartInfo(new_part_info) .build(); diff --git a/src/Storages/MergeTree/MergeTreeData.h b/src/Storages/MergeTree/MergeTreeData.h index 7a9730e8627..8438ac412c9 100644 --- a/src/Storages/MergeTree/MergeTreeData.h +++ b/src/Storages/MergeTree/MergeTreeData.h @@ -241,7 +241,7 @@ public: MergeTreeDataPartFormat choosePartFormat(size_t bytes_uncompressed, size_t rows_count) const; MergeTreeDataPartFormat choosePartFormatOnDisk(size_t bytes_uncompressed, size_t rows_count) const; - MergeTreeDataPartBuilder getDataPartBuilder(const String & name, const VolumePtr & volume, const String & part_dir) const; + MergeTreeDataPartBuilder getDataPartBuilder(const String & name, const VolumePtr & volume, const String & part_dir, const ReadSettings & read_settings_) const; /// Auxiliary object to add a set of parts into the working set in two steps: /// * First, as PreActive parts (the parts are ready, but not yet in the active set). diff --git a/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp b/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp index 37f578b0c25..6ec4bc31d90 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp @@ -14,20 +14,22 @@ namespace ErrorCodes } MergeTreeDataPartBuilder::MergeTreeDataPartBuilder( - const MergeTreeData & data_, String name_, VolumePtr volume_, String root_path_, String part_dir_) + const MergeTreeData & data_, String name_, VolumePtr volume_, String root_path_, String part_dir_, const ReadSettings & read_settings_) : data(data_) , name(std::move(name_)) , volume(std::move(volume_)) , root_path(std::move(root_path_)) , part_dir(std::move(part_dir_)) + , read_settings(read_settings_) { } MergeTreeDataPartBuilder::MergeTreeDataPartBuilder( - const MergeTreeData & data_, String name_, MutableDataPartStoragePtr part_storage_) + const MergeTreeData & data_, String name_, MutableDataPartStoragePtr part_storage_, const ReadSettings & read_settings_) : data(data_) , name(std::move(name_)) , part_storage(std::move(part_storage_)) + , read_settings(read_settings_) { } @@ -73,7 +75,8 @@ MutableDataPartStoragePtr MergeTreeDataPartBuilder::getPartStorageByType( MergeTreeDataPartStorageType storage_type_, const VolumePtr & volume_, const String & root_path_, - const String & part_dir_) + const String & part_dir_, + const ReadSettings &) /// Unused here, but used in private repo. { if (!volume_) throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot create part storage, because volume is not specified"); @@ -112,7 +115,7 @@ MergeTreeDataPartBuilder & MergeTreeDataPartBuilder::withPartType(MergeTreeDataP MergeTreeDataPartBuilder & MergeTreeDataPartBuilder::withPartStorageType(MergeTreeDataPartStorageType storage_type_) { - part_storage = getPartStorageByType(storage_type_, volume, root_path, part_dir); + part_storage = getPartStorageByType(storage_type_, volume, root_path, part_dir, read_settings); return *this; } @@ -126,7 +129,8 @@ MergeTreeDataPartBuilder::PartStorageAndMarkType MergeTreeDataPartBuilder::getPartStorageAndMarkType( const VolumePtr & volume_, const String & root_path_, - const String & part_dir_) + const String & part_dir_, + const ReadSettings & read_settings_) { auto disk = volume_->getDisk(); auto part_relative_path = fs::path(root_path_) / part_dir_; @@ -138,7 +142,7 @@ MergeTreeDataPartBuilder::getPartStorageAndMarkType( if (MarkType::isMarkFileExtension(ext)) { - auto storage = getPartStorageByType(MergeTreeDataPartStorageType::Full, volume_, root_path_, part_dir_); + auto storage = getPartStorageByType(MergeTreeDataPartStorageType::Full, volume_, root_path_, part_dir_, read_settings_); return {std::move(storage), MarkType(ext)}; } } @@ -156,7 +160,7 @@ MergeTreeDataPartBuilder & MergeTreeDataPartBuilder::withPartFormatFromDisk() MergeTreeDataPartBuilder & MergeTreeDataPartBuilder::withPartFormatFromVolume() { assert(volume); - auto [storage, mark_type] = getPartStorageAndMarkType(volume, root_path, part_dir); + auto [storage, mark_type] = getPartStorageAndMarkType(volume, root_path, part_dir, read_settings); if (!storage || !mark_type) { diff --git a/src/Storages/MergeTree/MergeTreeDataPartBuilder.h b/src/Storages/MergeTree/MergeTreeDataPartBuilder.h index 0f54ff0a631..bce881a1970 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartBuilder.h +++ b/src/Storages/MergeTree/MergeTreeDataPartBuilder.h @@ -21,8 +21,8 @@ using VolumePtr = std::shared_ptr; class MergeTreeDataPartBuilder { public: - MergeTreeDataPartBuilder(const MergeTreeData & data_, String name_, VolumePtr volume_, String root_path_, String part_dir_); - MergeTreeDataPartBuilder(const MergeTreeData & data_, String name_, MutableDataPartStoragePtr part_storage_); + MergeTreeDataPartBuilder(const MergeTreeData & data_, String name_, VolumePtr volume_, String root_path_, String part_dir_, const ReadSettings & read_settings_); + MergeTreeDataPartBuilder(const MergeTreeData & data_, String name_, MutableDataPartStoragePtr part_storage_, const ReadSettings & read_settings_); std::shared_ptr build(); @@ -42,7 +42,8 @@ public: static PartStorageAndMarkType getPartStorageAndMarkType( const VolumePtr & volume_, const String & root_path_, - const String & part_dir_); + const String & part_dir_, + const ReadSettings & read_settings); private: Self & withPartFormatFromVolume(); @@ -52,7 +53,8 @@ private: MergeTreeDataPartStorageType storage_type_, const VolumePtr & volume_, const String & root_path_, - const String & part_dir_); + const String & part_dir_, + const ReadSettings & read_settings); const MergeTreeData & data; const String name; @@ -64,6 +66,8 @@ private: std::optional part_type; MutableDataPartStoragePtr part_storage; const IMergeTreeDataPart * parent_part = nullptr; + + const ReadSettings read_settings; }; } diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.cpp b/src/Storages/MergeTree/MergeTreeDataWriter.cpp index 67fef759ed4..12dbd529f70 100644 --- a/src/Storages/MergeTree/MergeTreeDataWriter.cpp +++ b/src/Storages/MergeTree/MergeTreeDataWriter.cpp @@ -609,7 +609,7 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl( } } - auto new_data_part = data.getDataPartBuilder(part_name, data_part_volume, part_dir) + auto new_data_part = data.getDataPartBuilder(part_name, data_part_volume, part_dir, getReadSettings()) .withPartFormat(data.choosePartFormat(expected_size, block.rows())) .withPartInfo(new_part_info) .build(); diff --git a/src/Storages/MergeTree/MergeTreePartsMover.cpp b/src/Storages/MergeTree/MergeTreePartsMover.cpp index 48a4a37f444..e9c9f2b4b06 100644 --- a/src/Storages/MergeTree/MergeTreePartsMover.cpp +++ b/src/Storages/MergeTree/MergeTreePartsMover.cpp @@ -280,7 +280,7 @@ MergeTreePartsMover::TemporaryClonedPart MergeTreePartsMover::clonePart(const Me cloned_part_storage = part->makeCloneOnDisk(disk, MergeTreeData::MOVING_DIR_NAME, read_settings, write_settings, cancellation_hook); } - MergeTreeDataPartBuilder builder(*data, part->name, cloned_part_storage); + MergeTreeDataPartBuilder builder(*data, part->name, cloned_part_storage, getReadSettings()); cloned_part.part = std::move(builder).withPartFormatFromDisk().build(); LOG_TRACE(log, "Part {} was cloned to {}", part->name, cloned_part.part->getDataPartStorage().getFullPath()); diff --git a/src/Storages/MergeTree/MutateTask.cpp b/src/Storages/MergeTree/MutateTask.cpp index 2e7847fc99f..92e0193fff9 100644 --- a/src/Storages/MergeTree/MutateTask.cpp +++ b/src/Storages/MergeTree/MutateTask.cpp @@ -2286,7 +2286,7 @@ bool MutateTask::prepare() String tmp_part_dir_name = prefix + ctx->future_part->name; ctx->temporary_directory_lock = ctx->data->getTemporaryPartDirectoryHolder(tmp_part_dir_name); - auto builder = ctx->data->getDataPartBuilder(ctx->future_part->name, single_disk_volume, tmp_part_dir_name); + auto builder = ctx->data->getDataPartBuilder(ctx->future_part->name, single_disk_volume, tmp_part_dir_name, getReadSettings()); builder.withPartFormat(ctx->future_part->part_format); builder.withPartInfo(ctx->future_part->part_info); diff --git a/src/Storages/StorageReplicatedMergeTree.cpp b/src/Storages/StorageReplicatedMergeTree.cpp index b5b07a129bd..e5b40c07f69 100644 --- a/src/Storages/StorageReplicatedMergeTree.cpp +++ b/src/Storages/StorageReplicatedMergeTree.cpp @@ -2092,7 +2092,7 @@ MergeTreeData::MutableDataPartPtr StorageReplicatedMergeTree::attachPartHelperFo const auto part_old_name = part_info->getPartNameV1(); const auto volume = std::make_shared("volume_" + part_old_name, disk); - auto part = getDataPartBuilder(entry.new_part_name, volume, fs::path(DETACHED_DIR_NAME) / part_old_name) + auto part = getDataPartBuilder(entry.new_part_name, volume, fs::path(DETACHED_DIR_NAME) / part_old_name, getReadSettings()) .withPartFormatFromDisk() .build(); From 1fd66d0472d90bc6da1d0f04dce8140b83fd6bb7 Mon Sep 17 00:00:00 2001 From: Pavel Kruglov <48961922+Avogar@users.noreply.github.com> Date: Thu, 31 Oct 2024 14:58:27 +0100 Subject: [PATCH 358/680] Update SerializationObject.cpp --- src/DataTypes/Serializations/SerializationObject.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/DataTypes/Serializations/SerializationObject.cpp b/src/DataTypes/Serializations/SerializationObject.cpp index cf63797b0c2..19e12d777e4 100644 --- a/src/DataTypes/Serializations/SerializationObject.cpp +++ b/src/DataTypes/Serializations/SerializationObject.cpp @@ -365,7 +365,7 @@ ISerialization::DeserializeBinaryBulkStatePtr SerializationObject::deserializeOb auto structure_state = std::make_shared(serialization_version); if (structure_state->serialization_version.value == ObjectSerializationVersion::Value::V1 || structure_state->serialization_version.value == ObjectSerializationVersion::Value::V2) { - if (structure_state->structure_version.value == ObjectSerializationVersion::Value::V1) + if (structure_state->serialization_version.value == ObjectSerializationVersion::Value::V1) { /// Skip max_dynamic_paths parameter in V1 serialization version. size_t max_dynamic_paths; From 936d6b22518e7711adc4991663f6474b42805eb8 Mon Sep 17 00:00:00 2001 From: MikhailBurdukov Date: Thu, 31 Oct 2024 14:05:33 +0000 Subject: [PATCH 359/680] Fix unescaping in named collections --- .../NamedCollectionsMetadataStorage.cpp | 2 +- tests/integration/test_named_collections/test.py | 14 ++++++++++++++ 2 files changed, 15 insertions(+), 1 deletion(-) diff --git a/src/Common/NamedCollections/NamedCollectionsMetadataStorage.cpp b/src/Common/NamedCollections/NamedCollectionsMetadataStorage.cpp index b8413bfadd7..8bb411f1437 100644 --- a/src/Common/NamedCollections/NamedCollectionsMetadataStorage.cpp +++ b/src/Common/NamedCollections/NamedCollectionsMetadataStorage.cpp @@ -568,7 +568,7 @@ std::vector NamedCollectionsMetadataStorage::listCollections() cons std::vector collections; collections.reserve(paths.size()); for (const auto & path : paths) - collections.push_back(std::filesystem::path(path).stem()); + collections.push_back(unescapeForFileName(std::filesystem::path(path).stem())); return collections; } diff --git a/tests/integration/test_named_collections/test.py b/tests/integration/test_named_collections/test.py index ed80898ebc7..bd04bb9e3c8 100644 --- a/tests/integration/test_named_collections/test.py +++ b/tests/integration/test_named_collections/test.py @@ -794,3 +794,17 @@ def test_keeper_storage_remove_on_cluster(cluster, ignore, expected_raise): node.query( f"DROP NAMED COLLECTION test_nc ON CLUSTER `replicated_nc_nodes_cluster`" ) + + +@pytest.mark.parametrize( + "instance_name", + [("node"), ("node_with_keeper")], +) +def test_name_escaping(cluster, instance_name): + node = cluster.instances[instance_name] + + node.query("DROP NAMED COLLECTION IF EXISTS test;") + node.query("CREATE NAMED COLLECTION `test_!strange/symbols!` AS key1=1, key2=2") + node.restart_clickhouse() + + node.query("DROP NAMED COLLECTION `test_!strange/symbols!`") From fa5010ba181f7251ebcf9ce09ade01c48fdcdebc Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Thu, 31 Oct 2024 14:20:47 +0000 Subject: [PATCH 360/680] fix test --- ...der_optimization_with_virtual_row_special.sql | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql index ee7336bdf02..3d6f9ad391b 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql @@ -1,21 +1,19 @@ -- Tags: no-parallel --- modified from test_01155_ordinary -DROP DATABASE IF EXISTS test_01155_ordinary; +-- modified from test_01155_ordinary, to test special optimization path for virtual row +DROP DATABASE IF EXISTS test_03031; -SET allow_deprecated_database_ordinary = 1; +CREATE DATABASE test_03031; -CREATE DATABASE test_01155_ordinary ENGINE = Ordinary; - -USE test_01155_ordinary; +USE test_03031; CREATE TABLE src (s String) ENGINE = MergeTree() ORDER BY s; INSERT INTO src(s) VALUES ('before moving tables'); -CREATE TABLE dist (s String) ENGINE = Distributed(test_shard_localhost, test_01155_ordinary, src); +CREATE TABLE dist (s String) ENGINE = Distributed(test_shard_localhost, test_03031, src); SET enable_analyzer=0; -SELECT _table FROM merge('test_01155_ordinary', '') ORDER BY _table, s; +SELECT _table FROM merge('test_03031', '') ORDER BY _table, s; DROP TABLE src; DROP TABLE dist; -DROP DATABASE test_01155_ordinary; \ No newline at end of file +DROP DATABASE test_03031; \ No newline at end of file From 542dac1815858e55147a5db80e58690bb8b72df2 Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 28 Oct 2024 10:31:50 +0000 Subject: [PATCH 361/680] Implement simple CAST from Map/Tuple/Object to new JSON through serialization/deserialization from JSON string --- src/DataTypes/DataTypeObject.cpp | 10 +++++ src/DataTypes/DataTypeObject.h | 3 ++ .../Serializations/SerializationObject.cpp | 11 +---- .../Serializations/SerializationObject.h | 3 -- .../SerializationObjectDynamicPath.cpp | 8 ++-- .../Serializations/SerializationSubObject.cpp | 8 ++-- src/Functions/FunctionsConversion.cpp | 42 ++++++++++++++----- ...61_tuple_map_object_to_json_cast.reference | 23 ++++++++++ .../03261_tuple_map_object_to_json_cast.sql | 14 +++++++ 9 files changed, 91 insertions(+), 31 deletions(-) create mode 100644 tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.reference create mode 100644 tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql diff --git a/src/DataTypes/DataTypeObject.cpp b/src/DataTypes/DataTypeObject.cpp index 18bfed9c5c3..d744e851ea9 100644 --- a/src/DataTypes/DataTypeObject.cpp +++ b/src/DataTypes/DataTypeObject.cpp @@ -1,6 +1,9 @@ #include #include #include +#include +#include +#include #include #include #include @@ -522,6 +525,13 @@ static DataTypePtr createObject(const ASTPtr & arguments, const DataTypeObject:: return std::make_shared(schema_format, std::move(typed_paths), std::move(paths_to_skip), std::move(path_regexps_to_skip), max_dynamic_paths, max_dynamic_types); } +const DataTypePtr & DataTypeObject::getTypeOfSharedData() +{ + /// Array(Tuple(String, String)) + static const DataTypePtr type = std::make_shared(std::make_shared(DataTypes{std::make_shared(), std::make_shared()}, Names{"paths", "values"})); + return type; +} + static DataTypePtr createJSON(const ASTPtr & arguments) { auto context = CurrentThread::getQueryContext(); diff --git a/src/DataTypes/DataTypeObject.h b/src/DataTypes/DataTypeObject.h index 7eb2e7729de..32ed6a7ee86 100644 --- a/src/DataTypes/DataTypeObject.h +++ b/src/DataTypes/DataTypeObject.h @@ -63,6 +63,9 @@ public: size_t getMaxDynamicTypes() const { return max_dynamic_types; } size_t getMaxDynamicPaths() const { return max_dynamic_paths; } + /// Shared data has type Array(Tuple(String, String)). + static const DataTypePtr & getTypeOfSharedData(); + private: SchemaFormat schema_format; /// Set of paths with types that were specified in type declaration. diff --git a/src/DataTypes/Serializations/SerializationObject.cpp b/src/DataTypes/Serializations/SerializationObject.cpp index 0fbf8c54a22..3e1badb25ca 100644 --- a/src/DataTypes/Serializations/SerializationObject.cpp +++ b/src/DataTypes/Serializations/SerializationObject.cpp @@ -25,7 +25,7 @@ SerializationObject::SerializationObject( : typed_path_serializations(std::move(typed_path_serializations_)) , paths_to_skip(paths_to_skip_) , dynamic_serialization(std::make_shared()) - , shared_data_serialization(getTypeOfSharedData()->getDefaultSerialization()) + , shared_data_serialization(DataTypeObject::getTypeOfSharedData()->getDefaultSerialization()) { /// We will need sorted order of typed paths to serialize them in order for consistency. sorted_typed_paths.reserve(typed_path_serializations.size()); @@ -38,13 +38,6 @@ SerializationObject::SerializationObject( path_regexps_to_skip.emplace_back(regexp_str); } -const DataTypePtr & SerializationObject::getTypeOfSharedData() -{ - /// Array(Tuple(String, String)) - static const DataTypePtr type = std::make_shared(std::make_shared(DataTypes{std::make_shared(), std::make_shared()}, Names{"paths", "values"})); - return type; -} - bool SerializationObject::shouldSkipPath(const String & path) const { if (paths_to_skip.contains(path)) @@ -168,7 +161,7 @@ void SerializationObject::enumerateStreams(EnumerateStreamsSettings & settings, settings.path.push_back(Substream::ObjectSharedData); auto shared_data_substream_data = SubstreamData(shared_data_serialization) - .withType(getTypeOfSharedData()) + .withType(DataTypeObject::getTypeOfSharedData()) .withColumn(column_object ? column_object->getSharedDataPtr() : nullptr) .withSerializationInfo(data.serialization_info) .withDeserializeState(deserialize_state ? deserialize_state->shared_data_state : nullptr); diff --git a/src/DataTypes/Serializations/SerializationObject.h b/src/DataTypes/Serializations/SerializationObject.h index 420293ba428..8bc72312da1 100644 --- a/src/DataTypes/Serializations/SerializationObject.h +++ b/src/DataTypes/Serializations/SerializationObject.h @@ -111,9 +111,6 @@ private: DeserializeBinaryBulkSettings & settings, SubstreamsDeserializeStatesCache * cache); - /// Shared data has type Array(Tuple(String, String)). - static const DataTypePtr & getTypeOfSharedData(); - struct TypedPathSubcolumnCreator : public ISubcolumnCreator { String path; diff --git a/src/DataTypes/Serializations/SerializationObjectDynamicPath.cpp b/src/DataTypes/Serializations/SerializationObjectDynamicPath.cpp index 5323079c54b..c1f26eca792 100644 --- a/src/DataTypes/Serializations/SerializationObjectDynamicPath.cpp +++ b/src/DataTypes/Serializations/SerializationObjectDynamicPath.cpp @@ -18,7 +18,7 @@ SerializationObjectDynamicPath::SerializationObjectDynamicPath( , path(path_) , path_subcolumn(path_subcolumn_) , dynamic_serialization(std::make_shared()) - , shared_data_serialization(SerializationObject::getTypeOfSharedData()->getDefaultSerialization()) + , shared_data_serialization(DataTypeObject::getTypeOfSharedData()->getDefaultSerialization()) , max_dynamic_types(max_dynamic_types_) { } @@ -67,8 +67,8 @@ void SerializationObjectDynamicPath::enumerateStreams( { settings.path.push_back(Substream::ObjectSharedData); auto shared_data_substream_data = SubstreamData(shared_data_serialization) - .withType(data.type ? SerializationObject::getTypeOfSharedData() : nullptr) - .withColumn(data.column ? SerializationObject::getTypeOfSharedData()->createColumn() : nullptr) + .withType(data.type ? DataTypeObject::getTypeOfSharedData() : nullptr) + .withColumn(data.column ? DataTypeObject::getTypeOfSharedData()->createColumn() : nullptr) .withSerializationInfo(data.serialization_info) .withDeserializeState(deserialize_state->nested_state); settings.path.back().data = shared_data_substream_data; @@ -164,7 +164,7 @@ void SerializationObjectDynamicPath::deserializeBinaryBulkWithMultipleStreams( settings.path.push_back(Substream::ObjectSharedData); /// Initialize shared_data column if needed. if (result_column->empty()) - dynamic_path_state->shared_data = SerializationObject::getTypeOfSharedData()->createColumn(); + dynamic_path_state->shared_data = DataTypeObject::getTypeOfSharedData()->createColumn(); size_t prev_size = result_column->size(); shared_data_serialization->deserializeBinaryBulkWithMultipleStreams(dynamic_path_state->shared_data, limit, settings, dynamic_path_state->nested_state, cache); /// If we need to read a subcolumn from Dynamic column, create an empty Dynamic column, fill it and extract subcolumn. diff --git a/src/DataTypes/Serializations/SerializationSubObject.cpp b/src/DataTypes/Serializations/SerializationSubObject.cpp index 9084d46f9b2..ff61cb55572 100644 --- a/src/DataTypes/Serializations/SerializationSubObject.cpp +++ b/src/DataTypes/Serializations/SerializationSubObject.cpp @@ -17,7 +17,7 @@ SerializationSubObject::SerializationSubObject( : path_prefix(path_prefix_) , typed_paths_serializations(typed_paths_serializations_) , dynamic_serialization(std::make_shared()) - , shared_data_serialization(SerializationObject::getTypeOfSharedData()->getDefaultSerialization()) + , shared_data_serialization(DataTypeObject::getTypeOfSharedData()->getDefaultSerialization()) { } @@ -64,8 +64,8 @@ void SerializationSubObject::enumerateStreams( /// We will need to read shared data to find all paths with requested prefix. settings.path.push_back(Substream::ObjectSharedData); auto shared_data_substream_data = SubstreamData(shared_data_serialization) - .withType(data.type ? SerializationObject::getTypeOfSharedData() : nullptr) - .withColumn(data.column ? SerializationObject::getTypeOfSharedData()->createColumn() : nullptr) + .withType(data.type ? DataTypeObject::getTypeOfSharedData() : nullptr) + .withColumn(data.column ? DataTypeObject::getTypeOfSharedData()->createColumn() : nullptr) .withSerializationInfo(data.serialization_info) .withDeserializeState(deserialize_state ? deserialize_state->shared_data_state : nullptr); settings.path.back().data = shared_data_substream_data; @@ -208,7 +208,7 @@ void SerializationSubObject::deserializeBinaryBulkWithMultipleStreams( settings.path.push_back(Substream::ObjectSharedData); /// If it's a new object column, reinitialize column for shared data. if (result_column->empty()) - sub_object_state->shared_data = SerializationObject::getTypeOfSharedData()->createColumn(); + sub_object_state->shared_data = DataTypeObject::getTypeOfSharedData()->createColumn(); size_t prev_size = column_object.size(); shared_data_serialization->deserializeBinaryBulkWithMultipleStreams(sub_object_state->shared_data, limit, settings, sub_object_state->shared_data_state, cache); settings.path.pop_back(); diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp index 0f6311c9716..ee04916e7b4 100644 --- a/src/Functions/FunctionsConversion.cpp +++ b/src/Functions/FunctionsConversion.cpp @@ -3921,7 +3921,7 @@ private: } } - WrapperType createTupleToObjectWrapper(const DataTypeTuple & from_tuple, bool has_nullable_subcolumns) const + WrapperType createTupleToObjectDeprecatedWrapper(const DataTypeTuple & from_tuple, bool has_nullable_subcolumns) const { if (!from_tuple.haveExplicitNames()) throw Exception(ErrorCodes::TYPE_MISMATCH, @@ -3968,7 +3968,7 @@ private: }; } - WrapperType createMapToObjectWrapper(const DataTypeMap & from_map, bool has_nullable_subcolumns) const + WrapperType createMapToObjectDeprecatedWrapper(const DataTypeMap & from_map, bool has_nullable_subcolumns) const { auto key_value_types = from_map.getKeyValueTypes(); @@ -4048,11 +4048,11 @@ private: { if (const auto * from_tuple = checkAndGetDataType(from_type.get())) { - return createTupleToObjectWrapper(*from_tuple, to_type->hasNullableSubcolumns()); + return createTupleToObjectDeprecatedWrapper(*from_tuple, to_type->hasNullableSubcolumns()); } else if (const auto * from_map = checkAndGetDataType(from_type.get())) { - return createMapToObjectWrapper(*from_map, to_type->hasNullableSubcolumns()); + return createMapToObjectDeprecatedWrapper(*from_map, to_type->hasNullableSubcolumns()); } else if (checkAndGetDataType(from_type.get())) { @@ -4081,23 +4081,43 @@ private: "Cast to Object can be performed only from flatten named Tuple, Map or String. Got: {}", from_type->getName()); } + WrapperType createObjectWrapper(const DataTypePtr & from_type, const DataTypeObject * to_object) const { if (checkAndGetDataType(from_type.get())) { return [this](ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * nullable_source, size_t input_rows_count) { - auto res = ConvertImplGenericFromString::execute(arguments, result_type, nullable_source, input_rows_count, context)->assumeMutable(); - res->finalize(); - return res; + return ConvertImplGenericFromString::execute(arguments, result_type, nullable_source, input_rows_count, context); + }; + } + + /// Cast Tuple/Object/Map to JSON type through serializing into JSON string and parsing back into JSON column. + /// Potentially we can do smarter conversion Tuple -> JSON with type preservation, but it's questionable how exactly Tuple should be + /// converted to JSON (for example, should we recursively convert nested Array(Tuple) to Array(JSON) or not, should we infer types from String fields, etc). + if (checkAndGetDataType(from_type.get()) || checkAndGetDataType(from_type.get()) || checkAndGetDataType(from_type.get())) + { + return [this](ColumnsWithTypeAndName & arguments, const DataTypePtr & result_type, const ColumnNullable * nullable_source, size_t input_rows_count) + { + auto json_string = ColumnString::create(); + ColumnStringHelpers::WriteHelper write_helper(assert_cast(*json_string), input_rows_count); + auto & write_buffer = write_helper.getWriteBuffer(); + FormatSettings format_settings = context ? getFormatSettings(context) : FormatSettings{}; + auto serialization = arguments[0].type->getDefaultSerialization(); + for (size_t i = 0; i < input_rows_count; ++i) + { + serialization->serializeTextJSON(*arguments[0].column, i, write_buffer, format_settings); + write_helper.rowWritten(); + } + write_helper.finalize(); + + ColumnsWithTypeAndName args_with_json_string = {ColumnWithTypeAndName(json_string->getPtr(), std::make_shared(), "")}; + return ConvertImplGenericFromString::execute(args_with_json_string, result_type, nullable_source, input_rows_count, context); }; } /// TODO: support CAST between JSON types with different parameters - /// support CAST from Map to JSON - /// support CAST from Tuple to JSON - /// support CAST from Object('json') to JSON - throw Exception(ErrorCodes::TYPE_MISMATCH, "Cast to {} can be performed only from String. Got: {}", magic_enum::enum_name(to_object->getSchemaFormat()), from_type->getName()); + throw Exception(ErrorCodes::TYPE_MISMATCH, "Cast to {} can be performed only from String/Map/Object/Tuple. Got: {}", magic_enum::enum_name(to_object->getSchemaFormat()), from_type->getName()); } WrapperType createVariantToVariantWrapper(const DataTypeVariant & from_variant, const DataTypeVariant & to_variant) const diff --git a/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.reference b/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.reference new file mode 100644 index 00000000000..0ae94e68663 --- /dev/null +++ b/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.reference @@ -0,0 +1,23 @@ +Map to JSON +{"a":"0","b":"1970-01-01","c":[],"d":[{"e":"0"}]} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a":"1","b":"1970-01-02","c":["0"],"d":[{"e":"1"}]} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a":"2","b":"1970-01-03","c":["0","1"],"d":[{"e":"2"}]} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a":"3","b":"1970-01-04","c":["0","1","2"],"d":[{"e":"3"}]} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a":"4","b":"1970-01-05","c":["0","1","2","3"],"d":[{"e":"4"}]} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a0":"0","b0":"1970-01-01","c0":[],"d0":[{"e0":"0"}]} {'a0':'Int64','b0':'Date','c0':'Array(Nullable(String))','d0':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a1":"1","b1":"1970-01-02","c1":["0"],"d1":[{"e1":"1"}]} {'a1':'Int64','b1':'Date','c1':'Array(Nullable(String))','d1':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a2":"2","b2":"1970-01-03","c2":["0","1"],"d2":[{"e2":"2"}]} {'a2':'Int64','b2':'Date','c2':'Array(Nullable(String))','d2':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a0":"3","b0":"1970-01-04","c0":["0","1","2"],"d0":[{"e0":"3"}]} {'a0':'Int64','b0':'Date','c0':'Array(Nullable(String))','d0':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a1":"4","b1":"1970-01-05","c1":["0","1","2","3"],"d1":[{"e1":"4"}]} {'a1':'Int64','b1':'Date','c1':'Array(Nullable(String))','d1':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +Tuple to JSON +{"a":"0","b":"1970-01-01","c":[],"d":[{"e":"0"}]} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a":"1","b":"1970-01-02","c":["0"],"d":[{"e":"1"}]} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a":"2","b":"1970-01-03","c":["0","1"],"d":[{"e":"2"}]} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a":"3","b":"1970-01-04","c":["0","1","2"],"d":[{"e":"3"}]} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +{"a":"4","b":"1970-01-05","c":["0","1","2","3"],"d":[{"e":"4"}]} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d':'Array(JSON(max_dynamic_types=16, max_dynamic_paths=256))'} +Object to JSON +{"a":"0","b":"1970-01-01","c":[],"d":{"e":["0"]}} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d.e':'Array(Nullable(Int64))'} +{"a":"1","b":"1970-01-02","c":["0"],"d":{"e":["1"]}} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d.e':'Array(Nullable(Int64))'} +{"a":"2","b":"1970-01-03","c":["0","1"],"d":{"e":["2"]}} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d.e':'Array(Nullable(Int64))'} +{"a":"3","b":"1970-01-04","c":["0","1","2"],"d":{"e":["3"]}} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d.e':'Array(Nullable(Int64))'} +{"a":"4","b":"1970-01-05","c":["0","1","2","3"],"d":{"e":["4"]}} {'a':'Int64','b':'Date','c':'Array(Nullable(String))','d.e':'Array(Nullable(Int64))'} diff --git a/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql b/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql new file mode 100644 index 00000000000..fcec7eb3af4 --- /dev/null +++ b/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql @@ -0,0 +1,14 @@ +set allow_experimental_json_type = 1; +set allow_experimental_object_type = 1; +set allow_experimental_variant_type = 1; +set use_variant_as_common_type = 1; + +select 'Map to JSON'; +select map('a', number::UInt32, 'b', toDate(number), 'c', range(number), 'd', [map('e', number::UInt32)])::JSON as json, JSONAllPathsWithTypes(json) from numbers(5); +select map('a' || number % 3, number::UInt32, 'b' || number % 3, toDate(number), 'c' || number % 3, range(number), 'd' || number % 3, [map('e' || number % 3, number::UInt32)])::JSON as json, JSONAllPathsWithTypes(json) from numbers(5); + +select 'Tuple to JSON'; +select tuple(number::UInt32 as a, toDate(number) as b, range(number) as c, [tuple(number::UInt32 as e)] as d)::JSON as json, JSONAllPathsWithTypes(json) from numbers(5); + +select 'Object to JSON'; +select toJSONString(map('a', number::UInt32, 'b', toDate(number), 'c', range(number), 'd', [map('e', number::UInt32)]))::Object('json')::JSON as json, JSONAllPathsWithTypes(json) from numbers(5); From 83f434dffb6bad82abdc791179196b32e1a7f347 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Thu, 31 Oct 2024 16:25:17 +0000 Subject: [PATCH 362/680] fix simple path --- src/Processors/Transforms/FillingTransform.cpp | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index 4a8965dcfaa..dd116a9972a 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -608,9 +608,6 @@ void FillingTransform::transformRange( const auto current_value = (*input_fill_columns[i])[range_begin]; const auto & fill_from = filling_row.getFillDescription(i).fill_from; - logDebug("current value", current_value.dump()); - logDebug("fill from", fill_from.dump()); - if (!fill_from.isNull() && !equals(current_value, fill_from)) { filling_row.initUsingFrom(i); @@ -663,6 +660,7 @@ void FillingTransform::transformRange( interpolate(result_columns, interpolate_block); insertFromFillingRow(res_fill_columns, res_interpolate_columns, res_other_columns, interpolate_block); copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); + filling_row_changed = false; } /// Initialize staleness border for current row to generate it's prefix @@ -679,6 +677,7 @@ void FillingTransform::transformRange( interpolate(result_columns, interpolate_block); insertFromFillingRow(res_fill_columns, res_interpolate_columns, res_other_columns, interpolate_block); copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); + filling_row_changed = false; } while (filling_row.next(next_row, filling_row_changed)); } From 390429dee53f0c758d823166f9f09024dbed07ae Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Thu, 31 Oct 2024 17:34:17 +0100 Subject: [PATCH 363/680] Fix build --- src/Storages/ObjectStorage/StorageObjectStorageSource.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp index a1737c55c26..563bdc44760 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp @@ -522,7 +522,7 @@ std::unique_ptr StorageObjectStorageSource::createReadBu ? std::max(read_settings.remote_fs_buffer_size, DBMS_DEFAULT_BUFFER_SIZE) : read_settings.remote_fs_buffer_size; if (object_size) - buffer_size = std::min(object_size, buffer_size); + buffer_size = std::min(object_size, buffer_size); auto & reader = context_->getThreadPoolReader(FilesystemReaderType::ASYNCHRONOUS_REMOTE_FS_READER); impl = std::make_unique( From 1000ef0e022516536cbd680fa6a206bf5401295c Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Thu, 31 Oct 2024 16:39:31 +0000 Subject: [PATCH 364/680] some improves --- src/Interpreters/FillingRow.cpp | 20 ++++++++----- .../Transforms/FillingTransform.cpp | 30 +++++++++++-------- src/Processors/Transforms/FillingTransform.h | 1 + 3 files changed, 31 insertions(+), 20 deletions(-) diff --git a/src/Interpreters/FillingRow.cpp b/src/Interpreters/FillingRow.cpp index 98c18e9b2ae..384ad669206 100644 --- a/src/Interpreters/FillingRow.cpp +++ b/src/Interpreters/FillingRow.cpp @@ -13,7 +13,7 @@ namespace DB constexpr static bool debug_logging_enabled = false; template -inline static void logDebug(String fmt_str, Args&&... args) +inline static void logDebug(const char * fmt_str, Args&&... args) { if constexpr (debug_logging_enabled) LOG_DEBUG(getLogger("FillingRow"), "{}", fmt::format(fmt::runtime(fmt_str), std::forward(args)...)); @@ -117,7 +117,7 @@ bool FillingRow::isConstraintsSatisfied(size_t pos) const chassert(hasSomeConstraints(pos)); int direction = getDirection(pos); - logDebug("constraint: {}, row: {}, direction: {}", constraints[pos].dump(), row[pos].dump(), direction); + logDebug("constraint: {}, row: {}, direction: {}", constraints[pos], row[pos], direction); return less(row[pos], constraints[pos], direction); } @@ -230,7 +230,7 @@ bool FillingRow::next(const FillingRow & next_original_row, bool& value_changed) bool FillingRow::shift(const FillingRow & next_original_row, bool& value_changed) { - logDebug("next_original_row: {}, current: {}", next_original_row.dump(), dump()); + logDebug("next_original_row: {}, current: {}", next_original_row, *this); for (size_t pos = 0; pos < size(); ++pos) { @@ -318,15 +318,12 @@ void FillingRow::updateConstraintsWithStalenessRow(const Columns& base_row, size for (size_t i = 0; i < size(); ++i) { const auto& descr = getFillDescription(i); - constraints[i] = descr.fill_to; if (!descr.fill_staleness.isNull()) { Field staleness_border = (*base_row[i])[row_ind]; descr.staleness_step_func(staleness_border, 1); - - if (constraints[i].isNull() || less(staleness_border, constraints[i], getDirection(i))) - constraints[i] = std::move(staleness_border); + constraints[i] = findBorder(descr.fill_to, staleness_border, getDirection(i)); } } } @@ -350,3 +347,12 @@ WriteBuffer & operator<<(WriteBuffer & out, const FillingRow & row) } } + +template <> +struct fmt::formatter : fmt::formatter +{ + constexpr auto format(const DB::FillingRow & row, format_context & ctx) const + { + return fmt::format_to(ctx.out(), "{}", row.dump()); + } +}; diff --git a/src/Processors/Transforms/FillingTransform.cpp b/src/Processors/Transforms/FillingTransform.cpp index dd116a9972a..ab782f3e521 100644 --- a/src/Processors/Transforms/FillingTransform.cpp +++ b/src/Processors/Transforms/FillingTransform.cpp @@ -20,7 +20,7 @@ namespace DB constexpr static bool debug_logging_enabled = false; template -inline static void logDebug(String key, const T & value, const char * separator = " : ") +inline static void logDebug(const char * key, const T & value, const char * separator = " : ") { if constexpr (debug_logging_enabled) { @@ -235,6 +235,7 @@ FillingTransform::FillingTransform( fill_column_positions.push_back(block_position); auto & descr = filling_row.getFillDescription(i); + running_with_staleness |= !descr.fill_staleness.isNull(); const Block & output_header = getOutputPort().getHeader(); const DataTypePtr & type = removeNullable(output_header.getByPosition(block_position).type); @@ -663,23 +664,26 @@ void FillingTransform::transformRange( filling_row_changed = false; } - /// Initialize staleness border for current row to generate it's prefix - filling_row.updateConstraintsWithStalenessRow(input_fill_columns, row_ind); - - while (filling_row.shift(next_row, filling_row_changed)) + if (running_with_staleness) { - logDebug("filling_row after shift", filling_row); + /// Initialize staleness border for current row to generate it's prefix + filling_row.updateConstraintsWithStalenessRow(input_fill_columns, row_ind); - do + while (filling_row.shift(next_row, filling_row_changed)) { - logDebug("inserting prefix filling_row", filling_row); + logDebug("filling_row after shift", filling_row); - interpolate(result_columns, interpolate_block); - insertFromFillingRow(res_fill_columns, res_interpolate_columns, res_other_columns, interpolate_block); - copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); - filling_row_changed = false; + do + { + logDebug("inserting prefix filling_row", filling_row); - } while (filling_row.next(next_row, filling_row_changed)); + interpolate(result_columns, interpolate_block); + insertFromFillingRow(res_fill_columns, res_interpolate_columns, res_other_columns, interpolate_block); + copyRowFromColumns(res_sort_prefix_columns, input_sort_prefix_columns, row_ind); + filling_row_changed = false; + + } while (filling_row.next(next_row, filling_row_changed)); + } } /// new valid filling row was generated but not inserted, will use it during suffix generation diff --git a/src/Processors/Transforms/FillingTransform.h b/src/Processors/Transforms/FillingTransform.h index a8866a97103..92ca4fe6c9e 100644 --- a/src/Processors/Transforms/FillingTransform.h +++ b/src/Processors/Transforms/FillingTransform.h @@ -84,6 +84,7 @@ private: SortDescription sort_prefix; const InterpolateDescriptionPtr interpolate_description; /// Contains INTERPOLATE columns + bool running_with_staleness = false; /// True if STALENESS clause was used. FillingRow filling_row; /// Current row, which is used to fill gaps. FillingRow next_row; /// Row to which we need to generate filling rows. bool filling_row_inserted = false; From 9021aeaaff66f7a0c0daeb37d1cd42157c5a15aa Mon Sep 17 00:00:00 2001 From: avogar Date: Thu, 31 Oct 2024 16:57:51 +0000 Subject: [PATCH 365/680] Add docs --- docs/en/sql-reference/data-types/newjson.md | 46 +++++++++++++++++++-- 1 file changed, 43 insertions(+), 3 deletions(-) diff --git a/docs/en/sql-reference/data-types/newjson.md b/docs/en/sql-reference/data-types/newjson.md index 68952590eb9..2f54d45cd64 100644 --- a/docs/en/sql-reference/data-types/newjson.md +++ b/docs/en/sql-reference/data-types/newjson.md @@ -58,10 +58,10 @@ SELECT json FROM test; └───────────────────────────────────┘ ``` -Using CAST from 'String': +Using CAST from `String`: ```sql -SELECT '{"a" : {"b" : 42},"c" : [1, 2, 3], "d" : "Hello, World!"}'::JSON as json; +SELECT '{"a" : {"b" : 42},"c" : [1, 2, 3], "d" : "Hello, World!"}'::JSON AS json; ``` ```text @@ -70,7 +70,47 @@ SELECT '{"a" : {"b" : 42},"c" : [1, 2, 3], "d" : "Hello, World!"}'::JSON as json └────────────────────────────────────────────────┘ ``` -CAST from `JSON`, named `Tuple`, `Map` and `Object('json')` to `JSON` type will be supported later. +Using CAST from `Tuple`: + +```sql +SELECT (tuple(42 AS b) AS a, [1, 2, 3] AS c, 'Hello, World!' AS d)::JSON AS json; +``` + +```text +┌─json───────────────────────────────────────────┐ +│ {"a":{"b":42},"c":[1,2,3],"d":"Hello, World!"} │ +└────────────────────────────────────────────────┘ +``` + +Using CAST from `Map`: + +```sql +SELECT map('a', map('b', 42), 'c', [1,2,3], 'd', 'Hello, World!')::JSON AS json; +``` + +```text +┌─json───────────────────────────────────────────┐ +│ {"a":{"b":42},"c":[1,2,3],"d":"Hello, World!"} │ +└────────────────────────────────────────────────┘ +``` + +Using CAST from deprecated `Object('json')`: + +```sql + SELECT '{"a" : {"b" : 42},"c" : [1, 2, 3], "d" : "Hello, World!"}'::Object('json')::JSON AS json; + ``` + +```text +┌─json───────────────────────────────────────────┐ +│ {"a":{"b":42},"c":[1,2,3],"d":"Hello, World!"} │ +└────────────────────────────────────────────────┘ +``` + +:::note +CAST from `Tuple`/`Map`/`Object('json')` to `JSON` is implemented via serializing the column into `String` column containing JSON objects and deserializing it back to `JSON` type column. +::: + +CAST between `JSON` types with different arguments will be supported later. ## Reading JSON paths as subcolumns From ca389d0d71c96998f0c9feeca6ffae913a02fa77 Mon Sep 17 00:00:00 2001 From: kssenii Date: Thu, 31 Oct 2024 18:43:56 +0100 Subject: [PATCH 366/680] Move settings to cloud level --- src/Core/Settings.cpp | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 404f5a6b090..ee814e72447 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -4846,12 +4846,6 @@ Limit on size of a single batch of file segments that a read buffer can request )", 0) \ DECLARE(UInt64, filesystem_cache_reserve_space_wait_lock_timeout_milliseconds, 1000, R"( Wait time to lock cache for space reservation in filesystem cache -)", 0) \ - DECLARE(Bool, filesystem_cache_enable_background_download_for_metadata_files_in_packed_storage, true, R"( -Wait time to lock cache for space reservation in filesystem cache -)", 0) \ - DECLARE(Bool, filesystem_cache_enable_background_download_during_fetch, true, R"( -Wait time to lock cache for space reservation in filesystem cache )", 0) \ DECLARE(UInt64, temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds, (10 * 60 * 1000), R"( Wait time to lock cache for space reservation for temporary data in filesystem cache @@ -5112,6 +5106,12 @@ Only in ClickHouse Cloud. A maximum number of unacknowledged in-flight packets i )", 0) \ DECLARE(UInt64, distributed_cache_data_packet_ack_window, DistributedCache::ACK_DATA_PACKET_WINDOW, R"( Only in ClickHouse Cloud. A window for sending ACK for DataPacket sequence in a single distributed cache read request +)", 0) \ + DECLARE(Bool, filesystem_cache_enable_background_download_for_metadata_files_in_packed_storage, true, R"( +Only in ClickHouse Cloud. Wait time to lock cache for space reservation in filesystem cache +)", 0) \ + DECLARE(Bool, filesystem_cache_enable_background_download_during_fetch, true, R"( +Only in ClickHouse Cloud. Wait time to lock cache for space reservation in filesystem cache )", 0) \ \ DECLARE(Bool, parallelize_output_from_storages, true, R"( @@ -5122,6 +5122,7 @@ The setting allows a user to provide own deduplication semantic in MergeTree/Rep For example, by providing a unique value for the setting in each INSERT statement, user can avoid the same inserted data being deduplicated. + Possible values: - Any string From 77298ef479befda70073216255658f656bf5fba5 Mon Sep 17 00:00:00 2001 From: jsc0218 Date: Thu, 31 Oct 2024 18:23:06 +0000 Subject: [PATCH 367/680] add setting --- src/Core/Settings.cpp | 3 +++ src/Core/SettingsChangesHistory.cpp | 1 + src/Processors/QueryPlan/ReadFromMergeTree.cpp | 3 ++- tests/queries/0_stateless/01786_explain_merge_tree.sh | 2 +- tests/queries/0_stateless/02149_read_in_order_fixed_prefix.sql | 1 + .../03031_read_in_order_optimization_with_virtual_row.sql | 2 ++ ...031_read_in_order_optimization_with_virtual_row_explain.sql | 2 +- ...031_read_in_order_optimization_with_virtual_row_special.sql | 2 ++ 8 files changed, 13 insertions(+), 3 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 0aecb7cf941..37646dc86cb 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -2863,6 +2863,9 @@ Possible values: **See Also** - [ORDER BY Clause](../../sql-reference/statements/select/order-by.md/#optimize_read_in_order) +)", 0) \ + DECLARE(Bool, read_in_order_use_virtual_row, false, R"( +Use virtual row while reading in order of primary key or its monotonic function fashion. It is useful when searching over multiple parts as only relevant ones are touched. )", 0) \ DECLARE(Bool, optimize_read_in_window_order, true, R"( Enable ORDER BY optimization in window clause for reading data in corresponding order in MergeTree tables. diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 88d39d6d393..4b014e141ac 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -64,6 +64,7 @@ static std::initializer_listgetSettingsRef()[Setting::read_in_order_use_virtual_row]) virtual_row_conversion = std::make_shared(std::move(*virtual_row_conversion_)); updateSortDescription(); diff --git a/tests/queries/0_stateless/01786_explain_merge_tree.sh b/tests/queries/0_stateless/01786_explain_merge_tree.sh index 828012f56bc..9fb764dcd38 100755 --- a/tests/queries/0_stateless/01786_explain_merge_tree.sh +++ b/tests/queries/0_stateless/01786_explain_merge_tree.sh @@ -7,7 +7,7 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) for i in $(seq 0 1) do - CH_CLIENT="$CLICKHOUSE_CLIENT --optimize_move_to_prewhere=1 --convert_query_to_cnf=0 --optimize_read_in_order=1 --enable_analyzer=$i" + CH_CLIENT="$CLICKHOUSE_CLIENT --optimize_move_to_prewhere=1 --convert_query_to_cnf=0 --optimize_read_in_order=1 --read_in_order_use_virtual_row=1 --enable_analyzer=$i" $CH_CLIENT -q "drop table if exists test_index" $CH_CLIENT -q "drop table if exists idx" diff --git a/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.sql b/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.sql index 7bbdecf5501..4cc05203b6a 100644 --- a/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.sql +++ b/tests/queries/0_stateless/02149_read_in_order_fixed_prefix.sql @@ -2,6 +2,7 @@ SET max_threads=0; SET optimize_read_in_order=1; SET optimize_trivial_insert_select = 1; SET read_in_order_two_level_merge_threshold=100; +SET read_in_order_use_virtual_row = 1; DROP TABLE IF EXISTS t_read_in_order; diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql index 8826f2c27cf..0f100287815 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row.sql @@ -1,4 +1,6 @@ +SET read_in_order_use_virtual_row = 1; + DROP TABLE IF EXISTS t; CREATE TABLE t diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql index 8cdcb4628ec..8e3f37b37b8 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_explain.sql @@ -1,6 +1,6 @@ -- Tags: no-random-merge-tree-settings, no-object-storage -SET optimize_read_in_order = 1, merge_tree_min_rows_for_concurrent_read = 1000; +SET optimize_read_in_order = 1, merge_tree_min_rows_for_concurrent_read = 1000, read_in_order_use_virtual_row = 1; DROP TABLE IF EXISTS tab; diff --git a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql index 3d6f9ad391b..52aa71437db 100644 --- a/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql +++ b/tests/queries/0_stateless/03031_read_in_order_optimization_with_virtual_row_special.sql @@ -7,6 +7,8 @@ CREATE DATABASE test_03031; USE test_03031; +SET read_in_order_use_virtual_row = 1; + CREATE TABLE src (s String) ENGINE = MergeTree() ORDER BY s; INSERT INTO src(s) VALUES ('before moving tables'); CREATE TABLE dist (s String) ENGINE = Distributed(test_shard_localhost, test_03031, src); From b9232c20063054525f0c192f528d77d85e1af9ff Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Fri, 1 Nov 2024 10:09:54 +0800 Subject: [PATCH 368/680] add uts --- .../0_stateless/03258_quantile_exact_weighted_issue.reference | 2 ++ .../queries/0_stateless/03258_quantile_exact_weighted_issue.sql | 2 ++ 2 files changed, 4 insertions(+) create mode 100644 tests/queries/0_stateless/03258_quantile_exact_weighted_issue.reference create mode 100644 tests/queries/0_stateless/03258_quantile_exact_weighted_issue.sql diff --git a/tests/queries/0_stateless/03258_quantile_exact_weighted_issue.reference b/tests/queries/0_stateless/03258_quantile_exact_weighted_issue.reference new file mode 100644 index 00000000000..69afec5d545 --- /dev/null +++ b/tests/queries/0_stateless/03258_quantile_exact_weighted_issue.reference @@ -0,0 +1,2 @@ +AggregateFunction(quantilesExactWeighted(0.2, 0.4, 0.6, 0.8), UInt64, UInt8) +AggregateFunction(quantilesExactWeightedInterpolated(0.2, 0.4, 0.6, 0.8), UInt64, UInt8) diff --git a/tests/queries/0_stateless/03258_quantile_exact_weighted_issue.sql b/tests/queries/0_stateless/03258_quantile_exact_weighted_issue.sql new file mode 100644 index 00000000000..3069389f4e2 --- /dev/null +++ b/tests/queries/0_stateless/03258_quantile_exact_weighted_issue.sql @@ -0,0 +1,2 @@ +SELECT toTypeName(quantilesExactWeightedState(0.2, 0.4, 0.6, 0.8)(number + 1, 1) AS x) FROM numbers(49999); +SELECT toTypeName(quantilesExactWeightedInterpolatedState(0.2, 0.4, 0.6, 0.8)(number + 1, 1) AS x) FROM numbers(49999); From a77caf42149ab864a3c96df09d7fc8771362adaa Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Fri, 1 Nov 2024 03:41:03 +0000 Subject: [PATCH 369/680] Exempt refreshable materialized views from ignore_empty_sql_security_in_create_view_query --- src/Interpreters/InterpreterCreateQuery.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Interpreters/InterpreterCreateQuery.cpp b/src/Interpreters/InterpreterCreateQuery.cpp index a38a7ab45d1..f6586f8bfc2 100644 --- a/src/Interpreters/InterpreterCreateQuery.cpp +++ b/src/Interpreters/InterpreterCreateQuery.cpp @@ -1467,7 +1467,7 @@ BlockIO InterpreterCreateQuery::createTable(ASTCreateQuery & create) bool is_secondary_query = getContext()->getZooKeeperMetadataTransaction() && !getContext()->getZooKeeperMetadataTransaction()->isInitialQuery(); auto mode = getLoadingStrictnessLevel(create.attach, /*force_attach*/ false, /*has_force_restore_data_flag*/ false, is_secondary_query || is_restore_from_backup); - if (!create.sql_security && create.supportSQLSecurity() && !getContext()->getServerSettings()[ServerSetting::ignore_empty_sql_security_in_create_view_query]) + if (!create.sql_security && create.supportSQLSecurity() && (create.refresh_strategy || !getContext()->getServerSettings()[ServerSetting::ignore_empty_sql_security_in_create_view_query])) create.sql_security = std::make_shared(); if (create.sql_security) From e851e8f3e48df739ac270d7b8672b1cd38dbad2e Mon Sep 17 00:00:00 2001 From: MikhailBurdukov Date: Fri, 1 Nov 2024 08:29:12 +0000 Subject: [PATCH 370/680] Restart CI From a50bc3bac15867ce0ee2d90afa480efdc9c98670 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Fri, 1 Nov 2024 08:50:54 +0000 Subject: [PATCH 371/680] Update version_date.tsv and changelogs after v24.10.1.2812-stable --- SECURITY.md | 3 +- docker/keeper/Dockerfile | 2 +- docker/server/Dockerfile.alpine | 2 +- docker/server/Dockerfile.ubuntu | 2 +- docs/changelogs/v24.10.1.2812-stable.md | 412 ++++++++++++++++++++++++ utils/list-versions/version_date.tsv | 1 + 6 files changed, 418 insertions(+), 4 deletions(-) create mode 100644 docs/changelogs/v24.10.1.2812-stable.md diff --git a/SECURITY.md b/SECURITY.md index db302da8ecd..1b0648dc489 100644 --- a/SECURITY.md +++ b/SECURITY.md @@ -14,9 +14,10 @@ The following versions of ClickHouse server are currently supported with securit | Version | Supported | |:-|:-| +| 24.10 | ✔️ | | 24.9 | ✔️ | | 24.8 | ✔️ | -| 24.7 | ✔️ | +| 24.7 | ❌ | | 24.6 | ❌ | | 24.5 | ❌ | | 24.4 | ❌ | diff --git a/docker/keeper/Dockerfile b/docker/keeper/Dockerfile index dfe6a420260..bc76bdbb619 100644 --- a/docker/keeper/Dockerfile +++ b/docker/keeper/Dockerfile @@ -34,7 +34,7 @@ RUN arch=${TARGETARCH:-amd64} \ # lts / testing / prestable / etc ARG REPO_CHANNEL="stable" ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}" -ARG VERSION="24.9.2.42" +ARG VERSION="24.10.1.2812" ARG PACKAGES="clickhouse-keeper" ARG DIRECT_DOWNLOAD_URLS="" diff --git a/docker/server/Dockerfile.alpine b/docker/server/Dockerfile.alpine index 991c25ad142..93acf1a5773 100644 --- a/docker/server/Dockerfile.alpine +++ b/docker/server/Dockerfile.alpine @@ -35,7 +35,7 @@ RUN arch=${TARGETARCH:-amd64} \ # lts / testing / prestable / etc ARG REPO_CHANNEL="stable" ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}" -ARG VERSION="24.9.2.42" +ARG VERSION="24.10.1.2812" ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static" ARG DIRECT_DOWNLOAD_URLS="" diff --git a/docker/server/Dockerfile.ubuntu b/docker/server/Dockerfile.ubuntu index 5dc88b49e31..506a627b11c 100644 --- a/docker/server/Dockerfile.ubuntu +++ b/docker/server/Dockerfile.ubuntu @@ -28,7 +28,7 @@ RUN sed -i "s|http://archive.ubuntu.com|${apt_archive}|g" /etc/apt/sources.list ARG REPO_CHANNEL="stable" ARG REPOSITORY="deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb ${REPO_CHANNEL} main" -ARG VERSION="24.9.2.42" +ARG VERSION="24.10.1.2812" ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static" #docker-official-library:off diff --git a/docs/changelogs/v24.10.1.2812-stable.md b/docs/changelogs/v24.10.1.2812-stable.md new file mode 100644 index 00000000000..c26bbf706ff --- /dev/null +++ b/docs/changelogs/v24.10.1.2812-stable.md @@ -0,0 +1,412 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.10.1.2812-stable (9cd0a3738d5) FIXME as compared to v24.10.1.1-new (b12a3677418) + +#### Backward Incompatible Change +* Allow to write `SETTINGS` before `FORMAT` in a chain of queries with `UNION` when subqueries are inside parentheses. This closes [#39712](https://github.com/ClickHouse/ClickHouse/issues/39712). Change the behavior when a query has the SETTINGS clause specified twice in a sequence. The closest SETTINGS clause will have a preference for the corresponding subquery. In the previous versions, the outermost SETTINGS clause could take a preference over the inner one. [#68614](https://github.com/ClickHouse/ClickHouse/pull/68614) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Reordering of filter conditions from `[PRE]WHERE` clause is now allowed by default. It could be disabled by setting `allow_reorder_prewhere_conditions` to `false`. [#70657](https://github.com/ClickHouse/ClickHouse/pull/70657) ([Nikita Taranov](https://github.com/nickitat)). +* Fix `optimize_functions_to_subcolumns` optimization (previously could lead to `Invalid column type for ColumnUnique::insertRangeFrom. Expected String, got LowCardinality(String)` error), by preserving `LowCardinality` type in `mapKeys`/`mapValues`. [#70716](https://github.com/ClickHouse/ClickHouse/pull/70716) ([Azat Khuzhin](https://github.com/azat)). +* Remove the `idxd-config` library, which has an incompatible license. This also removes the experimental Intel DeflateQPL codec. [#70987](https://github.com/ClickHouse/ClickHouse/pull/70987) ([Alexey Milovidov](https://github.com/alexey-milovidov)). + +#### New Feature +* MongoDB integration refactored: migration to new driver mongocxx from deprecated Poco::MongoDB, remove support for deprecated old protocol, support for connection by URI, support for all MongoDB types, support for WHERE and ORDER BY statements on MongoDB side, restriction for expression unsupported by MongoDB. [#63279](https://github.com/ClickHouse/ClickHouse/pull/63279) ([Kirill Nikiforov](https://github.com/allmazz)). +* A new `--progress-table` option in clickhouse-client prints a table with metrics changing during query execution; a new `--enable-progress-table-toggle` is associated with the `--progress-table` option, and toggles the rendering of the progress table by pressing the control key (Space). [#63689](https://github.com/ClickHouse/ClickHouse/pull/63689) ([Maria Khristenko](https://github.com/mariaKhr)). +* This allows to grant access to the wildcard prefixes. `GRANT SELECT ON db.table_pefix_* TO user`. [#65311](https://github.com/ClickHouse/ClickHouse/pull/65311) ([pufit](https://github.com/pufit)). +* Add system.query_metric_log which contains history of memory and metric values from table system.events for individual queries, periodically flushed to disk. [#66532](https://github.com/ClickHouse/ClickHouse/pull/66532) ([Pablo Marcos](https://github.com/pamarcos)). +* A simple SELECT query can be written with implicit SELECT to enable calculator-style expressions, e.g., `ch "1 + 2"`. This is controlled by a new setting, `implicit_select`. [#68502](https://github.com/ClickHouse/ClickHouse/pull/68502) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Support --copy mode for clickhouse local as a shortcut for format conversion [#68503](https://github.com/ClickHouse/ClickHouse/issues/68503). [#68583](https://github.com/ClickHouse/ClickHouse/pull/68583) ([Denis Hananein](https://github.com/denis-hananein)). +* Add support for `arrayUnion` function. [#68989](https://github.com/ClickHouse/ClickHouse/pull/68989) ([Peter Nguyen](https://github.com/petern48)). +* Support aggreate function `quantileExactWeightedInterpolated`, which is a interpolated version based on quantileExactWeighted. Some people may wonder why we need a new `quantileExactWeightedInterpolated` since we already have `quantileExactInterpolatedWeighted`. The reason is the new one is more accurate than the old one. BTW, it is for spark compatiability in Apache Gluten. [#69619](https://github.com/ClickHouse/ClickHouse/pull/69619) ([李扬](https://github.com/taiyang-li)). +* Support function arrayElementOrNull. It returns null if array index is out of range or map key not found. [#69646](https://github.com/ClickHouse/ClickHouse/pull/69646) ([李扬](https://github.com/taiyang-li)). +* Allows users to specify regular expressions through new `message_regexp` and `message_regexp_negative` fields in the `config.xml` file to filter out logging. The logging is applied to the formatted un-colored text for the most intuitive developer experience. [#69657](https://github.com/ClickHouse/ClickHouse/pull/69657) ([Peter Nguyen](https://github.com/petern48)). +* Support Dynamic type in most functions by executing them on internal types inside Dynamic. [#69691](https://github.com/ClickHouse/ClickHouse/pull/69691) ([Pavel Kruglov](https://github.com/Avogar)). +* Re-added `RIPEMD160` function, which computes the RIPEMD-160 cryptographic hash of a string. Example: `SELECT HEX(RIPEMD160('The quick brown fox jumps over the lazy dog'))` returns `37F332F68DB77BD9D7EDD4969571AD671CF9DD3B`. [#70087](https://github.com/ClickHouse/ClickHouse/pull/70087) ([Dergousov Maxim](https://github.com/m7kss1)). +* Allow to cache read files for object storage table engines and data lakes using hash from ETag + file path as cache key. [#70135](https://github.com/ClickHouse/ClickHouse/pull/70135) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Support reading Iceberg tables on HDFS. [#70268](https://github.com/ClickHouse/ClickHouse/pull/70268) ([flynn](https://github.com/ucasfl)). +* Allow to read/write JSON type as binary string in RowBinary format under settings `input_format_binary_read_json_as_string/output_format_binary_write_json_as_string`. [#70288](https://github.com/ClickHouse/ClickHouse/pull/70288) ([Pavel Kruglov](https://github.com/Avogar)). +* Allow to serialize/deserialize JSON column as single String column in Native format. For output use setting `output_format_native_write_json_as_string`. For input, use serialization version `1` before the column data. [#70312](https://github.com/ClickHouse/ClickHouse/pull/70312) ([Pavel Kruglov](https://github.com/Avogar)). +* Supports standard CTE, `with insert`, as previously only supports `insert ... with ...`. [#70593](https://github.com/ClickHouse/ClickHouse/pull/70593) ([Shichao Jin](https://github.com/jsc0218)). + +#### Performance Improvement +* Support minmax index for `pointInPolygon`. [#62085](https://github.com/ClickHouse/ClickHouse/pull/62085) ([JackyWoo](https://github.com/JackyWoo)). +* Add support for parquet bloom filters. [#62966](https://github.com/ClickHouse/ClickHouse/pull/62966) ([Arthur Passos](https://github.com/arthurpassos)). +* Lock-free parts rename to avoid INSERT affect SELECT (due to parts lock) (under normal circumstances with `fsync_part_directory`, QPS of SELECT with INSERT in parallel, increased 2x, under heavy load the effect is even bigger). Note, this only includes `ReplicatedMergeTree` for now. [#64955](https://github.com/ClickHouse/ClickHouse/pull/64955) ([Azat Khuzhin](https://github.com/azat)). +* Respect `ttl_only_drop_parts` on `materialize ttl`; only read necessary columns to recalculate TTL and drop parts by replacing them with an empty one. [#65488](https://github.com/ClickHouse/ClickHouse/pull/65488) ([Andrey Zvonov](https://github.com/zvonand)). +* Refactor `IDisk` and `IObjectStorage` for better performance. Tables from `plain` and `plain_rewritable` object storages will initialize faster. [#68146](https://github.com/ClickHouse/ClickHouse/pull/68146) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Optimized thread creation in the ThreadPool to minimize lock contention. Thread creation is now performed outside of the critical section to avoid delays in job scheduling and thread management under high load conditions. This leads to a much more responsive ClickHouse under heavy concurrent load. [#68694](https://github.com/ClickHouse/ClickHouse/pull/68694) ([filimonov](https://github.com/filimonov)). +* Enable reading LowCardinality string columns from ORC. [#69481](https://github.com/ClickHouse/ClickHouse/pull/69481) ([李扬](https://github.com/taiyang-li)). +* Added an ability to parse data directly into sparse columns. [#69828](https://github.com/ClickHouse/ClickHouse/pull/69828) ([Anton Popov](https://github.com/CurtizJ)). +* Supports parallel reading of parquet row groups and prefetching of row groups in single-threaded mode. [#69862](https://github.com/ClickHouse/ClickHouse/pull/69862) ([LiuNeng](https://github.com/liuneng1994)). +* Improved performance of parsing formats with high number of missed values (e.g. `JSONEachRow`). [#69875](https://github.com/ClickHouse/ClickHouse/pull/69875) ([Anton Popov](https://github.com/CurtizJ)). +* Use `LowCardinality` for `ProfileEvents` in system logs such as `part_log`, `query_views_log`, `filesystem_cache_log`. [#70152](https://github.com/ClickHouse/ClickHouse/pull/70152) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Improve performance of FromUnixTimestamp/ToUnixTimestamp functions. [#71042](https://github.com/ClickHouse/ClickHouse/pull/71042) ([kevinyhzou](https://github.com/KevinyhZou)). + +#### Improvement +* Allow parametrised SQL aliases. [#50665](https://github.com/ClickHouse/ClickHouse/pull/50665) ([Anton Kozlov](https://github.com/tonickkozlov)). +* Fixed [#57616](https://github.com/ClickHouse/ClickHouse/issues/57616) this problem occurs because all positive number arguments are automatically identified as `uint64` type, leading to an inability to match int type data in `summapfiltered`. the issue of non-matching is indeed confusing, as the `uint64` parameters are not specified by the user. additionally, if the arguments are `[1,2,3,toint8(-3)]`, due to the `getleastsupertype()`, these parameters will be uniformly treated as `int` type, causing `'1,2,3'` to also fail in matching the `uint` type data in `summapfiltered`. [#58408](https://github.com/ClickHouse/ClickHouse/pull/58408) ([Chen768959](https://github.com/Chen768959)). +* `ALTER TABLE .. REPLACE PARTITION` doesn't wait anymore for mutations/merges that happen in other partitions. [#59138](https://github.com/ClickHouse/ClickHouse/pull/59138) ([Vasily Nemkov](https://github.com/Enmk)). +* Refreshable materialized views are now supported in Replicated databases. [#60669](https://github.com/ClickHouse/ClickHouse/pull/60669) ([Michael Kolupaev](https://github.com/al13n321)). +* Symbolic links for tables in the `data/database_name/` directory are created for the actual paths to the table's data, depending on the storage policy, instead of the `store/...` directory on the default disk. [#61777](https://github.com/ClickHouse/ClickHouse/pull/61777) ([Kirill](https://github.com/kirillgarbar)). +* Apply configuration updates in global context object. It fixes issues like [#62308](https://github.com/ClickHouse/ClickHouse/issues/62308). [#62944](https://github.com/ClickHouse/ClickHouse/pull/62944) ([Amos Bird](https://github.com/amosbird)). +* Reworked settings that control the behavior of parallel replicas algorithms. A quick recap: ClickHouse has four different algorithms for parallel reading involving multiple replicas, which is reflected in the setting `parallel_replicas_mode`, the default value for it is `read_tasks` Additionally, the toggle-switch setting `enable_parallel_replicas` has been added. [#63151](https://github.com/ClickHouse/ClickHouse/pull/63151) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix `ReadSettings` not using user set values, because defaults were only used. [#65625](https://github.com/ClickHouse/ClickHouse/pull/65625) ([Kseniia Sumarokova](https://github.com/kssenii)). +* While parsing an Enum field from JSON, a string containing an integer will be interpreted as the corresponding Enum element. This closes [#65119](https://github.com/ClickHouse/ClickHouse/issues/65119). [#66801](https://github.com/ClickHouse/ClickHouse/pull/66801) ([scanhex12](https://github.com/scanhex12)). +* Allow `TRIM` -ing `LEADING` or `TRAILING` empty string as a no-op. Closes [#67792](https://github.com/ClickHouse/ClickHouse/issues/67792). [#68455](https://github.com/ClickHouse/ClickHouse/pull/68455) ([Peter Nguyen](https://github.com/petern48)). +* Support creating a table with a query: `CREATE TABLE ... CLONE AS ...`. It clones the source table's schema and then attaches all partitions to the newly created table. This feature is only supported with tables of the `MergeTree` family Closes [#65015](https://github.com/ClickHouse/ClickHouse/issues/65015). [#69091](https://github.com/ClickHouse/ClickHouse/pull/69091) ([tuanpach](https://github.com/tuanpach)). +* In Gluten ClickHouse, Spark's timestamp type is mapped to ClickHouse's datetime64(6) type. When casting timestamp '2012-01-01 00:11:22' as a string, Spark returns '2012-01-01 00:11:22', while Gluten ClickHouse returns '2012-01-01 00:11:22.000000'. [#69179](https://github.com/ClickHouse/ClickHouse/pull/69179) ([Wenzheng Liu](https://github.com/lwz9103)). +* Always use the new analyzer to calculate constant expressions when `enable_analyzer` is set to `true`. Support calculation of `executable()` table function arguments without using `SELECT` query for constant expression. [#69292](https://github.com/ClickHouse/ClickHouse/pull/69292) ([Dmitry Novik](https://github.com/novikd)). +* Add `enable_secure_identifiers` to disallow insecure identifiers. [#69411](https://github.com/ClickHouse/ClickHouse/pull/69411) ([tuanpach](https://github.com/tuanpach)). +* Add `show_create_query_identifier_quoting_rule` to define identifier quoting behavior of the show create query result. Possible values: - `user_display`: When the identifiers is a keyword. - `when_necessary`: When the identifiers is one of `{"distinct", "all", "table"}`, or it can cause ambiguity: column names, dictionary attribute names. - `always`: Always quote identifiers. [#69448](https://github.com/ClickHouse/ClickHouse/pull/69448) ([tuanpach](https://github.com/tuanpach)). +* Follow-up to https://github.com/ClickHouse/ClickHouse/pull/69346 Point 4 described there will work now as well:. [#69563](https://github.com/ClickHouse/ClickHouse/pull/69563) ([Vitaly Baranov](https://github.com/vitlibar)). +* Implement generic SerDe between Avro Union and ClickHouse Variant type. Resolves [#69713](https://github.com/ClickHouse/ClickHouse/issues/69713). [#69712](https://github.com/ClickHouse/ClickHouse/pull/69712) ([Jiří Kozlovský](https://github.com/jirislav)). +* 1. CREATE TABLE AS will copy PRIMARY KEY, ORDER BY, and similar clauses. Now it is supported only for the MergeTree family of table engines. 2. For example, the follow SQL statements will trigger exception in the past, but this PR fixes it: if the destination table do not provide an `ORDER BY` or `PRIMARY KEY` expression in the table definition, we will copy that from source table. [#69739](https://github.com/ClickHouse/ClickHouse/pull/69739) ([sakulali](https://github.com/sakulali)). +* Added user-level settings `min_free_disk_bytes_to_throw_insert` and `min_free_disk_ratio_to_throw_insert` to prevent insertions on disks that are almost full. [#69755](https://github.com/ClickHouse/ClickHouse/pull/69755) ([Marco Vilas Boas](https://github.com/marco-vb)). +* If you run `clickhouse-client` or other CLI application and it starts up slowly due to an overloaded server, and you start typing your query, such as `SELECT`, the previous versions will display the remaining of the terminal echo contents before printing the greetings message, such as `SELECTClickHouse local version 24.10.1.1.` instead of `ClickHouse local version 24.10.1.1.`. Now it is fixed. This closes [#31696](https://github.com/ClickHouse/ClickHouse/issues/31696). [#69856](https://github.com/ClickHouse/ClickHouse/pull/69856) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add new column readonly_duration to the system.replicas table. Needed to be able to distinguish actual readonly replicas from sentinel ones in alerts. [#69871](https://github.com/ClickHouse/ClickHouse/pull/69871) ([Miсhael Stetsyuk](https://github.com/mstetsyuk)). +* Change the join to sort settings type to unsigned int. [#69886](https://github.com/ClickHouse/ClickHouse/pull/69886) ([kevinyhzou](https://github.com/KevinyhZou)). +* Support 64-bit XID in Keeper. It can be enabled with `use_xid_64` config. [#69908](https://github.com/ClickHouse/ClickHouse/pull/69908) ([Antonio Andelic](https://github.com/antonio2368)). +* New function getSettingOrDefault() added to return the default value and avoid exception if a custom setting is not found in the current profile. [#69917](https://github.com/ClickHouse/ClickHouse/pull/69917) ([Shankar](https://github.com/shiyer7474)). +* Allow empty needle in function replace, the same behavior with PostgreSQL. [#69918](https://github.com/ClickHouse/ClickHouse/pull/69918) ([zhanglistar](https://github.com/zhanglistar)). +* Enhance OpenTelemetry span logging to include query settings. [#70011](https://github.com/ClickHouse/ClickHouse/pull/70011) ([sharathks118](https://github.com/sharathks118)). +* Allow empty needle in functions replaceRegexp*, like https://github.com/ClickHouse/ClickHouse/pull/69918. [#70053](https://github.com/ClickHouse/ClickHouse/pull/70053) ([zhanglistar](https://github.com/zhanglistar)). +* Add info to higher-order array functions if lambda result type is unexpected. [#70093](https://github.com/ClickHouse/ClickHouse/pull/70093) ([ttanay](https://github.com/ttanay)). +* Keeper improvement: less blocking during cluster changes. [#70275](https://github.com/ClickHouse/ClickHouse/pull/70275) ([Antonio Andelic](https://github.com/antonio2368)). +* Embedded documentation for settings will be strictly more detailed and complete than the documentation on the website. This is the first step before making the website documentation always auto-generated from the source code. This has long-standing implications: - it will be guaranteed to have every setting; - there is no chance of having default values obsolete; - we can generate this documentation for each ClickHouse version; - the documentation can be displayed by the server itself even without Internet access. Generate the docs on the website from the source code. [#70289](https://github.com/ClickHouse/ClickHouse/pull/70289) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add `WITH IMPLICIT` and `FINAL` keywords to the `SHOW GRANTS` command. Fix a minor bug with implicit grants: [#70094](https://github.com/ClickHouse/ClickHouse/issues/70094). [#70293](https://github.com/ClickHouse/ClickHouse/pull/70293) ([pufit](https://github.com/pufit)). +* Don't disable nonblocking read from page cache for the entire server when reading from a blocking I/O. [#70299](https://github.com/ClickHouse/ClickHouse/pull/70299) ([Antonio Andelic](https://github.com/antonio2368)). +* Respect `compatibility` for MergeTree settings. The `compatibility` value is taken from the `default` profile on server startup, and default MergeTree settings are changed accordingly. Further changes of the `compatibility` setting do not affect MergeTree settings. [#70322](https://github.com/ClickHouse/ClickHouse/pull/70322) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Clickhouse-client realtime metrics follow-up: restore cursor when ctrl-c cancels query; immediately stop intercepting keystrokes when the query is canceled; display the metrics table if `--progress-table` is on, and toggling is disabled. [#70423](https://github.com/ClickHouse/ClickHouse/pull/70423) ([Julia Kartseva](https://github.com/jkartseva)). +* Command-line arguments for Bool settings are set to true when no value is provided for the argument (e.g. `clickhouse-client --optimize_aggregation_in_order --query "SELECT 1"`). [#70459](https://github.com/ClickHouse/ClickHouse/pull/70459) ([davidtsuk](https://github.com/davidtsuk)). +* Avoid spamming the logs with large HTTP response bodies in case of errors during inter-server communication. [#70487](https://github.com/ClickHouse/ClickHouse/pull/70487) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Added a new setting `max_parts_to_move` to control the maximum number of parts that can be moved at once. [#70520](https://github.com/ClickHouse/ClickHouse/pull/70520) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Limit the frequency of certain log messages. [#70601](https://github.com/ClickHouse/ClickHouse/pull/70601) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Don't do validation when synchronizing user_directories from keeper. [#70644](https://github.com/ClickHouse/ClickHouse/pull/70644) ([Raúl Marín](https://github.com/Algunenano)). +* Introduced a special (experimental) mode of a merge selector for MergeTree tables which makes it more aggressive for the partitions that are close to the limit by the number of parts. It is controlled by the `merge_selector_use_blurry_base` MergeTree-level setting. [#70645](https://github.com/ClickHouse/ClickHouse/pull/70645) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* `CHECK TABLE` with `PART` qualifier was incorrectly formatted in the client. [#70660](https://github.com/ClickHouse/ClickHouse/pull/70660) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Support write column index and offset index using parquet native writer. [#70669](https://github.com/ClickHouse/ClickHouse/pull/70669) ([LiuNeng](https://github.com/liuneng1994)). +* Support parse `DateTime64` for microseond and timezone in joda syntax. [#70737](https://github.com/ClickHouse/ClickHouse/pull/70737) ([kevinyhzou](https://github.com/KevinyhZou)). +* Changed an approach to figure out if a cloud storage supports [batch delete](https://docs.aws.amazon.com/AmazonS3/latest/API/API_DeleteObjects.html) or not. [#70786](https://github.com/ClickHouse/ClickHouse/pull/70786) ([Vitaly Baranov](https://github.com/vitlibar)). +* Support for Parquet page V2 on native reader. [#70807](https://github.com/ClickHouse/ClickHouse/pull/70807) ([Arthur Passos](https://github.com/arthurpassos)). +* Add an HTML page for visualizing merges. [#70821](https://github.com/ClickHouse/ClickHouse/pull/70821) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#71234](https://github.com/ClickHouse/ClickHouse/issues/71234): Do not call the object storage API when listing directories, as this may be cost-inefficient. Instead, store the list of filenames in the memory. The trade-offs are increased initial load time and memory required to store filenames. [#70823](https://github.com/ClickHouse/ClickHouse/pull/70823) ([Julia Kartseva](https://github.com/jkartseva)). +* A check if table has both `storage_policy` and `disk` set after alter query is added. A check if a new storage policy is compatible with an old one when using `disk` setting is added. [#70839](https://github.com/ClickHouse/ClickHouse/pull/70839) ([Kirill](https://github.com/kirillgarbar)). +* Add system.s3_queue_settings and system.azure_queue_settings. [#70841](https://github.com/ClickHouse/ClickHouse/pull/70841) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Functions `base58Encode` and `base58Decode` now accept arguments of type `FixedString`. Example: `SELECT base58Encode(toFixedString('plaintext', 9));`. [#70846](https://github.com/ClickHouse/ClickHouse/pull/70846) ([Faizan Patel](https://github.com/faizan2786)). +* Add the `partition` column to every entry type of the part log. Previously, it was set only for some entries. This closes [#70819](https://github.com/ClickHouse/ClickHouse/issues/70819). [#70848](https://github.com/ClickHouse/ClickHouse/pull/70848) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add merge start and mutate start events into `system.part_log` which helps with merges analysis and visualization. [#70850](https://github.com/ClickHouse/ClickHouse/pull/70850) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Do not call the LIST object storage API when determining if a file or directory exists on the plain rewritable disk, as it can be cost-inefficient. [#70852](https://github.com/ClickHouse/ClickHouse/pull/70852) ([Julia Kartseva](https://github.com/jkartseva)). +* Add a profile event about the number of merged source parts. It allows the monitoring of the fanout of the merge tree in production. [#70908](https://github.com/ClickHouse/ClickHouse/pull/70908) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Reduce the number of object storage HEAD API requests in the plain_rewritable disk. [#70915](https://github.com/ClickHouse/ClickHouse/pull/70915) ([Julia Kartseva](https://github.com/jkartseva)). +* Background downloads to filesystem cache was enabled back. [#70929](https://github.com/ClickHouse/ClickHouse/pull/70929) ([Nikita Taranov](https://github.com/nickitat)). +* Add a new merge selector algorithm, named `Trivial`, for professional usage only. It is worse than the `Simple` merge selector. [#70969](https://github.com/ClickHouse/ClickHouse/pull/70969) ([Alexey Milovidov](https://github.com/alexey-milovidov)). + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Fix toHour-like conversion functions' monotonicity when optional time zone argument is passed. [#60264](https://github.com/ClickHouse/ClickHouse/pull/60264) ([Amos Bird](https://github.com/amosbird)). +* Relax `supportsPrewhere` check for StorageMerge. This fixes [#61064](https://github.com/ClickHouse/ClickHouse/issues/61064). It was hardened unnecessarily in [#60082](https://github.com/ClickHouse/ClickHouse/issues/60082). [#61091](https://github.com/ClickHouse/ClickHouse/pull/61091) ([Amos Bird](https://github.com/amosbird)). +* Fix `use_concurrency_control` setting handling for proper `concurrent_threads_soft_limit_num` limit enforcing. This enables concurrency control by default because previously it was broken. [#61473](https://github.com/ClickHouse/ClickHouse/pull/61473) ([Sergei Trifonov](https://github.com/serxa)). +* Fix incorrect JOIN ON section optimization in case of `IS NULL` check under any other function (like `NOT`) that may lead to wrong results. Closes [#67915](https://github.com/ClickHouse/ClickHouse/issues/67915). [#68049](https://github.com/ClickHouse/ClickHouse/pull/68049) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Prevent `ALTER` queries that would make the `CREATE` query of tables invalid. [#68574](https://github.com/ClickHouse/ClickHouse/pull/68574) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Fix inconsistent AST formatting for `negate` (`-`) and `NOT` functions with tuples and arrays. [#68600](https://github.com/ClickHouse/ClickHouse/pull/68600) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Fix insertion of incomplete type into Dynamic during deserialization. It could lead to `Parameter out of bound` errors. [#69291](https://github.com/ClickHouse/ClickHouse/pull/69291) ([Pavel Kruglov](https://github.com/Avogar)). +* Fix inf loop after `restore replica` in the replicated merge tree with zero copy. [#69293](https://github.com/ClickHouse/ClickHouse/pull/69293) ([MikhailBurdukov](https://github.com/MikhailBurdukov)). +* Return back default value of `processing_threads_num` as number of cpu cores in storage `S3Queue`. [#69384](https://github.com/ClickHouse/ClickHouse/pull/69384) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Bypass try/catch flow when de/serializing nested repeated protobuf to nested columns ( fixes [#41971](https://github.com/ClickHouse/ClickHouse/issues/41971) ). [#69556](https://github.com/ClickHouse/ClickHouse/pull/69556) ([Eliot Hautefeuille](https://github.com/hileef)). +* Fix vrash during insertion into FixedString column in PostgreSQL engine. [#69584](https://github.com/ClickHouse/ClickHouse/pull/69584) ([Pavel Kruglov](https://github.com/Avogar)). +* Fix crash when executing `create view t as (with recursive 42 as ttt select ttt);`. [#69676](https://github.com/ClickHouse/ClickHouse/pull/69676) ([Han Fei](https://github.com/hanfei1991)). +* Added `strict_once` mode to aggregate function `windowFunnel` to avoid counting one event several times in case it matches multiple conditions, close [#21835](https://github.com/ClickHouse/ClickHouse/issues/21835). [#69738](https://github.com/ClickHouse/ClickHouse/pull/69738) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Fixed `maxMapState` throwing 'Bad get' if value type is DateTime64. [#69787](https://github.com/ClickHouse/ClickHouse/pull/69787) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix `getSubcolumn` with `LowCardinality` columns by overriding `useDefaultImplementationForLowCardinalityColumns` to return `true`. [#69831](https://github.com/ClickHouse/ClickHouse/pull/69831) ([Miсhael Stetsyuk](https://github.com/mstetsyuk)). +* Fix permanent blocked distributed sends if DROP of distributed table fails. [#69843](https://github.com/ClickHouse/ClickHouse/pull/69843) ([Azat Khuzhin](https://github.com/azat)). +* Fix non-cancellable queries containing WITH FILL with NaN keys. This closes [#69261](https://github.com/ClickHouse/ClickHouse/issues/69261). [#69845](https://github.com/ClickHouse/ClickHouse/pull/69845) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix analyzer default with old compatibility value. [#69895](https://github.com/ClickHouse/ClickHouse/pull/69895) ([Raúl Marín](https://github.com/Algunenano)). +* Don't check dependencies during CREATE OR REPLACE VIEW during DROP of old table. Previously CREATE OR REPLACE query failed when there are dependent tables of the recreated view. [#69907](https://github.com/ClickHouse/ClickHouse/pull/69907) ([Pavel Kruglov](https://github.com/Avogar)). +* Implement missing decimal cases for `zeroField`. Fixes [#69730](https://github.com/ClickHouse/ClickHouse/issues/69730). [#69978](https://github.com/ClickHouse/ClickHouse/pull/69978) ([Arthur Passos](https://github.com/arthurpassos)). +* Now SQL security will work with parameterized views correctly. [#69984](https://github.com/ClickHouse/ClickHouse/pull/69984) ([pufit](https://github.com/pufit)). +* Closes [#69752](https://github.com/ClickHouse/ClickHouse/issues/69752). [#69985](https://github.com/ClickHouse/ClickHouse/pull/69985) ([pufit](https://github.com/pufit)). +* Fixed a bug when the timezone could change the result of the query with a `Date` or `Date32` arguments. [#70036](https://github.com/ClickHouse/ClickHouse/pull/70036) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Fixes `Block structure mismatch` for queries with nested views and `WHERE` condition. Fixes [#66209](https://github.com/ClickHouse/ClickHouse/issues/66209). [#70054](https://github.com/ClickHouse/ClickHouse/pull/70054) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Avoid reusing columns among different named tuples when evaluating `tuple` functions. This fixes [#70022](https://github.com/ClickHouse/ClickHouse/issues/70022). [#70103](https://github.com/ClickHouse/ClickHouse/pull/70103) ([Amos Bird](https://github.com/amosbird)). +* Fix wrong LOGICAL_ERROR when replacing literals in ranges. [#70122](https://github.com/ClickHouse/ClickHouse/pull/70122) ([Pablo Marcos](https://github.com/pamarcos)). +* Check for Nullable(Nothing) type during ALTER TABLE MODIFY COLUMN/QUERY to prevent tables with such data type. [#70123](https://github.com/ClickHouse/ClickHouse/pull/70123) ([Pavel Kruglov](https://github.com/Avogar)). +* Proper error message for illegal query `JOIN ... ON *` , close [#68650](https://github.com/ClickHouse/ClickHouse/issues/68650). [#70124](https://github.com/ClickHouse/ClickHouse/pull/70124) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Fix wrong result with skipping index. [#70127](https://github.com/ClickHouse/ClickHouse/pull/70127) ([Raúl Marín](https://github.com/Algunenano)). +* Fix data race in ColumnObject/ColumnTuple decompress method that could lead to heap use after free. [#70137](https://github.com/ClickHouse/ClickHouse/pull/70137) ([Pavel Kruglov](https://github.com/Avogar)). +* Fix possible hung in ALTER COLUMN with Dynamic type. [#70144](https://github.com/ClickHouse/ClickHouse/pull/70144) ([Pavel Kruglov](https://github.com/Avogar)). +* Now ClickHouse will consider more errors as retriable and will not mark data parts as broken in case of such errors. [#70145](https://github.com/ClickHouse/ClickHouse/pull/70145) ([alesapin](https://github.com/alesapin)). +* Use correct `max_types` parameter during Dynamic type creation for JSON subcolumn. [#70147](https://github.com/ClickHouse/ClickHouse/pull/70147) ([Pavel Kruglov](https://github.com/Avogar)). +* Fix the password being displayed in `system.query_log` for users with bcrypt password authentication method. [#70148](https://github.com/ClickHouse/ClickHouse/pull/70148) ([Nikolay Degterinsky](https://github.com/evillique)). +* Fix event counter for native interface (InterfaceNativeSendBytes). [#70153](https://github.com/ClickHouse/ClickHouse/pull/70153) ([Yakov Olkhovskiy](https://github.com/yakov-olkhovskiy)). +* Fix possible crash in JSON column. [#70172](https://github.com/ClickHouse/ClickHouse/pull/70172) ([Pavel Kruglov](https://github.com/Avogar)). +* Fix multiple issues with arrayMin and arrayMax. [#70207](https://github.com/ClickHouse/ClickHouse/pull/70207) ([Raúl Marín](https://github.com/Algunenano)). +* Respect setting allow_simdjson in JSON type parser. [#70218](https://github.com/ClickHouse/ClickHouse/pull/70218) ([Pavel Kruglov](https://github.com/Avogar)). +* Fix server segfault on creating a materialized view with two selects and an `INTERSECT`, e.g. `CREATE MATERIALIZED VIEW v0 AS (SELECT 1) INTERSECT (SELECT 1);`. [#70264](https://github.com/ClickHouse/ClickHouse/pull/70264) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Don't modify global settings with startup scripts. Previously, changing a setting in a startup script would change it globally. [#70310](https://github.com/ClickHouse/ClickHouse/pull/70310) ([Antonio Andelic](https://github.com/antonio2368)). +* Fix ALTER of Dynamic type with reducing max_types parameter that could lead to server crash. [#70328](https://github.com/ClickHouse/ClickHouse/pull/70328) ([Pavel Kruglov](https://github.com/Avogar)). +* Fix crash when using WITH FILL incorrectly. [#70338](https://github.com/ClickHouse/ClickHouse/pull/70338) ([Raúl Marín](https://github.com/Algunenano)). +* Fix possible use-after-free in `SYSTEM DROP FORMAT SCHEMA CACHE FOR Protobuf`. [#70358](https://github.com/ClickHouse/ClickHouse/pull/70358) ([Azat Khuzhin](https://github.com/azat)). +* Fix crash during GROUP BY JSON sub-object subcolumn. [#70374](https://github.com/ClickHouse/ClickHouse/pull/70374) ([Pavel Kruglov](https://github.com/Avogar)). +* Don't prefetch parts for vertical merges if part has no rows. [#70452](https://github.com/ClickHouse/ClickHouse/pull/70452) ([Antonio Andelic](https://github.com/antonio2368)). +* Fix crash in WHERE with lambda functions. [#70464](https://github.com/ClickHouse/ClickHouse/pull/70464) ([Raúl Marín](https://github.com/Algunenano)). +* Fix table creation with `CREATE ... AS table_function()` with database `Replicated` and unavailable table function source on secondary replica. [#70511](https://github.com/ClickHouse/ClickHouse/pull/70511) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Ignore all output on async insert with `wait_for_async_insert=1`. Closes [#62644](https://github.com/ClickHouse/ClickHouse/issues/62644). [#70530](https://github.com/ClickHouse/ClickHouse/pull/70530) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Ignore frozen_metadata.txt while traversing shadow directory from system.remote_data_paths. [#70590](https://github.com/ClickHouse/ClickHouse/pull/70590) ([Aleksei Filatov](https://github.com/aalexfvk)). +* Fix creation of stateful window functions on misaligned memory. [#70631](https://github.com/ClickHouse/ClickHouse/pull/70631) ([Raúl Marín](https://github.com/Algunenano)). +* Fixed rare crashes in `SELECT`-s and merges after adding a column of `Array` type with non-empty default expression. [#70695](https://github.com/ClickHouse/ClickHouse/pull/70695) ([Anton Popov](https://github.com/CurtizJ)). +* Insert into table function s3 respect query settings. [#70696](https://github.com/ClickHouse/ClickHouse/pull/70696) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Fix infinite recursion when infering a proto schema with skip unsupported fields enabled. [#70697](https://github.com/ClickHouse/ClickHouse/pull/70697) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#71122](https://github.com/ClickHouse/ClickHouse/issues/71122): `GroupArraySortedData` uses a PODArray with non-POD elements, manually calling constructors and destructors for the elements as needed. But it wasn't careful enough: in two places it forgot to call destructor, in one place it left elements uninitialized if an exception is thrown when deserializing previous elements. Then `GroupArraySortedData`'s destructor called destructors on uninitialized elements and crashed: ``` 2024.10.17 22:58:23.523790 [ 5233 ] {} BaseDaemon: ########## Short fault info ############ 2024.10.17 22:58:23.523834 [ 5233 ] {} BaseDaemon: (version 24.6.1.4609 (official build), build id: 5423339A6571004018D55BBE05D464AFA35E6718, git hash: fa6cdfda8a94890eb19bc7f22f8b0b56292f7a26) (from thread 682) Received signal 11 2024.10.17 22:58:23.523862 [ 5233 ] {} BaseDaemon: Signal description: Segmentation fault 2024.10.17 22:58:23.523883 [ 5233 ] {} BaseDaemon: Address: 0x8f. Access: . Address not mapped to object. 2024.10.17 22:58:23.523908 [ 5233 ] {} BaseDaemon: Stack trace: 0x0000aaaac4b78308 0x0000ffffb7701850 0x0000aaaac0104855 0x0000aaaac01048a0 0x0000aaaac501e84c 0x0000aaaac7c510d0 0x0000aaaac7c4ba20 0x0000aaaac968bbfc 0x0000aaaac968fab0 0x0000aaaac969bf50 0x0000aaaac9b7520c 0x0000aaaac9b74c74 0x0000aaaac9b8a150 0x0000aaaac9b809f0 0x0000aaaac9b80574 0x0000aaaac9b8e364 0x0000aaaac9b8e4fc 0x0000aaaac94f4328 0x0000aaaac94f428c 0x0000aaaac94f7df0 0x0000aaaac98b5a3c 0x0000aaaac950b234 0x0000aaaac49ae264 0x0000aaaac49b1dd0 0x0000aaaac49b0a80 0x0000ffffb755d5c8 0x0000ffffb75c5edc 2024.10.17 22:58:23.523936 [ 5233 ] {} BaseDaemon: ######################################## 2024.10.17 22:58:23.523959 [ 5233 ] {} BaseDaemon: (version 24.6.1.4609 (official build), build id: 5423339A6571004018D55BBE05D464AFA35E6718, git hash: fa6cdfda8a94890eb19bc7f22f8b0b56292f7a26) (from thread 682) (query_id: 6c8a33a2-f45a-4a3b-bd71-ded6a1c9ccd3::202410_534066_534078_2) (query: ) Received signal Segmentation fault (11) 2024.10.17 22:58:23.523977 [ 5233 ] {} BaseDaemon: Address: 0x8f. Access: . Address not mapped to object. 2024.10.17 22:58:23.523993 [ 5233 ] {} BaseDaemon: Stack trace: 0x0000aaaac4b78308 0x0000ffffb7701850 0x0000aaaac0104855 0x0000aaaac01048a0 0x0000aaaac501e84c 0x0000aaaac7c510d0 0x0000aaaac7c4ba20 0x0000aaaac968bbfc 0x0000aaaac968fab0 0x0000aaaac969bf50 0x0000aaaac9b7520c 0x0000aaaac9b74c74 0x0000aaaac9b8a150 0x0000aaaac9b809f0 0x0000aaaac9b80574 0x0000aaaac9b8e364 0x0000aaaac9b8e4fc 0x0000aaaac94f4328 0x0000aaaac94f428c 0x0000aaaac94f7df0 0x0000aaaac98b5a3c 0x0000aaaac950b234 0x0000aaaac49ae264 0x0000aaaac49b1dd0 0x0000aaaac49b0a80 0x0000ffffb755d5c8 0x0000ffffb75c5edc 2024.10.17 22:58:23.524817 [ 5233 ] {} BaseDaemon: 0. signalHandler(int, siginfo_t*, void*) @ 0x000000000c6f8308 2024.10.17 22:58:23.524917 [ 5233 ] {} BaseDaemon: 1. ? @ 0x0000ffffb7701850 2024.10.17 22:58:23.524962 [ 5233 ] {} BaseDaemon: 2. DB::Field::~Field() @ 0x0000000007c84855 2024.10.17 22:58:23.525012 [ 5233 ] {} BaseDaemon: 3. DB::Field::~Field() @ 0x0000000007c848a0 2024.10.17 22:58:23.526626 [ 5233 ] {} BaseDaemon: 4. DB::IAggregateFunctionDataHelper, DB::(anonymous namespace)::GroupArraySorted, DB::Field>>::destroy(char*) const (.5a6a451027f732f9fd91c13f4a13200c) @ 0x000000000cb9e84c 2024.10.17 22:58:23.527322 [ 5233 ] {} BaseDaemon: 5. DB::SerializationAggregateFunction::deserializeBinaryBulk(DB::IColumn&, DB::ReadBuffer&, unsigned long, double) const @ 0x000000000f7d10d0 2024.10.17 22:58:23.528470 [ 5233 ] {} BaseDaemon: 6. DB::ISerialization::deserializeBinaryBulkWithMultipleStreams(COW::immutable_ptr&, unsigned long, DB::ISerialization::DeserializeBinaryBulkSettings&, std::shared_ptr&, std::unordered_map::immutable_ptr, std::hash, std::equal_to, std::allocator::immutable_ptr>>>*) const @ 0x000000000f7cba20 2024.10.17 22:58:23.529213 [ 5233 ] {} BaseDaemon: 7. DB::MergeTreeReaderCompact::readData(DB::NameAndTypePair const&, COW::immutable_ptr&, unsigned long, std::function const&) @ 0x000000001120bbfc 2024.10.17 22:58:23.529277 [ 5233 ] {} BaseDaemon: 8. DB::MergeTreeReaderCompactSingleBuffer::readRows(unsigned long, unsigned long, bool, unsigned long, std::vector::immutable_ptr, std::allocator::immutable_ptr>>&) @ 0x000000001120fab0 2024.10.17 22:58:23.529319 [ 5233 ] {} BaseDaemon: 9. DB::MergeTreeSequentialSource::generate() @ 0x000000001121bf50 2024.10.17 22:58:23.529346 [ 5233 ] {} BaseDaemon: 10. DB::ISource::tryGenerate() @ 0x00000000116f520c 2024.10.17 22:58:23.529653 [ 5233 ] {} BaseDaemon: 11. DB::ISource::work() @ 0x00000000116f4c74 2024.10.17 22:58:23.529679 [ 5233 ] {} BaseDaemon: 12. DB::ExecutionThreadContext::executeTask() @ 0x000000001170a150 2024.10.17 22:58:23.529733 [ 5233 ] {} BaseDaemon: 13. DB::PipelineExecutor::executeStepImpl(unsigned long, std::atomic*) @ 0x00000000117009f0 2024.10.17 22:58:23.529763 [ 5233 ] {} BaseDaemon: 14. DB::PipelineExecutor::executeStep(std::atomic*) @ 0x0000000011700574 2024.10.17 22:58:23.530089 [ 5233 ] {} BaseDaemon: 15. DB::PullingPipelineExecutor::pull(DB::Chunk&) @ 0x000000001170e364 2024.10.17 22:58:23.530277 [ 5233 ] {} BaseDaemon: 16. DB::PullingPipelineExecutor::pull(DB::Block&) @ 0x000000001170e4fc 2024.10.17 22:58:23.530295 [ 5233 ] {} BaseDaemon: 17. DB::MergeTask::ExecuteAndFinalizeHorizontalPart::executeImpl() @ 0x0000000011074328 2024.10.17 22:58:23.530318 [ 5233 ] {} BaseDaemon: 18. DB::MergeTask::ExecuteAndFinalizeHorizontalPart::execute() @ 0x000000001107428c 2024.10.17 22:58:23.530339 [ 5233 ] {} BaseDaemon: 19. DB::MergeTask::execute() @ 0x0000000011077df0 2024.10.17 22:58:23.530362 [ 5233 ] {} BaseDaemon: 20. DB::SharedMergeMutateTaskBase::executeStep() @ 0x0000000011435a3c 2024.10.17 22:58:23.530384 [ 5233 ] {} BaseDaemon: 21. DB::MergeTreeBackgroundExecutor::threadFunction() @ 0x000000001108b234 2024.10.17 22:58:23.530410 [ 5233 ] {} BaseDaemon: 22. ThreadPoolImpl>::worker(std::__list_iterator, void*>) @ 0x000000000c52e264 2024.10.17 22:58:23.530448 [ 5233 ] {} BaseDaemon: 23. void std::__function::__policy_invoker::__call_impl::ThreadFromGlobalPoolImpl>::scheduleImpl(std::function, Priority, std::optional, bool)::'lambda0'()>(void&&)::'lambda'(), void ()>>(std::__function::__policy_storage const*) @ 0x000000000c531dd0 2024.10.17 22:58:23.530476 [ 5233 ] {} BaseDaemon: 24. void* std::__thread_proxy[abi:v15000]>, void ThreadPoolImpl::scheduleImpl(std::function, Priority, std::optional, bool)::'lambda0'()>>(void*) @ 0x000000000c530a80 2024.10.17 22:58:23.530514 [ 5233 ] {} BaseDaemon: 25. ? @ 0x000000000007d5c8 2024.10.17 22:58:23.530534 [ 5233 ] {} BaseDaemon: 26. ? @ 0x00000000000e5edc 2024.10.17 22:58:23.530551 [ 5233 ] {} BaseDaemon: Integrity check of the executable skipped because the reference checksum could not be read. 2024.10.17 22:58:23.531083 [ 5233 ] {} BaseDaemon: Report this error to https://github.com/ClickHouse/ClickHouse/issues 2024.10.17 22:58:23.531294 [ 5233 ] {} BaseDaemon: Changed settings: max_insert_threads = 4, max_threads = 42, use_hedged_requests = false, distributed_foreground_insert = true, alter_sync = 0, enable_memory_bound_merging_of_aggregation_results = true, cluster_for_parallel_replicas = 'default', do_not_merge_across_partitions_select_final = false, log_queries = true, log_queries_probability = 1., max_http_get_redirects = 10, enable_deflate_qpl_codec = false, enable_zstd_qat_codec = false, query_profiler_real_time_period_ns = 0, query_profiler_cpu_time_period_ns = 0, max_bytes_before_external_group_by = 90194313216, max_bytes_before_external_sort = 90194313216, max_memory_usage = 180388626432, backup_restore_keeper_retry_max_backoff_ms = 60000, cancel_http_readonly_queries_on_client_close = true, max_table_size_to_drop = 1000000000000, max_partition_size_to_drop = 1000000000000, default_table_engine = 'ReplicatedMergeTree', mutations_sync = 0, optimize_trivial_insert_select = false, database_replicated_allow_only_replicated_engine = true, cloud_mode = true, cloud_mode_engine = 2, distributed_ddl_output_mode = 'none_only_active', distributed_ddl_entry_format_version = 6, async_insert_max_data_size = 10485760, async_insert_busy_timeout_max_ms = 1000, enable_filesystem_cache_on_write_operations = true, load_marks_asynchronously = true, allow_prefetched_read_pool_for_remote_filesystem = true, filesystem_prefetch_max_memory_usage = 18038862643, filesystem_prefetches_limit = 200, compatibility = '24.6', insert_keeper_max_retries = 20, allow_experimental_materialized_postgresql_table = false, date_time_input_format = 'best_effort' ```. [#70820](https://github.com/ClickHouse/ClickHouse/pull/70820) ([Michael Kolupaev](https://github.com/al13n321)). +* Disable enable_named_columns_in_function_tuple by default. [#70833](https://github.com/ClickHouse/ClickHouse/pull/70833) ([Raúl Marín](https://github.com/Algunenano)). +* Fix S3Queue table engine setting processing_threads_num not being effective in case it was deduced from the number of cpu cores on the server. [#70837](https://github.com/ClickHouse/ClickHouse/pull/70837) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Normalize named tuple arguments in aggregation states. This fixes [#69732](https://github.com/ClickHouse/ClickHouse/issues/69732) . [#70853](https://github.com/ClickHouse/ClickHouse/pull/70853) ([Amos Bird](https://github.com/amosbird)). +* Fix a logical error due to negative zeros in the two-level hash table. This closes [#70973](https://github.com/ClickHouse/ClickHouse/issues/70973). [#70979](https://github.com/ClickHouse/ClickHouse/pull/70979) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#71214](https://github.com/ClickHouse/ClickHouse/issues/71214): Fix logical error in `StorageS3Queue` "Cannot create a persistent node in /processed since it already exists". [#70984](https://github.com/ClickHouse/ClickHouse/pull/70984) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#71243](https://github.com/ClickHouse/ClickHouse/issues/71243): Fixed named sessions not being closed and hanging on forever under certain circumstances. [#70998](https://github.com/ClickHouse/ClickHouse/pull/70998) ([Márcio Martins](https://github.com/marcio-absmartly)). +* Backported in [#71157](https://github.com/ClickHouse/ClickHouse/issues/71157): Fix the bug that didn't consider _row_exists column in rebuild option of projection lightweight delete. [#71089](https://github.com/ClickHouse/ClickHouse/pull/71089) ([Shichao Jin](https://github.com/jsc0218)). +* Backported in [#71265](https://github.com/ClickHouse/ClickHouse/issues/71265): Fix wrong value in system.query_metric_log due to unexpected race condition. [#71124](https://github.com/ClickHouse/ClickHouse/pull/71124) ([Pablo Marcos](https://github.com/pamarcos)). +* Backported in [#71331](https://github.com/ClickHouse/ClickHouse/issues/71331): Fix async inserts with empty blocks via native protocol. [#71312](https://github.com/ClickHouse/ClickHouse/pull/71312) ([Anton Popov](https://github.com/CurtizJ)). + +#### Build/Testing/Packaging Improvement +* Docker in integration tests runner is updated to latest version. It was previously pinned u until patch release 24.0.3 was out. https://github.com/moby/moby/issues/45770#issuecomment-1618255130. - HDFS image was deprecated and not running with current docker version. Switched to newer version of a derivative image based on ubuntu. - HDFS tests were hardened to allow them to run with python-repeat. [#66867](https://github.com/ClickHouse/ClickHouse/pull/66867) ([Ilya Yatsishin](https://github.com/qoega)). +* Alpine docker images now use ubuntu 22.04 as glibc donor, results in upgrade of glibc version delivered with alpine images from 2.31 to 2.35. [#69033](https://github.com/ClickHouse/ClickHouse/pull/69033) ([filimonov](https://github.com/filimonov)). +* Makes dbms independent from clickhouse_functions. [#69914](https://github.com/ClickHouse/ClickHouse/pull/69914) ([Raúl Marín](https://github.com/Algunenano)). +* Fix FreeBSD compilation of the MariaDB connector. [#70007](https://github.com/ClickHouse/ClickHouse/pull/70007) ([Raúl Marín](https://github.com/Algunenano)). +* Building on Apple Mac OS X Darwin does not produce strange warnings anymore. [#70411](https://github.com/ClickHouse/ClickHouse/pull/70411) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix building with ARCH_NATIVE CMake flag. [#70585](https://github.com/ClickHouse/ClickHouse/pull/70585) ([Daniil Gentili](https://github.com/danog)). +* The universal installer will download Musl build on Alpine Linux. Some Docker containers are using Alpine Linux, but it was not possible to install ClickHouse there with `curl https://clickhouse.com/ | sh`. [#70767](https://github.com/ClickHouse/ClickHouse/pull/70767) ([Alexey Milovidov](https://github.com/alexey-milovidov)). + +#### NO CL CATEGORY + +* Backported in [#71259](https://github.com/ClickHouse/ClickHouse/issues/71259):. [#71220](https://github.com/ClickHouse/ClickHouse/pull/71220) ([Raúl Marín](https://github.com/Algunenano)). + +#### NO CL ENTRY + +* NO CL ENTRY: 'Revert "JSONCompactWithProgress query output format"'. [#69989](https://github.com/ClickHouse/ClickHouse/pull/69989) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* NO CL ENTRY: 'Revert "Support CREATE OR REPLACE VIEW atomically"'. [#70535](https://github.com/ClickHouse/ClickHouse/pull/70535) ([Raúl Marín](https://github.com/Algunenano)). +* NO CL ENTRY: 'Revert "Revert "Support CREATE OR REPLACE VIEW atomically""'. [#70536](https://github.com/ClickHouse/ClickHouse/pull/70536) ([Raúl Marín](https://github.com/Algunenano)). +* NO CL ENTRY: 'Revert "Add projections size to system.projections"'. [#70858](https://github.com/ClickHouse/ClickHouse/pull/70858) ([Alexey Milovidov](https://github.com/alexey-milovidov)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Allow writing argument of `has` or `hasAny` or `hasAll` as string values if array element type is `Enum`. [#56555](https://github.com/ClickHouse/ClickHouse/pull/56555) ([Duc Canh Le](https://github.com/canhld94)). +* Rename FileSegmentKind::Ephemeral and other changes. [#66600](https://github.com/ClickHouse/ClickHouse/pull/66600) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Closes [#67345](https://github.com/ClickHouse/ClickHouse/issues/67345). [#67346](https://github.com/ClickHouse/ClickHouse/pull/67346) ([KrJin](https://github.com/jincong8973)). +* Because it is too complicated to support. [#68410](https://github.com/ClickHouse/ClickHouse/pull/68410) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Fix 01600_parts_states_metrics_long flakiness. [#68521](https://github.com/ClickHouse/ClickHouse/pull/68521) ([Azat Khuzhin](https://github.com/azat)). +* Reduce client start time in debug/sanitizer mode. [#68980](https://github.com/ClickHouse/ClickHouse/pull/68980) ([Raúl Marín](https://github.com/Algunenano)). +* Closes [#69038](https://github.com/ClickHouse/ClickHouse/issues/69038). [#69040](https://github.com/ClickHouse/ClickHouse/pull/69040) ([Nikolay Degterinsky](https://github.com/evillique)). +* Better exception for unsupported full_text index with non-full parts. [#69067](https://github.com/ClickHouse/ClickHouse/pull/69067) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Catch additional zk connection erros while creating table and make sure to cleanup dirs if necessary for retries. [#69093](https://github.com/ClickHouse/ClickHouse/pull/69093) ([Sumit](https://github.com/sum12)). +* Update version_date.tsv and changelog after v24.7.5.37-stable. [#69185](https://github.com/ClickHouse/ClickHouse/pull/69185) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* DOCS: Replace live view with refreshable since the former is deprecated. [#69392](https://github.com/ClickHouse/ClickHouse/pull/69392) ([Damian Kula](https://github.com/heavelock)). +* Update ORC to the current HEAD. [#69473](https://github.com/ClickHouse/ClickHouse/pull/69473) ([Nikita Taranov](https://github.com/nickitat)). +* Make a test ready for flaky check. [#69586](https://github.com/ClickHouse/ClickHouse/pull/69586) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Support antlr parser to parse sql with some keywords as alias, make the behaviour same as the clickhouse-server - remove redundant `for` in the `keyword` field. [#69614](https://github.com/ClickHouse/ClickHouse/pull/69614) ([Z.H.](https://github.com/onlyacat)). +* Allow default implementations for null in function mapFromArrays for spark compatiability in apache gluten. Current change doesn't have any side effects on clickhouse in theory. [#69715](https://github.com/ClickHouse/ClickHouse/pull/69715) ([李扬](https://github.com/taiyang-li)). +* Fix exception message in AzureBlobStorage. [#69728](https://github.com/ClickHouse/ClickHouse/pull/69728) ([Pavel Kruglov](https://github.com/Avogar)). +* Add test parsing s3 URL with a bucket name including a dot. [#69743](https://github.com/ClickHouse/ClickHouse/pull/69743) ([Kaushik Iska](https://github.com/iskakaushik)). +* Make `clang-tidy` happy. [#69765](https://github.com/ClickHouse/ClickHouse/pull/69765) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Prepare to enable `clang-tidy` `readability-else-after-return`. [#69768](https://github.com/ClickHouse/ClickHouse/pull/69768) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* S3Queue: support having deprecated settings to not fail server startup. [#69769](https://github.com/ClickHouse/ClickHouse/pull/69769) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Use only adaptive heuristic to choose task sizes for remote reading. [#69778](https://github.com/ClickHouse/ClickHouse/pull/69778) ([Nikita Taranov](https://github.com/nickitat)). +* Remove unused buggy code. [#69780](https://github.com/ClickHouse/ClickHouse/pull/69780) ([Raúl Marín](https://github.com/Algunenano)). +* Fix bugfix check. [#69789](https://github.com/ClickHouse/ClickHouse/pull/69789) ([Antonio Andelic](https://github.com/antonio2368)). +* Followup for [#63279](https://github.com/ClickHouse/ClickHouse/issues/63279). [#69790](https://github.com/ClickHouse/ClickHouse/pull/69790) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Update version after release. [#69816](https://github.com/ClickHouse/ClickHouse/pull/69816) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Update ext-dict-functions.md. [#69819](https://github.com/ClickHouse/ClickHouse/pull/69819) ([kurikuQwQ](https://github.com/kurikuQwQ)). +* Allow cyrillic characters in generated contributor names. [#69820](https://github.com/ClickHouse/ClickHouse/pull/69820) ([Raúl Marín](https://github.com/Algunenano)). +* CI: praktika integration 1. [#69822](https://github.com/ClickHouse/ClickHouse/pull/69822) ([Max Kainov](https://github.com/maxknv)). +* Fix `test_delayed_replica_failover`. [#69826](https://github.com/ClickHouse/ClickHouse/pull/69826) ([Antonio Andelic](https://github.com/antonio2368)). +* minor change, less conflicts. [#69830](https://github.com/ClickHouse/ClickHouse/pull/69830) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Improve error message DDLWorker.cpp. [#69835](https://github.com/ClickHouse/ClickHouse/pull/69835) ([Denny Crane](https://github.com/den-crane)). +* Fix typo in description: mutation_sync -> mutations_sync. [#69838](https://github.com/ClickHouse/ClickHouse/pull/69838) ([Alexander Gololobov](https://github.com/davenger)). +* Fix changelog. [#69841](https://github.com/ClickHouse/ClickHouse/pull/69841) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* This closes [#49940](https://github.com/ClickHouse/ClickHouse/issues/49940). [#69842](https://github.com/ClickHouse/ClickHouse/pull/69842) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* This closes [#51036](https://github.com/ClickHouse/ClickHouse/issues/51036). [#69844](https://github.com/ClickHouse/ClickHouse/pull/69844) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Update README.md - Update meetups. [#69849](https://github.com/ClickHouse/ClickHouse/pull/69849) ([Tanya Bragin](https://github.com/tbragin)). +* Revert [#69790](https://github.com/ClickHouse/ClickHouse/issues/69790) and [#63279](https://github.com/ClickHouse/ClickHouse/issues/63279). [#69850](https://github.com/ClickHouse/ClickHouse/pull/69850) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* See [#63279](https://github.com/ClickHouse/ClickHouse/issues/63279). [#69851](https://github.com/ClickHouse/ClickHouse/pull/69851) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add a test for [#50928](https://github.com/ClickHouse/ClickHouse/issues/50928). [#69852](https://github.com/ClickHouse/ClickHouse/pull/69852) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add a test for [#55981](https://github.com/ClickHouse/ClickHouse/issues/55981). [#69853](https://github.com/ClickHouse/ClickHouse/pull/69853) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add a test for [#56823](https://github.com/ClickHouse/ClickHouse/issues/56823). [#69854](https://github.com/ClickHouse/ClickHouse/pull/69854) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* This closes [#62350](https://github.com/ClickHouse/ClickHouse/issues/62350). [#69855](https://github.com/ClickHouse/ClickHouse/pull/69855) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Refactor functions and variables in statistics code. [#69860](https://github.com/ClickHouse/ClickHouse/pull/69860) ([Robert Schulze](https://github.com/rschu1ze)). +* Resubmit [#63279](https://github.com/ClickHouse/ClickHouse/issues/63279). [#69861](https://github.com/ClickHouse/ClickHouse/pull/69861) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Improve stateless test runner. [#69864](https://github.com/ClickHouse/ClickHouse/pull/69864) ([Alexey Katsman](https://github.com/alexkats)). +* Adjust fast test time limit a bit. [#69874](https://github.com/ClickHouse/ClickHouse/pull/69874) ([Raúl Marín](https://github.com/Algunenano)). +* Add initial 24.9 CHANGELOG. [#69876](https://github.com/ClickHouse/ClickHouse/pull/69876) ([Raúl Marín](https://github.com/Algunenano)). +* Fix test `01278_random_string_utf8`. [#69878](https://github.com/ClickHouse/ClickHouse/pull/69878) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Fix minor fuzzer issue with experimental statistics. [#69881](https://github.com/ClickHouse/ClickHouse/pull/69881) ([Robert Schulze](https://github.com/rschu1ze)). +* Fix linking after settings refactoring. [#69882](https://github.com/ClickHouse/ClickHouse/pull/69882) ([Robert Schulze](https://github.com/rschu1ze)). +* Add Proj Obsolete Setting. [#69883](https://github.com/ClickHouse/ClickHouse/pull/69883) ([Shichao Jin](https://github.com/jsc0218)). +* Improve remote queries startup time. [#69884](https://github.com/ClickHouse/ClickHouse/pull/69884) ([Igor Nikonov](https://github.com/devcrafter)). +* Revert "Merge pull request [#69032](https://github.com/ClickHouse/ClickHouse/issues/69032) from alexon1234/include_real_time_execution_in_http_header". [#69885](https://github.com/ClickHouse/ClickHouse/pull/69885) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* A dedicated commits from https://github.com/ClickHouse/ClickHouse/pull/61473. [#69896](https://github.com/ClickHouse/ClickHouse/pull/69896) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Added aliases `time_bucket`(from TimescaleDB) and `date_bin`(from PostgreSQL) for `toStartOfInterval`. [#69900](https://github.com/ClickHouse/ClickHouse/pull/69900) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* RIPE is an acronym and thus should be capital. RIPE stands for **R**ACE **I**ntegrity **P**rimitives **E**valuation and RACE stands for **R**esearch and Development in **A**dvanced **C**ommunications **T**echnologies in **E**urope. [#69901](https://github.com/ClickHouse/ClickHouse/pull/69901) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Replace error codes with error names in stateless tests. [#69906](https://github.com/ClickHouse/ClickHouse/pull/69906) ([Dmitry Novik](https://github.com/novikd)). +* Move setting to 24.10. [#69913](https://github.com/ClickHouse/ClickHouse/pull/69913) ([Raúl Marín](https://github.com/Algunenano)). +* Minor: Reduce diff between public and private repo. [#69928](https://github.com/ClickHouse/ClickHouse/pull/69928) ([Robert Schulze](https://github.com/rschu1ze)). +* Followup for [#69861](https://github.com/ClickHouse/ClickHouse/issues/69861). [#69930](https://github.com/ClickHouse/ClickHouse/pull/69930) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Fix test_dictionaries_all_layouts_separate_sources. [#69962](https://github.com/ClickHouse/ClickHouse/pull/69962) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Fix test_keeper_mntr_data_size. [#69965](https://github.com/ClickHouse/ClickHouse/pull/69965) ([Antonio Andelic](https://github.com/antonio2368)). +* This closes [#49823](https://github.com/ClickHouse/ClickHouse/issues/49823). [#69981](https://github.com/ClickHouse/ClickHouse/pull/69981) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add changelog for 24.9. [#69982](https://github.com/ClickHouse/ClickHouse/pull/69982) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add a test for [#45303](https://github.com/ClickHouse/ClickHouse/issues/45303). [#69987](https://github.com/ClickHouse/ClickHouse/pull/69987) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Update CHANGELOG.md. [#69988](https://github.com/ClickHouse/ClickHouse/pull/69988) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Update README.md. [#69991](https://github.com/ClickHouse/ClickHouse/pull/69991) ([Tyler Hannan](https://github.com/tylerhannan)). +* Disable `03215_parallel_replicas_crash_after_refactoring.sql` for Azure. [#69992](https://github.com/ClickHouse/ClickHouse/pull/69992) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Update CHANGELOG.md. [#69993](https://github.com/ClickHouse/ClickHouse/pull/69993) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Update CHANGELOG.md. [#70004](https://github.com/ClickHouse/ClickHouse/pull/70004) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Revert "Add RIPEMD160 function". [#70005](https://github.com/ClickHouse/ClickHouse/pull/70005) ([Robert Schulze](https://github.com/rschu1ze)). +* Update CHANGELOG.md. [#70009](https://github.com/ClickHouse/ClickHouse/pull/70009) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Update CHANGELOG.md. [#70010](https://github.com/ClickHouse/ClickHouse/pull/70010) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Make the pylint stricter. [#70013](https://github.com/ClickHouse/ClickHouse/pull/70013) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Added a setting `restore_replace_external_dictionary_source_to_null` which enables replacing dictionary source with Null on restore for external dictionaries (useful for testing). [#70032](https://github.com/ClickHouse/ClickHouse/pull/70032) ([Alexander Tokmakov](https://github.com/tavplubix)). +* `isort` is a simple import sorter for the python to comply [pep-8](https://peps.python.org/pep-0008/#imports) requirements. It will allow to decrease conflicts during sync and beautify the code. The import block is divided into three sub-blocks: `standard library` -> `third-party libraries` -> `local imports` -> `.local imports`. Each sub-block is ordered alphabetically with sub-sub-blocks `import X` -> `from X import Y`. [#70038](https://github.com/ClickHouse/ClickHouse/pull/70038) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Update version_date.tsv and changelog after v24.9.1.3278-stable. [#70049](https://github.com/ClickHouse/ClickHouse/pull/70049) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Despite the fact that we set the org-level workflow parameter `PYTHONUNBUFFERED`, it's not inherited in workflows. [#70050](https://github.com/ClickHouse/ClickHouse/pull/70050) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Fix ubsan issue in function sqid. [#70061](https://github.com/ClickHouse/ClickHouse/pull/70061) ([Robert Schulze](https://github.com/rschu1ze)). +* Delete a setting change. [#70071](https://github.com/ClickHouse/ClickHouse/pull/70071) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Fix `test_distributed_ddl`. [#70075](https://github.com/ClickHouse/ClickHouse/pull/70075) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Remove unused placeholder from exception message string. [#70086](https://github.com/ClickHouse/ClickHouse/pull/70086) ([Alsu Giliazova](https://github.com/alsugiliazova)). +* Better exception message when some of the permission is missing. [#70088](https://github.com/ClickHouse/ClickHouse/pull/70088) ([pufit](https://github.com/pufit)). +* Make vector similarity indexes work with adaptive granularity. [#70101](https://github.com/ClickHouse/ClickHouse/pull/70101) ([Robert Schulze](https://github.com/rschu1ze)). +* Add missing columns `total_rows`, `data_compressed_bytes`, and `data_uncompressed_bytes` to `system.projections`. Part of https://github.com/ClickHouse/ClickHouse/pull/68901. [#70106](https://github.com/ClickHouse/ClickHouse/pull/70106) ([Jordi Villar](https://github.com/jrdi)). +* Make `00938_fix_rwlock_segfault_long` non flaky. [#70109](https://github.com/ClickHouse/ClickHouse/pull/70109) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Remove TODO. [#70110](https://github.com/ClickHouse/ClickHouse/pull/70110) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Change the default threshold to enable hyper threading. [#70111](https://github.com/ClickHouse/ClickHouse/pull/70111) ([Jiebin Sun](https://github.com/jiebinn)). +* Fixed [#69092](https://github.com/ClickHouse/ClickHouse/issues/69092): if `materialized_postgresql_tables_list=table1(id, code),table(id,name)` (`table1` has name that is a substring for `table`) `getTableAllowedColumns` method returns `[id, code]` for `table` before this fix. [#70114](https://github.com/ClickHouse/ClickHouse/pull/70114) ([Kruglov Kirill](https://github.com/1on)). +* Reduce log level. [#70117](https://github.com/ClickHouse/ClickHouse/pull/70117) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Rename `getNumberOfPhysicalCPUCores` and fix its decription. [#70130](https://github.com/ClickHouse/ClickHouse/pull/70130) ([Nikita Taranov](https://github.com/nickitat)). +* Adding 24.10. [#70132](https://github.com/ClickHouse/ClickHouse/pull/70132) ([Tyler Hannan](https://github.com/tylerhannan)). +* (Re?)-enable libcxx asserts for debug builds. [#70134](https://github.com/ClickHouse/ClickHouse/pull/70134) ([Robert Schulze](https://github.com/rschu1ze)). +* Refactor reading from object storage. [#70141](https://github.com/ClickHouse/ClickHouse/pull/70141) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Silence UBSAN for integer overflows in some datetime functions. [#70142](https://github.com/ClickHouse/ClickHouse/pull/70142) ([Michael Kolupaev](https://github.com/al13n321)). +* Improve pipdeptree generator for docker images. - Update requirements.txt for the integration tests runner container - Remove some small dependencies, improve `helpers/retry_decorator.py` - Upgrade docker-compose from EOL version 1 to version 2. [#70146](https://github.com/ClickHouse/ClickHouse/pull/70146) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Fix 'QueryPlan was not initialized' in 'loop' with empty MergeTree. [#70149](https://github.com/ClickHouse/ClickHouse/pull/70149) ([Michael Kolupaev](https://github.com/al13n321)). +* Remove QueryPlan DataStream. [#70158](https://github.com/ClickHouse/ClickHouse/pull/70158) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Update test_storage_s3_queue/test.py. [#70159](https://github.com/ClickHouse/ClickHouse/pull/70159) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Small docs fix. [#70160](https://github.com/ClickHouse/ClickHouse/pull/70160) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Test: PR local plan, non-constant in source stream. [#70173](https://github.com/ClickHouse/ClickHouse/pull/70173) ([Igor Nikonov](https://github.com/devcrafter)). +* Fix performance checks. [#70175](https://github.com/ClickHouse/ClickHouse/pull/70175) ([Antonio Andelic](https://github.com/antonio2368)). +* Simplify test 03246_range_literal_replacement_works. [#70176](https://github.com/ClickHouse/ClickHouse/pull/70176) ([Pablo Marcos](https://github.com/pamarcos)). +* Update 01079_parallel_alter_add_drop_column_zookeeper.sh. [#70196](https://github.com/ClickHouse/ClickHouse/pull/70196) ([Alexander Tokmakov](https://github.com/tavplubix)). +* Require bugfix job for a set of labels. [#70197](https://github.com/ClickHouse/ClickHouse/pull/70197) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* CI: Praktika integration, fast test. [#70239](https://github.com/ClickHouse/ClickHouse/pull/70239) ([Max Kainov](https://github.com/maxknv)). +* Avoid `Cannot schedule a task` error when loading parts. [#70257](https://github.com/ClickHouse/ClickHouse/pull/70257) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Bump usearch to v2.15.2 and SimSIMD to v5.0.0. [#70270](https://github.com/ClickHouse/ClickHouse/pull/70270) ([Robert Schulze](https://github.com/rschu1ze)). +* Instead of balancing tests by `crc32(file_name)` we'll use `add tests to a group with a minimal number of tests`. [#70272](https://github.com/ClickHouse/ClickHouse/pull/70272) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Closes [#70263](https://github.com/ClickHouse/ClickHouse/issues/70263). [#70273](https://github.com/ClickHouse/ClickHouse/pull/70273) ([flynn](https://github.com/ucasfl)). +* Hide MergeTreeSettings implementation. [#70285](https://github.com/ClickHouse/ClickHouse/pull/70285) ([Raúl Marín](https://github.com/Algunenano)). +* CI: Remove await feature from release branches. [#70294](https://github.com/ClickHouse/ClickHouse/pull/70294) ([Max Kainov](https://github.com/maxknv)). +* Fix `test_keeper_four_word_command`. [#70298](https://github.com/ClickHouse/ClickHouse/pull/70298) ([Antonio Andelic](https://github.com/antonio2368)). +* Update version_date.tsv and changelog after v24.9.2.42-stable. [#70301](https://github.com/ClickHouse/ClickHouse/pull/70301) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Synchronize settings with private. [#70320](https://github.com/ClickHouse/ClickHouse/pull/70320) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Add Ignore Option In DeduplicateMergeProjectionMode. [#70327](https://github.com/ClickHouse/ClickHouse/pull/70327) ([Shichao Jin](https://github.com/jsc0218)). +* CI: Enable Integration Tests for backport PRs. [#70329](https://github.com/ClickHouse/ClickHouse/pull/70329) ([Max Kainov](https://github.com/maxknv)). +* There is [a failed CI job](https://s3.amazonaws.com/clickhouse-test-reports/69778/2d81c38874958bd9d54a25524173bdb1ddf2b75c/stateless_tests__release_.html) which is triggered by [03237_create_or_replace_view_atomically_with_atomic_engine](https://github.com/ClickHouse/ClickHouse/blob/master/tests/queries/0_stateless/03237_create_or_replace_view_atomically_with_atomic_engine.sh). [#70330](https://github.com/ClickHouse/ClickHouse/pull/70330) ([tuanpach](https://github.com/tuanpach)). +* Fix flaky test `03237_insert_sparse_columns_mem`. [#70333](https://github.com/ClickHouse/ClickHouse/pull/70333) ([Anton Popov](https://github.com/CurtizJ)). +* Rename enable_secure_identifiers -> enforce_strict_identifier_format. [#70335](https://github.com/ClickHouse/ClickHouse/pull/70335) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Attempt to fix flaky RabbitMQ tests. Maybe closes [#45160](https://github.com/ClickHouse/ClickHouse/issues/45160). [#70336](https://github.com/ClickHouse/ClickHouse/pull/70336) ([filimonov](https://github.com/filimonov)). +* Don't fail the stateless check script if we can't collect minio logs. [#70350](https://github.com/ClickHouse/ClickHouse/pull/70350) ([Raúl Marín](https://github.com/Algunenano)). +* Fix tiny mistake, responsible for some of kafka test flaps. Example [report](https://s3.amazonaws.com/clickhouse-test-reports/0/3198aafac59c368993e7b5f49d95674cc1b1be18/integration_tests__release__[2_4].html). [#70352](https://github.com/ClickHouse/ClickHouse/pull/70352) ([filimonov](https://github.com/filimonov)). +* Closes [#69634](https://github.com/ClickHouse/ClickHouse/issues/69634). [#70354](https://github.com/ClickHouse/ClickHouse/pull/70354) ([pufit](https://github.com/pufit)). +* Fix 02346_fulltext_index_bug52019. [#70357](https://github.com/ClickHouse/ClickHouse/pull/70357) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Use new JSON for collecting minio logs. [#70359](https://github.com/ClickHouse/ClickHouse/pull/70359) ([Antonio Andelic](https://github.com/antonio2368)). +* Update comments in VectorSimilarityCondition (WHERE is not supported). [#70360](https://github.com/ClickHouse/ClickHouse/pull/70360) ([Azat Khuzhin](https://github.com/azat)). +* Remove 02492_clickhouse_local_context_uaf test. [#70363](https://github.com/ClickHouse/ClickHouse/pull/70363) ([Azat Khuzhin](https://github.com/azat)). +* Fix `clang-19` build issues. [#70412](https://github.com/ClickHouse/ClickHouse/pull/70412) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Ignore "Invalid multibyte data detected" error during completion. [#70422](https://github.com/ClickHouse/ClickHouse/pull/70422) ([Azat Khuzhin](https://github.com/azat)). +* Make QueryPlan explain methods const. [#70444](https://github.com/ClickHouse/ClickHouse/pull/70444) ([Alexander Gololobov](https://github.com/davenger)). +* Fix 0.1 second delay for interactive queries (due to keystroke interceptor). [#70445](https://github.com/ClickHouse/ClickHouse/pull/70445) ([Azat Khuzhin](https://github.com/azat)). +* Increase lock timeout in attempt to fix 02125_many_mutations. [#70448](https://github.com/ClickHouse/ClickHouse/pull/70448) ([Azat Khuzhin](https://github.com/azat)). +* Fix order in 03249_dynamic_alter_consistency. [#70453](https://github.com/ClickHouse/ClickHouse/pull/70453) ([Alexander Gololobov](https://github.com/davenger)). +* Fix refreshable MV in system database breaking server startup. [#70460](https://github.com/ClickHouse/ClickHouse/pull/70460) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix flaky test_refreshable_mv_in_replicated_db. [#70462](https://github.com/ClickHouse/ClickHouse/pull/70462) ([Michael Kolupaev](https://github.com/al13n321)). +* Update version_date.tsv and changelog after v24.8.5.115-lts. [#70463](https://github.com/ClickHouse/ClickHouse/pull/70463) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Decrease probability of "Server died" due to 00913_many_threads. [#70473](https://github.com/ClickHouse/ClickHouse/pull/70473) ([Azat Khuzhin](https://github.com/azat)). +* Fixes for killing leftovers in clikhouse-test. [#70474](https://github.com/ClickHouse/ClickHouse/pull/70474) ([Azat Khuzhin](https://github.com/azat)). +* Update version_date.tsv and changelog after v24.3.12.75-lts. [#70485](https://github.com/ClickHouse/ClickHouse/pull/70485) ([robot-clickhouse](https://github.com/robot-clickhouse)). +* Use logging instead of print. [#70505](https://github.com/ClickHouse/ClickHouse/pull/70505) ([János Benjamin Antal](https://github.com/antaljanosbenjamin)). +* Remove slow poll() logs in keeper. [#70508](https://github.com/ClickHouse/ClickHouse/pull/70508) ([Raúl Marín](https://github.com/Algunenano)). +* Add timeouts for retry loops in test_storage_rabbitmq. It should prevent cascading failures of the whole test suite caused by deadloop in one of the test scenarios. Also added small sleeps in a 'tight' loops to make retries bit less agressive. [#70510](https://github.com/ClickHouse/ClickHouse/pull/70510) ([filimonov](https://github.com/filimonov)). +* CI: Fix for canceled Sync workflow. [#70521](https://github.com/ClickHouse/ClickHouse/pull/70521) ([Max Kainov](https://github.com/maxknv)). +* Debug build faild with clang-18 after https://github.com/ClickHouse/ClickHouse/pull/70412, don't know why it's ok in release build, simply changing `_` to `_1` is ok for both release and debug build. [#70532](https://github.com/ClickHouse/ClickHouse/pull/70532) ([Chang chen](https://github.com/baibaichen)). +* Refreshable materialized views are not experimental anymore. [#70550](https://github.com/ClickHouse/ClickHouse/pull/70550) ([Michael Kolupaev](https://github.com/al13n321)). +* Fix 24.9 setting compatibility `database_replicated_allow_explicit_uuid`. [#70565](https://github.com/ClickHouse/ClickHouse/pull/70565) ([Nikita Fomichev](https://github.com/fm4v)). +* Fix typos. [#70588](https://github.com/ClickHouse/ClickHouse/pull/70588) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Vector search: allow to specify HNSW parameter `ef_search` at query time. [#70616](https://github.com/ClickHouse/ClickHouse/pull/70616) ([Robert Schulze](https://github.com/rschu1ze)). +* Increase max_rows_to_read limit in some tests. [#70617](https://github.com/ClickHouse/ClickHouse/pull/70617) ([Raúl Marín](https://github.com/Algunenano)). +* Reduce sync efforts with private. [#70634](https://github.com/ClickHouse/ClickHouse/pull/70634) ([Raúl Marín](https://github.com/Algunenano)). +* Fix parsing of some formats into sparse columns. [#70635](https://github.com/ClickHouse/ClickHouse/pull/70635) ([Anton Popov](https://github.com/CurtizJ)). +* Fix typos. [#70637](https://github.com/ClickHouse/ClickHouse/pull/70637) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Try fix 00180_no_seek_avoiding_when_reading_from_cache. [#70640](https://github.com/ClickHouse/ClickHouse/pull/70640) ([Kseniia Sumarokova](https://github.com/kssenii)). +* When the `PR Check` status is set, it's a valid RunConfig job failure. [#70643](https://github.com/ClickHouse/ClickHouse/pull/70643) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Fix timeout in materialized pg tests. [#70646](https://github.com/ClickHouse/ClickHouse/pull/70646) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Introduced MergeTree setting which allow to change merge selecting algorithm. However we still have only one algorithm and it's mostly for future experiments. [#70647](https://github.com/ClickHouse/ClickHouse/pull/70647) ([alesapin](https://github.com/alesapin)). +* Docs: Follow-up for [#70585](https://github.com/ClickHouse/ClickHouse/issues/70585). [#70654](https://github.com/ClickHouse/ClickHouse/pull/70654) ([Robert Schulze](https://github.com/rschu1ze)). +* Remove strange file. [#70662](https://github.com/ClickHouse/ClickHouse/pull/70662) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Locally I had lots of errors like `'AllocList' does not refer to a value` around places which used `offsetof`. Changing it to `__builtin_offsetof ` helped and I didn't debug any further. [#70671](https://github.com/ClickHouse/ClickHouse/pull/70671) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)). +* Adding the report link to a test result and files' list. [#70677](https://github.com/ClickHouse/ClickHouse/pull/70677) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* materialized postgres: minor fixes. [#70710](https://github.com/ClickHouse/ClickHouse/pull/70710) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Probably fix flaky test_refreshable_mv_in_replicated_db. [#70714](https://github.com/ClickHouse/ClickHouse/pull/70714) ([Michael Kolupaev](https://github.com/al13n321)). +* Move more setting structs to pImpl. [#70739](https://github.com/ClickHouse/ClickHouse/pull/70739) ([Raúl Marín](https://github.com/Algunenano)). +* Reduce sync effort. [#70747](https://github.com/ClickHouse/ClickHouse/pull/70747) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#71198](https://github.com/ClickHouse/ClickHouse/issues/71198): Check number of arguments for function with Dynamic argument. [#70749](https://github.com/ClickHouse/ClickHouse/pull/70749) ([Nikita Taranov](https://github.com/nickitat)). +* Add s3queue settings check for cloud. [#70750](https://github.com/ClickHouse/ClickHouse/pull/70750) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix readiness/health check for OpenLDAP container. [#70755](https://github.com/ClickHouse/ClickHouse/pull/70755) ([Julian Maicher](https://github.com/jmaicher)). +* Allow update plan headers for all the steps. [#70761](https://github.com/ClickHouse/ClickHouse/pull/70761) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Autogenerate documentation for settings. [#70768](https://github.com/ClickHouse/ClickHouse/pull/70768) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Not a logical error. [#70770](https://github.com/ClickHouse/ClickHouse/pull/70770) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* CI: Aarch64 build with Asan. [#70778](https://github.com/ClickHouse/ClickHouse/pull/70778) ([Max Kainov](https://github.com/maxknv)). +* Minor fix. [#70783](https://github.com/ClickHouse/ClickHouse/pull/70783) ([Anton Popov](https://github.com/CurtizJ)). +* The docs for settings should be located in the source code. Now, the CI supports that. [#70784](https://github.com/ClickHouse/ClickHouse/pull/70784) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Update style-test image. [#70785](https://github.com/ClickHouse/ClickHouse/pull/70785) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Avoid double finalization of `WriteBuffer` in library bridge. [#70799](https://github.com/ClickHouse/ClickHouse/pull/70799) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* Make Array Field serialization consistent. [#70803](https://github.com/ClickHouse/ClickHouse/pull/70803) ([Nikolai Kochetov](https://github.com/KochetovNicolai)). +* A follow-up for [#70785](https://github.com/ClickHouse/ClickHouse/issues/70785), [jwt](https://pypi.org/project/jwt/#history) looks very outdated, and we have issue with conflicting paths. [#70815](https://github.com/ClickHouse/ClickHouse/pull/70815) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Remove inneficient code. [#70816](https://github.com/ClickHouse/ClickHouse/pull/70816) ([Raúl Marín](https://github.com/Algunenano)). +* Allow large object files if OMIT_HEAVY_DEBUG_SYMBOLS = 0. [#70818](https://github.com/ClickHouse/ClickHouse/pull/70818) ([Michael Kolupaev](https://github.com/al13n321)). +* Add test with distributed queries for 15768. [#70834](https://github.com/ClickHouse/ClickHouse/pull/70834) ([Nikita Taranov](https://github.com/nickitat)). +* More setting structs to pImpl and reuse code. [#70840](https://github.com/ClickHouse/ClickHouse/pull/70840) ([Raúl Marín](https://github.com/Algunenano)). +* Update default HNSW parameter settings. [#70873](https://github.com/ClickHouse/ClickHouse/pull/70873) ([Robert Schulze](https://github.com/rschu1ze)). +* Limiting logging some lines about configs. [#70879](https://github.com/ClickHouse/ClickHouse/pull/70879) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* Fix `limit by`, `limit with ties` for distributed and parallel replicas. [#70880](https://github.com/ClickHouse/ClickHouse/pull/70880) ([Nikita Taranov](https://github.com/nickitat)). +* Fix darwin build. [#70894](https://github.com/ClickHouse/ClickHouse/pull/70894) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Add dots for consistency. [#70909](https://github.com/ClickHouse/ClickHouse/pull/70909) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Logical error fix for substrings, found by fuzzer. [#70914](https://github.com/ClickHouse/ClickHouse/pull/70914) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). +* More setting structs to pImpl. [#70942](https://github.com/ClickHouse/ClickHouse/pull/70942) ([Raúl Marín](https://github.com/Algunenano)). +* Add logging for mock HTTP servers used in minio integration tests. [#70943](https://github.com/ClickHouse/ClickHouse/pull/70943) ([Vitaly Baranov](https://github.com/vitlibar)). +* Minor fixups of [#70011](https://github.com/ClickHouse/ClickHouse/issues/70011) and [#69918](https://github.com/ClickHouse/ClickHouse/issues/69918). [#70959](https://github.com/ClickHouse/ClickHouse/pull/70959) ([Robert Schulze](https://github.com/rschu1ze)). +* CI: Do not skip Build report and status fix. [#70965](https://github.com/ClickHouse/ClickHouse/pull/70965) ([Max Kainov](https://github.com/maxknv)). +* Fix Keeper entry serialization compatibility. [#70972](https://github.com/ClickHouse/ClickHouse/pull/70972) ([Antonio Andelic](https://github.com/antonio2368)). +* Update exception message. [#70975](https://github.com/ClickHouse/ClickHouse/pull/70975) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Fix `utils/c++expr` option `-b`. [#70978](https://github.com/ClickHouse/ClickHouse/pull/70978) ([Sergei Trifonov](https://github.com/serxa)). +* Fix `test_keeper_broken_logs`. [#70982](https://github.com/ClickHouse/ClickHouse/pull/70982) ([Antonio Andelic](https://github.com/antonio2368)). +* Fix `01039_test_setting_parse`. [#70986](https://github.com/ClickHouse/ClickHouse/pull/70986) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Tests for languages support for Embedded Dictionaries. [#71004](https://github.com/ClickHouse/ClickHouse/pull/71004) ([Max Vostrikov](https://github.com/max-vostrikov)). +* Required for internal test runs with the same image build in public CI. [#71008](https://github.com/ClickHouse/ClickHouse/pull/71008) ([Ilya Yatsishin](https://github.com/qoega)). +* Move remaining settings objects to pImpl and start simplification. [#71019](https://github.com/ClickHouse/ClickHouse/pull/71019) ([Raúl Marín](https://github.com/Algunenano)). +* CI: Rearrange directories for praktika ci. [#71029](https://github.com/ClickHouse/ClickHouse/pull/71029) ([Max Kainov](https://github.com/maxknv)). +* Fix assert in RemoteSource::onAsyncJobReady(). [#71034](https://github.com/ClickHouse/ClickHouse/pull/71034) ([Igor Nikonov](https://github.com/devcrafter)). +* Fix showing error message in ReadBufferFromS3 when retrying. Without this PR information about a retryable failure in `ReadBufferFromS3` could look like this:. [#71038](https://github.com/ClickHouse/ClickHouse/pull/71038) ([Vitaly Baranov](https://github.com/vitlibar)). +* Fix `test_truncate_database`. [#71057](https://github.com/ClickHouse/ClickHouse/pull/71057) ([Antonio Andelic](https://github.com/antonio2368)). +* Fix clickhouse-test useless 5 second delay in case of multiple threads are used. [#71069](https://github.com/ClickHouse/ClickHouse/pull/71069) ([Azat Khuzhin](https://github.com/azat)). +* Backported in [#71142](https://github.com/ClickHouse/ClickHouse/issues/71142): Followup [#70520](https://github.com/ClickHouse/ClickHouse/issues/70520). [#71129](https://github.com/ClickHouse/ClickHouse/pull/71129) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Backported in [#71189](https://github.com/ClickHouse/ClickHouse/issues/71189): Update compatibility setting for `hnsw_candidate_list_size_for_search`. [#71133](https://github.com/ClickHouse/ClickHouse/pull/71133) ([Robert Schulze](https://github.com/rschu1ze)). +* Backported in [#71222](https://github.com/ClickHouse/ClickHouse/issues/71222): Fixes for interactive metrics. [#71173](https://github.com/ClickHouse/ClickHouse/pull/71173) ([Julia Kartseva](https://github.com/jkartseva)). +* Backported in [#71205](https://github.com/ClickHouse/ClickHouse/issues/71205): Maybe not GWPAsan by default. [#71174](https://github.com/ClickHouse/ClickHouse/pull/71174) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#71277](https://github.com/ClickHouse/ClickHouse/issues/71277): Fix LOGICAL_ERROR on wrong scalar subquery argument to table functions. [#71216](https://github.com/ClickHouse/ClickHouse/pull/71216) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#71253](https://github.com/ClickHouse/ClickHouse/issues/71253): Disable enable_named_columns_in_function_tuple for 24.10. [#71219](https://github.com/ClickHouse/ClickHouse/pull/71219) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#71303](https://github.com/ClickHouse/ClickHouse/issues/71303): Improve system.query_metric_log to remove flakiness. [#71295](https://github.com/ClickHouse/ClickHouse/pull/71295) ([Pablo Marcos](https://github.com/pamarcos)). +* Backported in [#71317](https://github.com/ClickHouse/ClickHouse/issues/71317): Fix debug log timestamp. [#71311](https://github.com/ClickHouse/ClickHouse/pull/71311) ([Pablo Marcos](https://github.com/pamarcos)). + +#### Not for changeling + +* Reverted. [#69812](https://github.com/ClickHouse/ClickHouse/pull/69812) ([tuanpach](https://github.com/tuanpach)). + diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 10c55aa4bf5..da7ad3ebd88 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -1,3 +1,4 @@ +v24.10.1.2812-stable 2024-11-01 v24.9.2.42-stable 2024-10-03 v24.9.1.3278-stable 2024-09-26 v24.8.5.115-lts 2024-10-08 From 9015454b37627712eac4eae5126378ae68d8e98c Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Fri, 1 Nov 2024 11:06:21 +0000 Subject: [PATCH 372/680] Add setting --- src/Core/Settings.cpp | 5 +++++ src/Core/SettingsChangesHistory.cpp | 1 + src/Interpreters/MutationsInterpreter.cpp | 6 ++++-- 3 files changed, 10 insertions(+), 2 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 6c269e22c35..17e2e1cc599 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -3640,6 +3640,11 @@ Given that, for example, dictionaries, can be out of sync across nodes, mutation ``` +)", 0) \ + DECLARE(Bool, validate_mutation_query, true, R"( +Validate mutation queries before accepting them. Mutations are executed in the background, and running an invalid query will cause mutations to get stuck, requiring manual intervention. + +Only change this setting if you encounter a backward-incompatible bug. )", 0) \ DECLARE(Seconds, lock_acquire_timeout, DBMS_DEFAULT_LOCK_ACQUIRE_TIMEOUT_SEC, R"( Defines how many seconds a locking request waits before failing. diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 3fe3e960dc6..613b9e2281a 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -64,6 +64,7 @@ static std::initializer_listgetSettingsRef()[Setting::validate_mutation_query]) + // Make sure the mutation query is valid + prepareQueryAffectedQueryTree(commands, source.getStorage(), context); QueryPlan plan; From 7691b7dd4435d1df5cd43cdf9169277aa9e81996 Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Fri, 1 Nov 2024 11:06:49 +0000 Subject: [PATCH 373/680] Fix test --- .../integration/test_failed_mutations/test.py | 32 +++++++------------ .../03256_invalid_mutation_query.sql | 2 ++ 2 files changed, 14 insertions(+), 20 deletions(-) diff --git a/tests/integration/test_failed_mutations/test.py b/tests/integration/test_failed_mutations/test.py index 5a2bf874da2..8d2ee46e748 100644 --- a/tests/integration/test_failed_mutations/test.py +++ b/tests/integration/test_failed_mutations/test.py @@ -27,6 +27,9 @@ REPLICATED_POSTPONE_MUTATION_LOG = ( POSTPONE_MUTATION_LOG = ( "According to exponential backoff policy, do not perform mutations for the part" ) +FAILING_MUTATION_QUERY = ( + "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT throwIf(1))" +) all_nodes = [node_with_backoff, node_no_backoff] @@ -83,17 +86,13 @@ def test_exponential_backoff_with_merge_tree(started_cluster, node, found_in_log assert not node.contains_in_log(POSTPONE_MUTATION_LOG) # Executing incorrect mutation. - node.query( - "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT x FROM notexist_table) SETTINGS allow_nondeterministic_mutations=1" - ) + node.query(FAILING_MUTATION_QUERY) check_logs() node.query("KILL MUTATION WHERE table='test_mutations'") # Check that after kill new parts mutations are postponing. - node.query( - "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT x FROM notexist_table) SETTINGS allow_nondeterministic_mutations=1" - ) + node.query(FAILING_MUTATION_QUERY) check_logs() @@ -101,9 +100,7 @@ def test_exponential_backoff_with_merge_tree(started_cluster, node, found_in_log def test_exponential_backoff_with_replicated_tree(started_cluster): prepare_cluster(True) - node_with_backoff.query( - "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT x FROM notexist_table) SETTINGS allow_nondeterministic_mutations=1" - ) + node_with_backoff.query(FAILING_MUTATION_QUERY) assert node_with_backoff.wait_for_log_line(REPLICATED_POSTPONE_MUTATION_LOG) assert not node_no_backoff.contains_in_log(REPLICATED_POSTPONE_MUTATION_LOG) @@ -114,7 +111,7 @@ def test_exponential_backoff_create_dependent_table(started_cluster): # Executing incorrect mutation. node_with_backoff.query( - "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT x FROM dep_table) SETTINGS allow_nondeterministic_mutations=1" + "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT x FROM dep_table) SETTINGS validate_mutation_query = 0" ) # Creating dependent table for mutation. @@ -148,9 +145,7 @@ def test_exponential_backoff_setting_override(started_cluster): node.query("INSERT INTO test_mutations SELECT * FROM system.numbers LIMIT 10") # Executing incorrect mutation. - node.query( - "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT x FROM dep_table) SETTINGS allow_nondeterministic_mutations=1" - ) + node.query(FAILING_MUTATION_QUERY) assert not node.contains_in_log(POSTPONE_MUTATION_LOG) @@ -166,9 +161,7 @@ def test_backoff_clickhouse_restart(started_cluster, replicated_table): node = node_with_backoff # Executing incorrect mutation. - node.query( - "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT x FROM dep_table) SETTINGS allow_nondeterministic_mutations=1" - ) + node.query(FAILING_MUTATION_QUERY) assert node.wait_for_log_line( REPLICATED_POSTPONE_MUTATION_LOG if replicated_table else POSTPONE_MUTATION_LOG ) @@ -193,11 +186,10 @@ def test_no_backoff_after_killing_mutation(started_cluster, replicated_table): node = node_with_backoff # Executing incorrect mutation. - node.query( - "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT x FROM dep_table) SETTINGS allow_nondeterministic_mutations=1" - ) + node.query(FAILING_MUTATION_QUERY) + # Executing correct mutation. - node.query("ALTER TABLE test_mutations DELETE WHERE x=1") + node.query("ALTER TABLE test_mutations DELETE WHERE x=1") assert node.wait_for_log_line( REPLICATED_POSTPONE_MUTATION_LOG if replicated_table else POSTPONE_MUTATION_LOG ) diff --git a/tests/queries/0_stateless/03256_invalid_mutation_query.sql b/tests/queries/0_stateless/03256_invalid_mutation_query.sql index 010f96414d4..2c554cabb9e 100644 --- a/tests/queries/0_stateless/03256_invalid_mutation_query.sql +++ b/tests/queries/0_stateless/03256_invalid_mutation_query.sql @@ -9,6 +9,8 @@ DELETE FROM t WHERE x IN (SELECT * FROM t2); -- { serverError 60 } ALTER TABLE t DELETE WHERE x in (SELECT y FROM t); -- { serverError 47 } ALTER TABLE t UPDATE x = 1 WHERE x IN (SELECT y FROM t); -- { serverError 47 } +DELETE FROM t WHERE x IN (SELECT foo FROM bar) SETTINGS validate_mutation_query = 0; + ALTER TABLE t ADD COLUMN y int; DELETE FROM t WHERE y in (SELECT y FROM t); From d0394719c6da6c3a7d647332b7ae977f703636b6 Mon Sep 17 00:00:00 2001 From: kssenii Date: Fri, 1 Nov 2024 12:11:07 +0100 Subject: [PATCH 374/680] More assertions --- .../IO/CachedOnDiskReadBufferFromFile.cpp | 1 + src/Interpreters/Cache/FileCache.cpp | 2 + src/Interpreters/Cache/FileSegment.cpp | 91 ++++++++++++++----- src/Interpreters/Cache/FileSegment.h | 2 +- src/Interpreters/Cache/Metadata.cpp | 21 +++-- 5 files changed, 89 insertions(+), 28 deletions(-) diff --git a/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp b/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp index 51c6045cb68..0f0cc4c4139 100644 --- a/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp +++ b/src/Disks/IO/CachedOnDiskReadBufferFromFile.cpp @@ -784,6 +784,7 @@ bool CachedOnDiskReadBufferFromFile::writeCache(char * data, size_t size, size_t LOG_INFO(log, "Insert into cache is skipped due to insufficient disk space. ({})", e.displayText()); return false; } + chassert(file_segment.state() == FileSegment::State::PARTIALLY_DOWNLOADED_NO_CONTINUATION); throw; } diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index f7b7ffc5aea..ae3c9c58fc5 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -1438,6 +1438,8 @@ void FileCache::loadMetadataForKeys(const fs::path & keys_dir) "cached file `{}` does not fit in cache anymore (size: {})", size_limit, offset_it->path().string(), size); + chassert(false); /// TODO: remove before merge. + fs::remove(offset_it->path()); } } diff --git a/src/Interpreters/Cache/FileSegment.cpp b/src/Interpreters/Cache/FileSegment.cpp index c356800fa57..f5a7011833a 100644 --- a/src/Interpreters/Cache/FileSegment.cpp +++ b/src/Interpreters/Cache/FileSegment.cpp @@ -139,7 +139,7 @@ FileSegmentGuard::Lock FileSegment::lock() const void FileSegment::setDownloadState(State state, const FileSegmentGuard::Lock & lock) { - if (isCompleted(false) && state != State::DETACHED) + if (isCompleted(false)) { throw Exception( ErrorCodes::LOGICAL_ERROR, @@ -700,6 +700,8 @@ void FileSegment::complete() case State::PARTIALLY_DOWNLOADED: { chassert(current_downloaded_size > 0); + chassert(fs::exists(getPath())); + chassert(fs::file_size(getPath()) > 0); if (is_last_holder) { @@ -841,29 +843,60 @@ bool FileSegment::assertCorrectnessUnlocked(const FileSegmentGuard::Lock & lock) } } - if (download_state == State::DOWNLOADED) + switch (download_state.load()) { - chassert(downloader_id.empty()); - chassert(downloaded_size == reserved_size); - chassert(downloaded_size == range().size()); - chassert(downloaded_size > 0); - chassert(std::filesystem::file_size(getPath()) > 0); - check_iterator(queue_iterator); - } - else - { - if (download_state == State::DOWNLOADING) - { - chassert(!downloader_id.empty()); - } - else if (download_state == State::PARTIALLY_DOWNLOADED - || download_state == State::EMPTY) + case State::EMPTY: { chassert(downloader_id.empty()); + chassert(!fs::exists(getPath())); + chassert(!queue_iterator); + break; } + case State::DOWNLOADED: + { + chassert(downloader_id.empty()); - chassert(reserved_size >= downloaded_size); - check_iterator(queue_iterator); + chassert(downloaded_size == reserved_size); + chassert(downloaded_size == range().size()); + chassert(downloaded_size > 0); + chassert(fs::file_size(getPath()) > 0); + + chassert(queue_iterator); + check_iterator(queue_iterator); + break; + } + case State::DOWNLOADING: + { + chassert(!downloader_id.empty()); + if (downloaded_size) + { + chassert(queue_iterator); + chassert(fs::file_size(getPath()) > 0); + } + break; + } + case State::PARTIALLY_DOWNLOADED: + { + chassert(downloader_id.empty()); + + chassert(reserved_size >= downloaded_size); + chassert(downloaded_size > 0); + chassert(fs::file_size(getPath()) > 0); + + chassert(queue_iterator); + check_iterator(queue_iterator); + break; + } + case State::PARTIALLY_DOWNLOADED_NO_CONTINUATION: + { + chassert(reserved_size >= downloaded_size); + check_iterator(queue_iterator); + break; + } + case State::DETACHED: + { + break; + } } return true; @@ -991,7 +1024,12 @@ FileSegmentsHolder::FileSegmentsHolder(FileSegments && file_segments_) FileSegmentPtr FileSegmentsHolder::getSingleFileSegment() const { if (file_segments.size() != 1) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected single file segment, got: {} in holder {}", file_segments.size(), toString()); + { + throw Exception( + ErrorCodes::LOGICAL_ERROR, + "Expected single file segment, got: {} in holder {}", + file_segments.size(), toString()); + } return file_segments.front(); } @@ -1001,7 +1039,18 @@ void FileSegmentsHolder::reset() ProfileEvents::increment(ProfileEvents::FilesystemCacheUnusedHoldFileSegments, file_segments.size()); for (auto file_segment_it = file_segments.begin(); file_segment_it != file_segments.end();) - file_segment_it = completeAndPopFrontImpl(); + { + try + { + file_segment_it = completeAndPopFrontImpl(); + } + catch (...) + { + chassert(false); + tryLogCurrentException(__PRETTY_FUNCTION__); + continue; + } + } file_segments.clear(); } diff --git a/src/Interpreters/Cache/FileSegment.h b/src/Interpreters/Cache/FileSegment.h index ee9aee1e354..79adc342329 100644 --- a/src/Interpreters/Cache/FileSegment.h +++ b/src/Interpreters/Cache/FileSegment.h @@ -254,7 +254,7 @@ private: const FileSegmentKind segment_kind; /// Size of the segment is not known until it is downloaded and /// can be bigger than max_file_segment_size. - const bool is_unbound = false; + const bool is_unbound; const bool background_download_enabled; std::atomic download_state; diff --git a/src/Interpreters/Cache/Metadata.cpp b/src/Interpreters/Cache/Metadata.cpp index 99ea01aa4f1..49dbbc71fa2 100644 --- a/src/Interpreters/Cache/Metadata.cpp +++ b/src/Interpreters/Cache/Metadata.cpp @@ -940,7 +940,16 @@ KeyMetadata::iterator LockedKey::removeFileSegmentImpl( if (file_segment->queue_iterator && invalidate_queue_entry) file_segment->queue_iterator->invalidate(); - file_segment->detach(segment_lock, *this); + try + { + file_segment->detach(segment_lock, *this); + } + catch (...) + { + tryLogCurrentException(__PRETTY_FUNCTION__); + chassert(false); + /// Do not rethrow, we much delete the file below. + } try { @@ -990,8 +999,8 @@ void LockedKey::shrinkFileSegmentToDownloadedSize( * because of no space left in cache, we need to be able to cut file segment's size to downloaded_size. */ - auto metadata = getByOffset(offset); - const auto & file_segment = metadata->file_segment; + auto file_segment_metadata = getByOffset(offset); + const auto & file_segment = file_segment_metadata->file_segment; chassert(file_segment->assertCorrectnessUnlocked(segment_lock)); const size_t downloaded_size = file_segment->getDownloadedSize(); @@ -1006,15 +1015,15 @@ void LockedKey::shrinkFileSegmentToDownloadedSize( chassert(file_segment->reserved_size >= downloaded_size); int64_t diff = file_segment->reserved_size - downloaded_size; - metadata->file_segment = std::make_shared( + file_segment_metadata->file_segment = std::make_shared( getKey(), offset, downloaded_size, FileSegment::State::DOWNLOADED, CreateFileSegmentSettings(file_segment->getKind()), false, file_segment->cache, key_metadata, file_segment->queue_iterator); if (diff) - metadata->getQueueIterator()->decrementSize(diff); + file_segment_metadata->getQueueIterator()->decrementSize(diff); - chassert(file_segment->assertCorrectnessUnlocked(segment_lock)); + chassert(file_segment_metadata->file_segment->assertCorrectnessUnlocked(segment_lock)); } bool LockedKey::addToDownloadQueue(size_t offset, const FileSegmentGuard::Lock &) From ce12f652c728df9513f5e8a940462558413bd58a Mon Sep 17 00:00:00 2001 From: avogar Date: Fri, 1 Nov 2024 11:25:21 +0000 Subject: [PATCH 375/680] Fix test flakiness --- .../queries/0_stateless/03246_alter_from_string_to_json.sql.j2 | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 b/tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 index e8760b659dc..2ccf2153699 100644 --- a/tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 +++ b/tests/queries/0_stateless/03246_alter_from_string_to_json.sql.j2 @@ -1,3 +1,6 @@ +-- Random settings limits: index_granularity=(None, 60000) +-- Tags: long + set allow_experimental_json_type = 1; set max_block_size = 20000; From e83cff7360e1a7ec0459a09bf95c954263b4c27c Mon Sep 17 00:00:00 2001 From: kssenii Date: Fri, 1 Nov 2024 12:47:03 +0100 Subject: [PATCH 376/680] Fix typo --- src/Interpreters/Cache/FileSegment.cpp | 2 +- src/Interpreters/Cache/Metadata.cpp | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/Interpreters/Cache/FileSegment.cpp b/src/Interpreters/Cache/FileSegment.cpp index f5a7011833a..080b54feb06 100644 --- a/src/Interpreters/Cache/FileSegment.cpp +++ b/src/Interpreters/Cache/FileSegment.cpp @@ -1046,8 +1046,8 @@ void FileSegmentsHolder::reset() } catch (...) { - chassert(false); tryLogCurrentException(__PRETTY_FUNCTION__); + chassert(false); continue; } } diff --git a/src/Interpreters/Cache/Metadata.cpp b/src/Interpreters/Cache/Metadata.cpp index 49dbbc71fa2..231545212cd 100644 --- a/src/Interpreters/Cache/Metadata.cpp +++ b/src/Interpreters/Cache/Metadata.cpp @@ -948,7 +948,7 @@ KeyMetadata::iterator LockedKey::removeFileSegmentImpl( { tryLogCurrentException(__PRETTY_FUNCTION__); chassert(false); - /// Do not rethrow, we much delete the file below. + /// Do not rethrow, we must delete the file below. } try From 2bafaa2fc675132d70d7683e16db4571dcddbd0e Mon Sep 17 00:00:00 2001 From: Pavel Kruglov <48961922+Avogar@users.noreply.github.com> Date: Fri, 1 Nov 2024 14:08:45 +0100 Subject: [PATCH 377/680] Update 03261_tuple_map_object_to_json_cast.sql --- .../queries/0_stateless/03261_tuple_map_object_to_json_cast.sql | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql b/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql index fcec7eb3af4..c0199452843 100644 --- a/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql +++ b/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql @@ -1,3 +1,5 @@ +-- Tags: no-fasttest + set allow_experimental_json_type = 1; set allow_experimental_object_type = 1; set allow_experimental_variant_type = 1; From 7a34fbc5b2f322341ae6c378920670f6a2258698 Mon Sep 17 00:00:00 2001 From: Anton Popov Date: Fri, 1 Nov 2024 14:20:37 +0000 Subject: [PATCH 378/680] allow to prewarm mark cache without enabled setting --- src/Interpreters/InterpreterSystemQuery.cpp | 2 +- src/Storages/MergeTree/MergeTreeData.cpp | 7 ++++- src/Storages/MergeTree/MergeTreeData.h | 1 + src/Storages/StorageMergeTree.cpp | 2 +- src/Storages/StorageReplicatedMergeTree.cpp | 2 +- .../03254_system_prewarm_mark_cache.reference | 4 +++ .../03254_system_prewarm_mark_cache.sql | 27 +++++++++++++++++++ 7 files changed, 41 insertions(+), 4 deletions(-) create mode 100644 tests/queries/0_stateless/03254_system_prewarm_mark_cache.reference create mode 100644 tests/queries/0_stateless/03254_system_prewarm_mark_cache.sql diff --git a/src/Interpreters/InterpreterSystemQuery.cpp b/src/Interpreters/InterpreterSystemQuery.cpp index 45636ab40b9..4c875026ace 100644 --- a/src/Interpreters/InterpreterSystemQuery.cpp +++ b/src/Interpreters/InterpreterSystemQuery.cpp @@ -1310,7 +1310,7 @@ RefreshTaskList InterpreterSystemQuery::getRefreshTasks() void InterpreterSystemQuery::prewarmMarkCache() { if (table_id.empty()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Table is not specified for prewarming marks cache"); + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Table is not specified for PREWARM MARK CACHE command"); getContext()->checkAccess(AccessType::SYSTEM_PREWARM_MARK_CACHE, table_id); diff --git a/src/Storages/MergeTree/MergeTreeData.cpp b/src/Storages/MergeTree/MergeTreeData.cpp index 4ed8c67469d..69979809c31 100644 --- a/src/Storages/MergeTree/MergeTreeData.cpp +++ b/src/Storages/MergeTree/MergeTreeData.cpp @@ -2343,11 +2343,16 @@ void MergeTreeData::stopOutdatedAndUnexpectedDataPartsLoadingTask() } } -void MergeTreeData::prewarmMarkCache(ThreadPool & pool) +void MergeTreeData::prewarmMarkCacheIfNeeded(ThreadPool & pool) { if (!(*getSettings())[MergeTreeSetting::prewarm_mark_cache]) return; + prewarmMarkCache(pool); +} + +void MergeTreeData::prewarmMarkCache(ThreadPool & pool) +{ auto * mark_cache = getContext()->getMarkCache().get(); if (!mark_cache) return; diff --git a/src/Storages/MergeTree/MergeTreeData.h b/src/Storages/MergeTree/MergeTreeData.h index a32106f76bb..8da4329a93b 100644 --- a/src/Storages/MergeTree/MergeTreeData.h +++ b/src/Storages/MergeTree/MergeTreeData.h @@ -508,6 +508,7 @@ public: /// Prewarm mark cache for the most recent data parts. void prewarmMarkCache(ThreadPool & pool); + void prewarmMarkCacheIfNeeded(ThreadPool & pool); String getLogName() const { return log.loadName(); } diff --git a/src/Storages/StorageMergeTree.cpp b/src/Storages/StorageMergeTree.cpp index 40cd6e01dba..1ba0617d8ae 100644 --- a/src/Storages/StorageMergeTree.cpp +++ b/src/Storages/StorageMergeTree.cpp @@ -155,7 +155,7 @@ StorageMergeTree::StorageMergeTree( loadMutations(); loadDeduplicationLog(); - prewarmMarkCache(getActivePartsLoadingThreadPool().get()); + prewarmMarkCacheIfNeeded(getActivePartsLoadingThreadPool().get()); } diff --git a/src/Storages/StorageReplicatedMergeTree.cpp b/src/Storages/StorageReplicatedMergeTree.cpp index bbfedb2f355..15341cca976 100644 --- a/src/Storages/StorageReplicatedMergeTree.cpp +++ b/src/Storages/StorageReplicatedMergeTree.cpp @@ -509,7 +509,7 @@ StorageReplicatedMergeTree::StorageReplicatedMergeTree( } loadDataParts(skip_sanity_checks, expected_parts_on_this_replica); - prewarmMarkCache(getActivePartsLoadingThreadPool().get()); + prewarmMarkCacheIfNeeded(getActivePartsLoadingThreadPool().get()); if (LoadingStrictnessLevel::ATTACH <= mode) { diff --git a/tests/queries/0_stateless/03254_system_prewarm_mark_cache.reference b/tests/queries/0_stateless/03254_system_prewarm_mark_cache.reference new file mode 100644 index 00000000000..86674e7765a --- /dev/null +++ b/tests/queries/0_stateless/03254_system_prewarm_mark_cache.reference @@ -0,0 +1,4 @@ +20000 +20000 +1 +0 diff --git a/tests/queries/0_stateless/03254_system_prewarm_mark_cache.sql b/tests/queries/0_stateless/03254_system_prewarm_mark_cache.sql new file mode 100644 index 00000000000..f9e77365836 --- /dev/null +++ b/tests/queries/0_stateless/03254_system_prewarm_mark_cache.sql @@ -0,0 +1,27 @@ +-- Tags: no-parallel, no-shared-merge-tree + +DROP TABLE IF EXISTS t_prewarm_cache; + +CREATE TABLE t_prewarm_cache (a UInt64, b UInt64, c UInt64) +ENGINE = ReplicatedMergeTree('/clickhouse/tables/{database}/03254_prewarm_mark_cache_smt/t_prewarm_cache', '1') +ORDER BY a SETTINGS prewarm_mark_cache = 0; + +SYSTEM DROP MARK CACHE; + +INSERT INTO t_prewarm_cache SELECT number, rand(), rand() FROM numbers(20000); + +SELECT count() FROM t_prewarm_cache WHERE NOT ignore(*); + +SYSTEM DROP MARK CACHE; + +SYSTEM PREWARM MARK CACHE t_prewarm_cache; + +SELECT count() FROM t_prewarm_cache WHERE NOT ignore(*); + +SYSTEM FLUSH LOGS; + +SELECT ProfileEvents['LoadedMarksCount'] > 0 FROM system.query_log +WHERE current_database = currentDatabase() AND type = 'QueryFinish' AND query LIKE 'SELECT count() FROM t_prewarm_cache%' +ORDER BY event_time_microseconds; + +DROP TABLE IF EXISTS t_prewarm_cache; From 47ddd7fb6b230e0d9b0d2341e118bd88ba871d07 Mon Sep 17 00:00:00 2001 From: avogar Date: Fri, 1 Nov 2024 14:33:03 +0000 Subject: [PATCH 379/680] Check suspicious and experimental types in JSON type hints --- src/DataTypes/DataTypeObject.cpp | 9 +++++++++ src/DataTypes/DataTypeObject.h | 2 ++ .../0_stateless/03261_json_hints_types_check.reference | 0 .../queries/0_stateless/03261_json_hints_types_check.sql | 9 +++++++++ 4 files changed, 20 insertions(+) create mode 100644 tests/queries/0_stateless/03261_json_hints_types_check.reference create mode 100644 tests/queries/0_stateless/03261_json_hints_types_check.sql diff --git a/src/DataTypes/DataTypeObject.cpp b/src/DataTypes/DataTypeObject.cpp index 18bfed9c5c3..69ae9b8e906 100644 --- a/src/DataTypes/DataTypeObject.cpp +++ b/src/DataTypes/DataTypeObject.cpp @@ -230,6 +230,15 @@ MutableColumnPtr DataTypeObject::createColumn() const return ColumnObject::create(std::move(typed_path_columns), max_dynamic_paths, max_dynamic_types); } +void DataTypeObject::forEachChild(const ChildCallback & callback) const +{ + for (const auto & [path, type] : typed_paths) + { + callback(*type); + type->forEachChild(callback); + } +} + namespace { diff --git a/src/DataTypes/DataTypeObject.h b/src/DataTypes/DataTypeObject.h index 7eb2e7729de..9321570fb75 100644 --- a/src/DataTypes/DataTypeObject.h +++ b/src/DataTypes/DataTypeObject.h @@ -50,6 +50,8 @@ public: bool equals(const IDataType & rhs) const override; + void forEachChild(const ChildCallback &) const override; + bool hasDynamicSubcolumnsData() const override { return true; } std::unique_ptr getDynamicSubcolumnData(std::string_view subcolumn_name, const SubstreamData & data, bool throw_if_null) const override; diff --git a/tests/queries/0_stateless/03261_json_hints_types_check.reference b/tests/queries/0_stateless/03261_json_hints_types_check.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03261_json_hints_types_check.sql b/tests/queries/0_stateless/03261_json_hints_types_check.sql new file mode 100644 index 00000000000..a407aa9474b --- /dev/null +++ b/tests/queries/0_stateless/03261_json_hints_types_check.sql @@ -0,0 +1,9 @@ +set allow_experimental_json_type=1; +set allow_experimental_variant_type=0; +set allow_experimental_object_type=0; + +select '{}'::JSON(a LowCardinality(Int128)); -- {serverError SUSPICIOUS_TYPE_FOR_LOW_CARDINALITY} +select '{}'::JSON(a FixedString(100000)); -- {serverError ILLEGAL_COLUMN} +select '{}'::JSON(a Variant(Int32)); -- {serverError ILLEGAL_COLUMN} +select '{}'::JSON(a Object('json')); -- {serverError ILLEGAL_COLUMN} + From 8f86168c65ad74e6203c59620f4667d0083e3c9e Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Fri, 1 Nov 2024 14:53:06 +0000 Subject: [PATCH 380/680] Fix test --- tests/integration/test_failed_mutations/test.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/integration/test_failed_mutations/test.py b/tests/integration/test_failed_mutations/test.py index 8d2ee46e748..c7e571ae171 100644 --- a/tests/integration/test_failed_mutations/test.py +++ b/tests/integration/test_failed_mutations/test.py @@ -28,7 +28,7 @@ POSTPONE_MUTATION_LOG = ( "According to exponential backoff policy, do not perform mutations for the part" ) FAILING_MUTATION_QUERY = ( - "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT throwIf(1))" + "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT throwIf(1)) SETTINGS allow_nondeterministic_mutations = 1" ) all_nodes = [node_with_backoff, node_no_backoff] @@ -111,7 +111,7 @@ def test_exponential_backoff_create_dependent_table(started_cluster): # Executing incorrect mutation. node_with_backoff.query( - "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT x FROM dep_table) SETTINGS validate_mutation_query = 0" + "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT x FROM dep_table) SETTINGS allow_nondeterministic_mutations = 1, validate_mutation_query = 0" ) # Creating dependent table for mutation. From 67b773dcddc61c01d603b16ac59632e9a8cc4f26 Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Fri, 1 Nov 2024 16:01:17 +0000 Subject: [PATCH 381/680] Fix style --- tests/integration/test_failed_mutations/test.py | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/tests/integration/test_failed_mutations/test.py b/tests/integration/test_failed_mutations/test.py index c7e571ae171..24b67ff86e5 100644 --- a/tests/integration/test_failed_mutations/test.py +++ b/tests/integration/test_failed_mutations/test.py @@ -27,9 +27,7 @@ REPLICATED_POSTPONE_MUTATION_LOG = ( POSTPONE_MUTATION_LOG = ( "According to exponential backoff policy, do not perform mutations for the part" ) -FAILING_MUTATION_QUERY = ( - "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT throwIf(1)) SETTINGS allow_nondeterministic_mutations = 1" -) +FAILING_MUTATION_QUERY = "ALTER TABLE test_mutations DELETE WHERE x IN (SELECT throwIf(1)) SETTINGS allow_nondeterministic_mutations = 1" all_nodes = [node_with_backoff, node_no_backoff] From 3fb4836f635a92ca59eda9dda519c8a466428bf9 Mon Sep 17 00:00:00 2001 From: Alexandre Snarskii Date: Fri, 1 Nov 2024 19:21:54 +0300 Subject: [PATCH 382/680] memory_worker shall be started on non-Linux OS too --- programs/server/Server.cpp | 2 ++ 1 file changed, 2 insertions(+) diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp index 1f481381b2b..5159f95419e 100644 --- a/programs/server/Server.cpp +++ b/programs/server/Server.cpp @@ -1353,9 +1353,11 @@ try } FailPointInjection::enableFromGlobalConfig(config()); +#endif memory_worker.start(); +#if defined(OS_LINUX) int default_oom_score = 0; #if !defined(NDEBUG) From 52fe2f18b08fdaff1d93abf2730096676eb55228 Mon Sep 17 00:00:00 2001 From: Michael Stetsyuk Date: Fri, 1 Nov 2024 16:42:01 +0000 Subject: [PATCH 383/680] rm metadata_version znode creation from restarting thread --- .../ReplicatedMergeTreeRestartingThread.cpp | 25 ++++--------------- 1 file changed, 5 insertions(+), 20 deletions(-) diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp index 93124e634bd..c73c9f6d048 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp @@ -31,6 +31,7 @@ namespace ErrorCodes extern const int REPLICA_IS_ALREADY_ACTIVE; extern const int REPLICA_STATUS_CHANGED; extern const int LOGICAL_ERROR; + extern const int SUPPORT_IS_DISABLED; } namespace FailPoints @@ -217,26 +218,10 @@ bool ReplicatedMergeTreeRestartingThread::tryStartup() } else { - /// Table was created before 20.4 and was never altered, - /// let's initialize replica metadata version from global metadata version. - - const String & zookeeper_path = storage.zookeeper_path, & replica_path = storage.replica_path; - - Coordination::Stat table_metadata_version_stat; - zookeeper->get(zookeeper_path + "/metadata", &table_metadata_version_stat); - - Coordination::Requests ops; - ops.emplace_back(zkutil::makeCheckRequest(zookeeper_path + "/metadata", table_metadata_version_stat.version)); - ops.emplace_back(zkutil::makeCreateRequest(replica_path + "/metadata_version", toString(table_metadata_version_stat.version), zkutil::CreateMode::Persistent)); - - Coordination::Responses res; - auto code = zookeeper->tryMulti(ops, res); - - if (code == Coordination::Error::ZBADVERSION) - throw Exception(ErrorCodes::REPLICA_STATUS_CHANGED, "Failed to initialize metadata_version " - "because table was concurrently altered, will retry"); - - zkutil::KeeperMultiException::check(code, ops, res); + throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, + "It seems you have upgraded from a version earlier than 20.4 straight to one later than 24.10. " + "ClickHouse does not support upgrades that span more than a year. " + "Please update gradually (through intermediate versions)."); } storage.queue.removeCurrentPartsFromMutations(); From 7e476b62d286326445d1a720f483e64fd8eae9d7 Mon Sep 17 00:00:00 2001 From: avogar Date: Fri, 1 Nov 2024 17:09:00 +0000 Subject: [PATCH 384/680] Fix tests --- tests/queries/0_stateless/03214_json_typed_dynamic_path.sql | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/03214_json_typed_dynamic_path.sql b/tests/queries/0_stateless/03214_json_typed_dynamic_path.sql index 1f6a025825a..eee3d70b8da 100644 --- a/tests/queries/0_stateless/03214_json_typed_dynamic_path.sql +++ b/tests/queries/0_stateless/03214_json_typed_dynamic_path.sql @@ -1,6 +1,7 @@ -- Tags: no-fasttest set allow_experimental_json_type = 1; +set allow_experimental_dynamic_type = 1; drop table if exists test; create table test (json JSON(a Dynamic)) engine=MergeTree order by tuple() settings min_rows_for_wide_part=1, min_bytes_for_wide_part=1; insert into test select '{"a" : 42}'; From 22e48f6852adcb3b3092b9b5a9e78674d52c0997 Mon Sep 17 00:00:00 2001 From: Pavel Kruglov <48961922+Avogar@users.noreply.github.com> Date: Fri, 1 Nov 2024 18:16:16 +0100 Subject: [PATCH 385/680] Update 03261_tuple_map_object_to_json_cast.sql --- .../queries/0_stateless/03261_tuple_map_object_to_json_cast.sql | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql b/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql index c0199452843..91d3f504f92 100644 --- a/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql +++ b/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql @@ -4,6 +4,7 @@ set allow_experimental_json_type = 1; set allow_experimental_object_type = 1; set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; +set enable_named_columns_in_function_tuple = 1; select 'Map to JSON'; select map('a', number::UInt32, 'b', toDate(number), 'c', range(number), 'd', [map('e', number::UInt32)])::JSON as json, JSONAllPathsWithTypes(json) from numbers(5); From bbde6ba51224c43cf88978adb92cd1a72b767313 Mon Sep 17 00:00:00 2001 From: Igor Nikonov Date: Fri, 1 Nov 2024 17:53:32 +0000 Subject: [PATCH 386/680] update test --- ...rallel_replicas_join_algo_and_analyzer_1.sh | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh index 1d43f540138..8d54c2eed13 100755 --- a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh @@ -27,6 +27,8 @@ inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 SETTINGS allow_experimental_analyzer=1" +PARALLEL_REPLICAS_SETTINGS="allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join = 0" + ############## echo echo "simple (global) join with analyzer and parallel replicas" @@ -35,17 +37,13 @@ $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, -max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0, parallel_replicas_local_plan=0" +SETTINGS enable_analyzer=1, $PARALLEL_REPLICAS_SETTING, parallel_replicas_local_plan=0" $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, send_logs_level='trace', -max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0, parallel_replicas_local_plan=0" 2>&1 | +SETTINGS enable_analyzer=1, send_logs_level='trace', $PARALLEL_REPLICAS_SETTING, parallel_replicas_local_plan=0" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' @@ -57,17 +55,13 @@ $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, -max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0, parallel_replicas_local_plan=0" +SETTINGS enable_analyzer=1, $PARALLEL_REPLICAS_SETTING, parallel_replicas_local_plan=1" $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS allow_experimental_analyzer=1, allow_experimental_parallel_reading_from_replicas = 2, send_logs_level='trace', -max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, -cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join=0, parallel_replicas_local_plan=1" 2>&1 | +SETTINGS enable_analyzer=1, send_logs_level='trace', $PARALLEL_REPLICAS_SETTING, parallel_replicas_local_plan=1" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' From 7315ad482052f50a98ea2eda433df34353e8d8d0 Mon Sep 17 00:00:00 2001 From: Igor Nikonov Date: Fri, 1 Nov 2024 17:55:49 +0000 Subject: [PATCH 387/680] Polishing --- .../02967_parallel_replicas_join_algo_and_analyzer_1.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh index 8d54c2eed13..d315257dbac 100755 --- a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh @@ -27,7 +27,7 @@ inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 SETTINGS allow_experimental_analyzer=1" -PARALLEL_REPLICAS_SETTINGS="allow_experimental_parallel_reading_from_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join = 0" +PARALLEL_REPLICAS_SETTINGS="enable_parallel_replicas = 2, max_parallel_replicas = 2, parallel_replicas_for_non_replicated_merge_tree = 1, cluster_for_parallel_replicas = 'test_cluster_one_shard_three_replicas_localhost', parallel_replicas_prefer_local_join = 0" ############## echo From 2cc2f31d9aebcf170b771be4d21cda63efcaf34e Mon Sep 17 00:00:00 2001 From: avogar Date: Fri, 1 Nov 2024 18:18:12 +0000 Subject: [PATCH 388/680] Fix error Invalid number of rows in Chunk with Variant column --- src/Columns/ColumnVariant.cpp | 2 +- .../0_stateless/03261_variant_permutation_bug.reference | 0 tests/queries/0_stateless/03261_variant_permutation_bug.sql | 6 ++++++ 3 files changed, 7 insertions(+), 1 deletion(-) create mode 100644 tests/queries/0_stateless/03261_variant_permutation_bug.reference create mode 100644 tests/queries/0_stateless/03261_variant_permutation_bug.sql diff --git a/src/Columns/ColumnVariant.cpp b/src/Columns/ColumnVariant.cpp index 564b60e1c1d..d5c8386d35f 100644 --- a/src/Columns/ColumnVariant.cpp +++ b/src/Columns/ColumnVariant.cpp @@ -952,7 +952,7 @@ ColumnPtr ColumnVariant::permute(const Permutation & perm, size_t limit) const if (hasOnlyNulls()) { if (limit) - return cloneResized(limit); + return cloneResized(limit ? std::min(size(), limit) : size()); /// If no limit, we can just return current immutable column. return this->getPtr(); diff --git a/tests/queries/0_stateless/03261_variant_permutation_bug.reference b/tests/queries/0_stateless/03261_variant_permutation_bug.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03261_variant_permutation_bug.sql b/tests/queries/0_stateless/03261_variant_permutation_bug.sql new file mode 100644 index 00000000000..373dd9e19fa --- /dev/null +++ b/tests/queries/0_stateless/03261_variant_permutation_bug.sql @@ -0,0 +1,6 @@ +set allow_experimental_variant_type=1; +create table test (x UInt64, d Variant(UInt64)) engine=Memory; +insert into test select number, null from numbers(200000); +select d from test order by d::String limit 32213 format Null; +drop table test; + From 9d0f256dfe87d0b914655570513e64f167cadeb0 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Fri, 1 Nov 2024 12:17:40 +0000 Subject: [PATCH 389/680] Enable SimSIMD backend in Usearch --- contrib/SimSIMD | 2 +- contrib/SimSIMD-cmake/CMakeLists.txt | 10 +++-- contrib/usearch-cmake/CMakeLists.txt | 64 +++++++++++++++++++++++++--- 3 files changed, 64 insertions(+), 12 deletions(-) diff --git a/contrib/SimSIMD b/contrib/SimSIMD index 935fef2964b..d7798ac6cb7 160000 --- a/contrib/SimSIMD +++ b/contrib/SimSIMD @@ -1 +1 @@ -Subproject commit 935fef2964bc38e995c5f465b42259a35b8cf0d3 +Subproject commit d7798ac6cb78ac1cb1cdc590f391643f983a2fd7 diff --git a/contrib/SimSIMD-cmake/CMakeLists.txt b/contrib/SimSIMD-cmake/CMakeLists.txt index f5dc4d63604..1d434490c7c 100644 --- a/contrib/SimSIMD-cmake/CMakeLists.txt +++ b/contrib/SimSIMD-cmake/CMakeLists.txt @@ -1,4 +1,6 @@ -set(SIMSIMD_PROJECT_DIR "${ClickHouse_SOURCE_DIR}/contrib/SimSIMD") - -add_library(_simsimd INTERFACE) -target_include_directories(_simsimd SYSTEM INTERFACE "${SIMSIMD_PROJECT_DIR}/include") +# See contrib/usearch-cmake/CMakeLists.txt, why only enabled on x86 +if (ARCH_AMD64) + set(SIMSIMD_PROJECT_DIR "${ClickHouse_SOURCE_DIR}/contrib/SimSIMD") + add_library(_simsimd INTERFACE) + target_include_directories(_simsimd SYSTEM INTERFACE "${SIMSIMD_PROJECT_DIR}/include") +endif() diff --git a/contrib/usearch-cmake/CMakeLists.txt b/contrib/usearch-cmake/CMakeLists.txt index 25f6ca82a74..69a986de192 100644 --- a/contrib/usearch-cmake/CMakeLists.txt +++ b/contrib/usearch-cmake/CMakeLists.txt @@ -6,12 +6,62 @@ target_include_directories(_usearch SYSTEM INTERFACE ${USEARCH_PROJECT_DIR}/incl target_link_libraries(_usearch INTERFACE _fp16) target_compile_definitions(_usearch INTERFACE USEARCH_USE_FP16LIB) -# target_compile_definitions(_usearch INTERFACE USEARCH_USE_SIMSIMD) -# ^^ simsimd is not enabled at the moment. Reasons: -# - Vectorization is important for raw scans but not so much for HNSW. We use usearch only for HNSW. -# - Simsimd does compile-time dispatch (choice of SIMD kernels determined by capabilities of the build machine) or dynamic dispatch (SIMD -# kernels chosen at runtime based on cpuid instruction). Since current builds are limited to SSE 4.2 (x86) and NEON (ARM), the speedup of -# the former would be moderate compared to AVX-512 / SVE. The latter is at the moment too fragile with respect to portability across x86 -# and ARM machines ... certain conbinations of quantizations / distance functions / SIMD instructions are not implemented at the moment. +# Only x86 for now. On ARM, the linker goes down in flames. To make SimSIMD compile, I had to remove a macro checks in SimSIMD +# for AVX512 (x86, worked nicely) and __ARM_BF16_FORMAT_ALTERNATIVE. It is probably because of that. +if (ARCH_AMD64) + target_link_libraries(_usearch INTERFACE _simsimd) + target_compile_definitions(_usearch INTERFACE USEARCH_USE_SIMSIMD) + + target_compile_definitions(_usearch INTERFACE USEARCH_CAN_COMPILE_FLOAT16) + target_compile_definitions(_usearch INTERFACE USEARCH_CAN_COMPILE_BF16) +endif () add_library(ch_contrib::usearch ALIAS _usearch) + + + +# LLVM ERROR: Cannot select: 0x7996e7a73150: f32,ch = load<(load (s16) from %ir.22, !tbaa !54231), anyext from bf16> 0x79961cb737c0, 0x7996e7a1a500, undef:i64, ./contrib/SimSIMD/include/simsimd/dot.h:215:1 +# 0x7996e7a1a500: i64 = add 0x79961e770d00, Constant:i64<-16>, ./contrib/SimSIMD/include/simsimd/dot.h:215:1 +# 0x79961e770d00: i64,ch = CopyFromReg 0x79961cb737c0, Register:i64 %4, ./contrib/SimSIMD/include/simsimd/dot.h:215:1 +# 0x7996e7a1ae10: i64 = Register %4 +# 0x7996e7a1b5f0: i64 = Constant<-16> +# 0x7996e7a1a730: i64 = undef +# In function: _ZL23simsimd_dot_bf16_serialPKu6__bf16S0_yPd +# PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and include the crash backtrace. +# Stack dump: +# 0. Running pass 'Function Pass Manager' on module 'src/libdbms.a(MergeTreeIndexVectorSimilarity.cpp.o at 2312737440)'. +# 1. Running pass 'AArch64 Instruction Selection' on function '@_ZL23simsimd_dot_bf16_serialPKu6__bf16S0_yPd' +# #0 0x00007999e83a63bf llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0xda63bf) +# #1 0x00007999e83a44f9 llvm::sys::RunSignalHandlers() (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0xda44f9) +# #2 0x00007999e83a6b00 (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0xda6b00) +# #3 0x00007999e6e45320 (/lib/x86_64-linux-gnu/libc.so.6+0x45320) +# #4 0x00007999e6e9eb1c pthread_kill (/lib/x86_64-linux-gnu/libc.so.6+0x9eb1c) +# #5 0x00007999e6e4526e raise (/lib/x86_64-linux-gnu/libc.so.6+0x4526e) +# #6 0x00007999e6e288ff abort (/lib/x86_64-linux-gnu/libc.so.6+0x288ff) +# #7 0x00007999e82fe0c2 llvm::report_fatal_error(llvm::Twine const&, bool) (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0xcfe0c2) +# #8 0x00007999e8c2f8e3 (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x162f8e3) +# #9 0x00007999e8c2ed76 llvm::SelectionDAGISel::SelectCodeCommon(llvm::SDNode*, unsigned char const*, unsigned int) (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x162ed76) +# #10 0x00007999ea1adbcb (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x2badbcb) +# #11 0x00007999e8c2611f llvm::SelectionDAGISel::DoInstructionSelection() (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x162611f) +# #12 0x00007999e8c25790 llvm::SelectionDAGISel::CodeGenAndEmitDAG() (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x1625790) +# #13 0x00007999e8c248de llvm::SelectionDAGISel::SelectAllBasicBlocks(llvm::Function const&) (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x16248de) +# #14 0x00007999e8c22934 llvm::SelectionDAGISel::runOnMachineFunction(llvm::MachineFunction&) (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x1622934) +# #15 0x00007999e87826b9 llvm::MachineFunctionPass::runOnFunction(llvm::Function&) (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x11826b9) +# #16 0x00007999e84f7772 llvm::FPPassManager::runOnFunction(llvm::Function&) (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0xef7772) +# #17 0x00007999e84fd2f4 llvm::FPPassManager::runOnModule(llvm::Module&) (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0xefd2f4) +# #18 0x00007999e84f7e9f llvm::legacy::PassManagerImpl::run(llvm::Module&) (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0xef7e9f) +# #19 0x00007999e99f7d61 (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x23f7d61) +# #20 0x00007999e99f8c91 (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x23f8c91) +# #21 0x00007999e99f8b10 llvm::lto::thinBackend(llvm::lto::Config const&, unsigned int, std::function>> (unsigned int, llvm::Twine const&)>, llvm::Module&, llvm::ModuleSummaryIndex const&, llvm::DenseMap, std::equal_to, std::allocator>, llvm::DenseMapInfo, llvm::detail::DenseMapPair, std::equal_to, std::allocator>>> const&, llvm::DenseMap, llvm::detail::DenseMapPair> const&, llvm::MapVector, llvm::detail::DenseMapPair>, llvm::SmallVector, 0u>>*, std::vector> const&) (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x23f8b10) +# #22 0x00007999e99f248d (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x23f248d) +# #23 0x00007999e99f1cd6 (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0x23f1cd6) +# #24 0x00007999e82c9beb (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0xcc9beb) +# #25 0x00007999e834ebe3 llvm::ThreadPool::processTasks(llvm::ThreadPoolTaskGroup*) (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0xd4ebe3) +# #26 0x00007999e834f704 (/usr/lib/llvm-18/bin/../lib/libLLVM.so.18.1+0xd4f704) +# #27 0x00007999e6e9ca94 (/lib/x86_64-linux-gnu/libc.so.6+0x9ca94) +# #28 0x00007999e6f29c3c (/lib/x86_64-linux-gnu/libc.so.6+0x129c3c) +# clang++-18: error: unable to execute command: Aborted (core dumped) +# clang++-18: error: linker command failed due to signal (use -v to see invocation) +# ^[[A^Cninja: build stopped: interrupted by user. From 3a042c080473957ffe40c5e299b06714868ab841 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Fri, 1 Nov 2024 12:55:02 +0000 Subject: [PATCH 390/680] Enable dynamic dispatch in SimSIMD --- contrib/SimSIMD-cmake/CMakeLists.txt | 6 ++++-- src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp | 2 ++ 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/contrib/SimSIMD-cmake/CMakeLists.txt b/contrib/SimSIMD-cmake/CMakeLists.txt index 1d434490c7c..8350417479a 100644 --- a/contrib/SimSIMD-cmake/CMakeLists.txt +++ b/contrib/SimSIMD-cmake/CMakeLists.txt @@ -1,6 +1,8 @@ # See contrib/usearch-cmake/CMakeLists.txt, why only enabled on x86 if (ARCH_AMD64) set(SIMSIMD_PROJECT_DIR "${ClickHouse_SOURCE_DIR}/contrib/SimSIMD") - add_library(_simsimd INTERFACE) - target_include_directories(_simsimd SYSTEM INTERFACE "${SIMSIMD_PROJECT_DIR}/include") + set(SIMSIMD_SRCS ${SIMSIMD_PROJECT_DIR}/c/lib.c) + add_library(_simsimd ${SIMSIMD_SRCS}) + target_include_directories(_simsimd SYSTEM PUBLIC "${SIMSIMD_PROJECT_DIR}/include") + target_compile_definitions(_simsimd PUBLIC SIMSIMD_DYNAMIC_DISPATCH) endif() diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp index 5a725922e14..0b5ffa659dc 100644 --- a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp @@ -118,6 +118,8 @@ USearchIndexWithSerialization::USearchIndexWithSerialization( if (!result) throw Exception(ErrorCodes::INCORRECT_DATA, "Could not create vector similarity index. Error: {}", String(result.error.release())); swap(result.index); + + /// LOG_TRACE(getLogger("XXX"), "{}", simsimd_uses_dynamic_dispatch()); } void USearchIndexWithSerialization::serialize(WriteBuffer & ostr) const From a4e576924b16ed199e3726313f96c241b604d4b6 Mon Sep 17 00:00:00 2001 From: 0xMihalich Date: Sat, 2 Nov 2024 18:48:57 +1000 Subject: [PATCH 391/680] Fix: ERROR: column "attgenerated" does not exist for old PostgreSQL databases Restore support for GreenPlum and older versions of PostgreSQL without affecting existing functionality. --- .../PostgreSQL/fetchPostgreSQLTableStructure.cpp | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.cpp b/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.cpp index 45fd52f27ab..5268dbcb59f 100644 --- a/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.cpp +++ b/src/Databases/PostgreSQL/fetchPostgreSQLTableStructure.cpp @@ -307,6 +307,13 @@ PostgreSQLTableStructure fetchPostgreSQLTableStructure( if (!columns.empty()) columns_part = fmt::format(" AND attname IN ('{}')", boost::algorithm::join(columns, "','")); + /// Bypassing the error of the missing column `attgenerated` in the system table `pg_attribute` for PostgreSQL versions below 12. + /// This trick involves executing a special query to the DBMS in advance to obtain the correct line with comment /// if column has GENERATED. + /// The result of the query will be the name of the column `attgenerated` or an empty string declaration for PostgreSQL version 11 and below. + /// This change does not degrade the function's performance but restores support for older versions and fix ERROR: column "attgenerated" does not exist. + pqxx::result gen_result{tx.exec("select case when current_setting('server_version_num')::int < 120000 then '''''' else 'attgenerated' end as generated")}; + std::string generated = gen_result[0][0].as(); + std::string query = fmt::format( "SELECT attname AS name, " /// column name "format_type(atttypid, atttypmod) AS type, " /// data type @@ -315,11 +322,11 @@ PostgreSQLTableStructure fetchPostgreSQLTableStructure( "atttypid as type_id, " "atttypmod as type_modifier, " "attnum as att_num, " - "attgenerated as generated " /// if column has GENERATED + "{} as generated " /// if column has GENERATED "FROM pg_attribute " "WHERE attrelid = (SELECT oid FROM pg_class WHERE {}) {}" "AND NOT attisdropped AND attnum > 0 " - "ORDER BY attnum ASC", where, columns_part); + "ORDER BY attnum ASC", generated, where, columns_part); /// Now we use variable `generated` to form query string. End of trick. auto postgres_table_with_schema = postgres_schema.empty() ? postgres_table : doubleQuoteString(postgres_schema) + '.' + doubleQuoteString(postgres_table); table.physical_columns = readNamesAndTypesList(tx, postgres_table_with_schema, query, use_nulls, false); From 64b405254c0c7dbe2217bd6251f3767556d01d75 Mon Sep 17 00:00:00 2001 From: Igor Nikonov Date: Sat, 2 Nov 2024 19:50:45 +0000 Subject: [PATCH 392/680] Fix --- .../02967_parallel_replicas_join_algo_and_analyzer_1.sh | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh index d315257dbac..a6e755ebc35 100755 --- a/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh +++ b/tests/queries/0_stateless/02967_parallel_replicas_join_algo_and_analyzer_1.sh @@ -37,13 +37,13 @@ $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS enable_analyzer=1, $PARALLEL_REPLICAS_SETTING, parallel_replicas_local_plan=0" +SETTINGS enable_analyzer=1, $PARALLEL_REPLICAS_SETTINGS, parallel_replicas_local_plan=0" $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS enable_analyzer=1, send_logs_level='trace', $PARALLEL_REPLICAS_SETTING, parallel_replicas_local_plan=0" 2>&1 | +SETTINGS enable_analyzer=1, send_logs_level='trace', $PARALLEL_REPLICAS_SETTINGS, parallel_replicas_local_plan=0" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' @@ -55,13 +55,13 @@ $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS enable_analyzer=1, $PARALLEL_REPLICAS_SETTING, parallel_replicas_local_plan=1" +SETTINGS enable_analyzer=1, $PARALLEL_REPLICAS_SETTINGS, parallel_replicas_local_plan=1" $CLICKHOUSE_CLIENT -q " select * from (select key, value from num_1) l inner join (select key, value from num_2) r on l.key = r.key order by l.key limit 10 offset 700000 -SETTINGS enable_analyzer=1, send_logs_level='trace', $PARALLEL_REPLICAS_SETTING, parallel_replicas_local_plan=1" 2>&1 | +SETTINGS enable_analyzer=1, send_logs_level='trace', $PARALLEL_REPLICAS_SETTINGS, parallel_replicas_local_plan=1" 2>&1 | grep "executeQuery\|.*Coordinator: Coordination done" | grep -o "SELECT.*WithMergeableState)\|.*Coordinator: Coordination done" | sed -re 's/_data_[[:digit:]]+_[[:digit:]]+/_data_/g' From 1d83bb2ddaeab407af0fa7d93307bb2465568b2b Mon Sep 17 00:00:00 2001 From: Igor Nikonov Date: Sun, 3 Nov 2024 07:39:38 +0000 Subject: [PATCH 393/680] Update settings changes history --- src/Core/SettingsChangesHistory.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 317037070fc..9f314788505 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -71,6 +71,7 @@ static std::initializer_list Date: Sun, 3 Nov 2024 15:10:26 +0000 Subject: [PATCH 394/680] Fix test --- .../0_stateless/02354_vector_search_expansion_search.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/02354_vector_search_expansion_search.sql b/tests/queries/0_stateless/02354_vector_search_expansion_search.sql index fcbe9ee42b9..f0cd5374be7 100644 --- a/tests/queries/0_stateless/02354_vector_search_expansion_search.sql +++ b/tests/queries/0_stateless/02354_vector_search_expansion_search.sql @@ -14,7 +14,7 @@ CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similar -- Generate random values but with a fixed seed (conceptually), so that the data is deterministic. -- Unfortunately, no random functions in ClickHouse accepts a seed. Instead, abuse the numbers table + hash functions to provide -- deterministic randomness. -INSERT INTO tab SELECT number, [sipHash64(number)/18446744073709551615, wyHash64(number)/18446744073709551615] FROM numbers(370000); -- 18446744073709551615 is the biggest UInt64 +INSERT INTO tab SELECT number, [sipHash64(number)/18446744073709551615, wyHash64(number)/18446744073709551615] FROM numbers(660000); -- 18446744073709551615 is the biggest UInt64 -- hnsw_candidate_list_size_for_search = 0 is illegal WITH [0.5, 0.5] AS reference_vec From 27241b484f8c26197ec4329212a8a5ef11d02007 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Sun, 3 Nov 2024 16:00:33 +0000 Subject: [PATCH 395/680] Fix linker warning --- contrib/usearch | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/usearch b/contrib/usearch index 53799b84ca9..7efe8b710c9 160000 --- a/contrib/usearch +++ b/contrib/usearch @@ -1 +1 @@ -Subproject commit 53799b84ca9ad708b060d0b1cfa5f039371721cd +Subproject commit 7efe8b710c9831bfe06573b1df0fad001b04a2b5 From 27049f2cb599b4f93ae327783ab0cc588bef7dd1 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Sun, 3 Nov 2024 19:16:35 +0000 Subject: [PATCH 396/680] Demote log level for failed authentication --- src/Access/AccessControl.cpp | 7 ++++--- src/Common/Exception.cpp | 31 +++++++++++++++++++++---------- src/Common/Exception.h | 9 +++++---- src/Server/TCPHandler.cpp | 3 ++- 4 files changed, 32 insertions(+), 18 deletions(-) diff --git a/src/Access/AccessControl.cpp b/src/Access/AccessControl.cpp index e8ee363be1a..9b3b8d2a977 100644 --- a/src/Access/AccessControl.cpp +++ b/src/Access/AccessControl.cpp @@ -608,7 +608,7 @@ AuthResult AccessControl::authenticate(const Credentials & credentials, const Po } catch (...) { - tryLogCurrentException(getLogger(), "from: " + address.toString() + ", user: " + credentials.getUserName() + ": Authentication failed"); + tryLogCurrentException(getLogger(), "from: " + address.toString() + ", user: " + credentials.getUserName() + ": Authentication failed", LogsLevel::information); WriteBufferFromOwnString message; message << credentials.getUserName() << ": Authentication failed: password is incorrect, or there is no user with such name."; @@ -622,8 +622,9 @@ AuthResult AccessControl::authenticate(const Credentials & credentials, const Po << "and deleting this file will reset the password.\n" << "See also /etc/clickhouse-server/users.xml on the server where ClickHouse is installed.\n\n"; - /// We use the same message for all authentication failures because we don't want to give away any unnecessary information for security reasons, - /// only the log will show the exact reason. + /// We use the same message for all authentication failures because we don't want to give away any unnecessary information for security reasons. + /// Only the log ((*), above) will show the exact reason. Note that (*) logs at information level instead of the default error level as + /// authentication failures are not an unusual event. throw Exception(PreformattedMessage{message.str(), "{}: Authentication failed: password is incorrect, or there is no user with such name", std::vector{credentials.getUserName()}}, diff --git a/src/Common/Exception.cpp b/src/Common/Exception.cpp index 320fc06cb2f..644c9a19738 100644 --- a/src/Common/Exception.cpp +++ b/src/Common/Exception.cpp @@ -251,7 +251,7 @@ void Exception::setThreadFramePointers(ThreadFramePointersBase frame_pointers) thread_frame_pointers.frame_pointers = std::move(frame_pointers); } -static void tryLogCurrentExceptionImpl(Poco::Logger * logger, const std::string & start_of_message) +static void tryLogCurrentExceptionImpl(Poco::Logger * logger, const std::string & start_of_message, LogsLevel level) { if (!isLoggingEnabled()) return; @@ -262,14 +262,25 @@ static void tryLogCurrentExceptionImpl(Poco::Logger * logger, const std::string if (!start_of_message.empty()) message.text = fmt::format("{}: {}", start_of_message, message.text); - LOG_ERROR(logger, message); + switch (level) + { + case LogsLevel::none: break; + case LogsLevel::test: LOG_TEST(logger, message); break; + case LogsLevel::trace: LOG_TRACE(logger, message); break; + case LogsLevel::debug: LOG_DEBUG(logger, message); break; + case LogsLevel::information: LOG_INFO(logger, message); break; + case LogsLevel::warning: LOG_WARNING(logger, message); break; + case LogsLevel::error: LOG_ERROR(logger, message); break; + case LogsLevel::fatal: LOG_FATAL(logger, message); break; + } + } catch (...) // NOLINT(bugprone-empty-catch) { } } -void tryLogCurrentException(const char * log_name, const std::string & start_of_message) +void tryLogCurrentException(const char * log_name, const std::string & start_of_message, LogsLevel level) { if (!isLoggingEnabled()) return; @@ -283,10 +294,10 @@ void tryLogCurrentException(const char * log_name, const std::string & start_of_ /// getLogger can allocate memory too auto logger = getLogger(log_name); - tryLogCurrentExceptionImpl(logger.get(), start_of_message); + tryLogCurrentExceptionImpl(logger.get(), start_of_message, level); } -void tryLogCurrentException(Poco::Logger * logger, const std::string & start_of_message) +void tryLogCurrentException(Poco::Logger * logger, const std::string & start_of_message, LogsLevel level) { /// Under high memory pressure, new allocations throw a /// MEMORY_LIMIT_EXCEEDED exception. @@ -295,17 +306,17 @@ void tryLogCurrentException(Poco::Logger * logger, const std::string & start_of_ /// MemoryTracker until the exception will be logged. LockMemoryExceptionInThread lock_memory_tracker(VariableContext::Global); - tryLogCurrentExceptionImpl(logger, start_of_message); + tryLogCurrentExceptionImpl(logger, start_of_message, level); } -void tryLogCurrentException(LoggerPtr logger, const std::string & start_of_message) +void tryLogCurrentException(LoggerPtr logger, const std::string & start_of_message, LogsLevel level) { - tryLogCurrentException(logger.get(), start_of_message); + tryLogCurrentException(logger.get(), start_of_message, level); } -void tryLogCurrentException(const AtomicLogger & logger, const std::string & start_of_message) +void tryLogCurrentException(const AtomicLogger & logger, const std::string & start_of_message, LogsLevel level) { - tryLogCurrentException(logger.load(), start_of_message); + tryLogCurrentException(logger.load(), start_of_message, level); } static void getNoSpaceLeftInfoMessage(std::filesystem::path path, String & msg) diff --git a/src/Common/Exception.h b/src/Common/Exception.h index 8ec640ff642..edc1b95bca4 100644 --- a/src/Common/Exception.h +++ b/src/Common/Exception.h @@ -7,6 +7,7 @@ #include #include #include +#include #include #include @@ -276,10 +277,10 @@ using Exceptions = std::vector; * Can be used in destructors in the catch-all block. */ /// TODO: Logger leak constexpr overload -void tryLogCurrentException(const char * log_name, const std::string & start_of_message = ""); -void tryLogCurrentException(Poco::Logger * logger, const std::string & start_of_message = ""); -void tryLogCurrentException(LoggerPtr logger, const std::string & start_of_message = ""); -void tryLogCurrentException(const AtomicLogger & logger, const std::string & start_of_message = ""); +void tryLogCurrentException(const char * log_name, const std::string & start_of_message = "", LogsLevel level = LogsLevel::error); +void tryLogCurrentException(Poco::Logger * logger, const std::string & start_of_message = "", LogsLevel level = LogsLevel::error); +void tryLogCurrentException(LoggerPtr logger, const std::string & start_of_message = "", LogsLevel level = LogsLevel::error); +void tryLogCurrentException(const AtomicLogger & logger, const std::string & start_of_message = "", LogsLevel level = LogsLevel::error); /** Prints current exception in canonical format. diff --git a/src/Server/TCPHandler.cpp b/src/Server/TCPHandler.cpp index e7e4ae25a68..ea5507c3155 100644 --- a/src/Server/TCPHandler.cpp +++ b/src/Server/TCPHandler.cpp @@ -1614,7 +1614,8 @@ void TCPHandler::receiveHello() if (e.code() != DB::ErrorCodes::AUTHENTICATION_FAILED) throw; - tryLogCurrentException(log, "SSL authentication failed, falling back to password authentication"); + tryLogCurrentException(log, "SSL authentication failed, falling back to password authentication", LogsLevel::debug); + /// ^^ Log at debug level instead of default error level as authentication failures are not an unusual event. } } } From 7f1ccc30c9e192a00ca624bcfcd05c9b2837d27d Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Sun, 3 Nov 2024 21:19:27 +0000 Subject: [PATCH 397/680] Try to suppress msan warnings --- contrib/SimSIMD | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/SimSIMD b/contrib/SimSIMD index d7798ac6cb7..c03d065a766 160000 --- a/contrib/SimSIMD +++ b/contrib/SimSIMD @@ -1 +1 @@ -Subproject commit d7798ac6cb78ac1cb1cdc590f391643f983a2fd7 +Subproject commit c03d065a7661004a9a18fe52753efafa170c67f9 From 5aba66e50a98f040daaa3c2235310e68cfa45e55 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Mon, 4 Nov 2024 03:13:42 +0000 Subject: [PATCH 398/680] adjust CI timeout, use TIMEOUT variable for setting fuzzers timeout --- docker/test/libfuzzer/Dockerfile | 2 -- tests/ci/ci_config.py | 2 +- tests/ci/libfuzzer_test_check.py | 3 +++ tests/fuzz/runner.py | 8 ++------ 4 files changed, 6 insertions(+), 9 deletions(-) diff --git a/docker/test/libfuzzer/Dockerfile b/docker/test/libfuzzer/Dockerfile index 3ffae0cd921..46e305c90ab 100644 --- a/docker/test/libfuzzer/Dockerfile +++ b/docker/test/libfuzzer/Dockerfile @@ -33,8 +33,6 @@ RUN apt-get update \ COPY requirements.txt / RUN pip3 install --no-cache-dir -r /requirements.txt -ENV FUZZER_ARGS="-max_total_time=60" - SHELL ["/bin/bash", "-c"] # docker run --network=host --volume :/workspace -e PR_TO_TEST=<> -e SHA_TO_TEST=<> clickhouse/libfuzzer diff --git a/tests/ci/ci_config.py b/tests/ci/ci_config.py index b4b7dbee59c..80da822652f 100644 --- a/tests/ci/ci_config.py +++ b/tests/ci/ci_config.py @@ -530,7 +530,7 @@ class CI: JobNames.LIBFUZZER_TEST: JobConfig( required_builds=[BuildNames.FUZZERS], run_by_labels=[Tags.libFuzzer], - timeout=10800, + timeout=5400, run_command='libfuzzer_test_check.py "$CHECK_NAME"', runner_type=Runners.FUNC_TESTER, ), diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index 379d681cb3e..d0936eb2323 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -22,6 +22,7 @@ from stopwatch import Stopwatch from tee_popen import TeePopen NO_CHANGES_MSG = "Nothing to run" +TIMEOUT = 60 s3 = S3Helper() @@ -264,6 +265,8 @@ def main(): check_name, run_by_hash_num, run_by_hash_total ) + additional_envs.append(f"TIMEOUT={TIMEOUT}") + ci_logs_credentials = CiLogsCredentials(Path(temp_path) / "export-logs-config.sh") ci_logs_args = ci_logs_credentials.get_docker_arguments( pr_info, stopwatch.start_time_str, check_name diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index af73a989ec3..0880940aabd 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -9,7 +9,7 @@ import subprocess from pathlib import Path DEBUGGER = os.getenv("DEBUGGER", "") -FUZZER_ARGS = os.getenv("FUZZER_ARGS", "") +TIMEOUT = int(os.getenv("TIMEOUT", "0")) OUTPUT = "/test_output" @@ -150,11 +150,7 @@ def main(): subprocess.check_call("ls -al", shell=True) - timeout = 60 - - match = re.search(r"(^|\s+)-max_total_time=(\d+)($|\s)", FUZZER_ARGS) - if match: - timeout = int(match.group(2)) + timeout = 30 if TIMEOUT == 0 else TIMEOUT with Path() as current: for fuzzer in current.iterdir(): From e2d64ea30254ce7e126c4442fe393429cfbd1c21 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Mon, 4 Nov 2024 03:37:46 +0000 Subject: [PATCH 399/680] fix style --- tests/fuzz/runner.py | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/fuzz/runner.py b/tests/fuzz/runner.py index 0880940aabd..f4c66e00117 100644 --- a/tests/fuzz/runner.py +++ b/tests/fuzz/runner.py @@ -4,7 +4,6 @@ import configparser import datetime import logging import os -import re import subprocess from pathlib import Path From a6c98a4a7f6c650c84dc750972176427a6e8c479 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Mon, 4 Nov 2024 05:17:46 +0000 Subject: [PATCH 400/680] take some changes from private --- tests/ci/s3_helper.py | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/tests/ci/s3_helper.py b/tests/ci/s3_helper.py index 46c206f0540..ced6d29e5c7 100644 --- a/tests/ci/s3_helper.py +++ b/tests/ci/s3_helper.py @@ -322,17 +322,23 @@ class S3Helper: return result def list_prefix_non_recursive( - self, s3_prefix_path: str, bucket: str = S3_BUILDS_BUCKET + self, + s3_prefix_path: str, + bucket: str = S3_BUILDS_BUCKET, + only_dirs: bool = False, ) -> List[str]: paginator = self.client.get_paginator("list_objects_v2") - pages = paginator.paginate(Bucket=bucket, Prefix=s3_prefix_path) + pages = paginator.paginate( + Bucket=bucket, Prefix=s3_prefix_path, Delimiter="/" + ) result = [] for page in pages: - if "Contents" in page: + if not only_dirs and "Contents" in page: for obj in page["Contents"]: - if "/" not in obj["Key"][len(s3_prefix_path) + 1 :]: - result.append(obj["Key"]) - + result.append(obj["Key"]) + if "CommonPrefixes" in page: + for obj in page["CommonPrefixes"]: + result.append(obj["Prefix"]) return result def url_if_exists(self, key: str, bucket: str = S3_BUILDS_BUCKET) -> str: From 94c8e6e6c201194fc6eea0784e9200fdf5d639a4 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Mon, 4 Nov 2024 05:31:15 +0000 Subject: [PATCH 401/680] Automatic style fix --- tests/ci/s3_helper.py | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/tests/ci/s3_helper.py b/tests/ci/s3_helper.py index ced6d29e5c7..d0aa034258a 100644 --- a/tests/ci/s3_helper.py +++ b/tests/ci/s3_helper.py @@ -328,9 +328,7 @@ class S3Helper: only_dirs: bool = False, ) -> List[str]: paginator = self.client.get_paginator("list_objects_v2") - pages = paginator.paginate( - Bucket=bucket, Prefix=s3_prefix_path, Delimiter="/" - ) + pages = paginator.paginate(Bucket=bucket, Prefix=s3_prefix_path, Delimiter="/") result = [] for page in pages: if not only_dirs and "Contents" in page: From 12c21dc7df4ea2a538a1c59bfa7eb05dd76df08d Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Mon, 4 Nov 2024 09:00:01 +0000 Subject: [PATCH 402/680] Minor fixups --- contrib/SimSIMD | 2 +- .../0_stateless/02354_vector_search_expansion_search.sql | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/contrib/SimSIMD b/contrib/SimSIMD index c03d065a766..ee3c9c9c00b 160000 --- a/contrib/SimSIMD +++ b/contrib/SimSIMD @@ -1 +1 @@ -Subproject commit c03d065a7661004a9a18fe52753efafa170c67f9 +Subproject commit ee3c9c9c00b51645f62a1a9e99611b78c0052a21 diff --git a/tests/queries/0_stateless/02354_vector_search_expansion_search.sql b/tests/queries/0_stateless/02354_vector_search_expansion_search.sql index f0cd5374be7..427148b829f 100644 --- a/tests/queries/0_stateless/02354_vector_search_expansion_search.sql +++ b/tests/queries/0_stateless/02354_vector_search_expansion_search.sql @@ -1,4 +1,4 @@ --- Tags: no-fasttest, long, no-asan, no-asan, no-ubsan, no-debug +-- Tags: no-fasttest, long, no-asan, no-ubsan, no-debug -- ^^ Disable test for slow builds: generating data takes time but a sufficiently large data set -- is necessary for different hnsw_candidate_list_size_for_search settings to make a difference From c7f970405885d6dae54c9eb94201c662528ab965 Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Mon, 4 Nov 2024 09:45:26 +0000 Subject: [PATCH 403/680] Try fix integration test --- tests/integration/test_quorum_inserts/test.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index eefc4882e8e..66f96d61b3e 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -366,7 +366,7 @@ def test_insert_quorum_with_ttl(started_cluster): zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster") -def test_insert_quorum_with_keeper_loss_connection(): +def test_insert_quorum_with_keeper_loss_connection(started_cluster): zero.query( "DROP TABLE IF EXISTS test_insert_quorum_with_keeper_fail ON CLUSTER cluster" ) From 6471034082e931a602fafd2530b218d4b1d386b3 Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Mon, 4 Nov 2024 13:02:58 +0100 Subject: [PATCH 404/680] impl --- base/base/StringRef.h | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) diff --git a/base/base/StringRef.h b/base/base/StringRef.h index af3441c2a75..ee62be2c4eb 100644 --- a/base/base/StringRef.h +++ b/base/base/StringRef.h @@ -86,7 +86,7 @@ using StringRefs = std::vector; * For more information, see hash_map_string_2.cpp */ -inline bool compare8(const char * p1, const char * p2) +inline bool compare16(const char * p1, const char * p2) { return 0xFFFF == _mm_movemask_epi8(_mm_cmpeq_epi8( _mm_loadu_si128(reinterpret_cast(p1)), @@ -115,7 +115,7 @@ inline bool compare64(const char * p1, const char * p2) #elif defined(__aarch64__) && defined(__ARM_NEON) -inline bool compare8(const char * p1, const char * p2) +inline bool compare16(const char * p1, const char * p2) { uint64_t mask = getNibbleMask(vceqq_u8( vld1q_u8(reinterpret_cast(p1)), vld1q_u8(reinterpret_cast(p2)))); @@ -185,13 +185,22 @@ inline bool memequalWide(const char * p1, const char * p2, size_t size) switch (size / 16) // NOLINT(bugprone-switch-missing-default-case) { - case 3: if (!compare8(p1 + 32, p2 + 32)) return false; [[fallthrough]]; - case 2: if (!compare8(p1 + 16, p2 + 16)) return false; [[fallthrough]]; - case 1: if (!compare8(p1, p2)) return false; [[fallthrough]]; + case 3: + if (!compare16(p1 + 32, p2 + 32)) + return false; + [[fallthrough]]; + case 2: + if (!compare16(p1 + 16, p2 + 16)) + return false; + [[fallthrough]]; + case 1: + if (!compare16(p1, p2)) + return false; + [[fallthrough]]; default: ; } - return compare8(p1 + size - 16, p2 + size - 16); + return compare16(p1 + size - 16, p2 + size - 16); } #endif From a37c1134b99e75df1df7320c1cad6420d2014a04 Mon Sep 17 00:00:00 2001 From: divanik Date: Mon, 4 Nov 2024 12:32:14 +0000 Subject: [PATCH 405/680] Resolve issues --- src/Storages/ObjectStorage/StorageObjectStorage.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Storages/ObjectStorage/StorageObjectStorage.cpp b/src/Storages/ObjectStorage/StorageObjectStorage.cpp index a72fd16abc2..fd2fe0400bb 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorage.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorage.cpp @@ -102,7 +102,7 @@ StorageObjectStorage::StorageObjectStorage( } else { - tryLogCurrentException(__PRETTY_FUNCTION__); + tryLogCurrentException(log); } } From c3471ef20d5a3c375d632bd600438d555cd51595 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Mon, 4 Nov 2024 13:33:34 +0100 Subject: [PATCH 406/680] Update AccessControl.cpp --- src/Access/AccessControl.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Access/AccessControl.cpp b/src/Access/AccessControl.cpp index 9b3b8d2a977..647fb238d48 100644 --- a/src/Access/AccessControl.cpp +++ b/src/Access/AccessControl.cpp @@ -608,7 +608,7 @@ AuthResult AccessControl::authenticate(const Credentials & credentials, const Po } catch (...) { - tryLogCurrentException(getLogger(), "from: " + address.toString() + ", user: " + credentials.getUserName() + ": Authentication failed", LogsLevel::information); + tryLogCurrentException(getLogger(), "from: " + address.toString() + ", user: " + credentials.getUserName() + ": Authentication failed", LogsLevel::debug); WriteBufferFromOwnString message; message << credentials.getUserName() << ": Authentication failed: password is incorrect, or there is no user with such name."; From 24a7e0f4ee52e47cadd00a41bff80eb3ac614960 Mon Sep 17 00:00:00 2001 From: Azat Khuzhin Date: Mon, 4 Nov 2024 13:44:36 +0100 Subject: [PATCH 407/680] Fix missing cluster startup for test_quorum_inserts::test_insert_quorum_with_keeper_fail def test_insert_quorum_with_keeper_loss_connection(): > zero.query( "DROP TABLE IF EXISTS test_insert_quorum_with_keeper_fail ON CLUSTER cluster" ) def query( > return self.client.query( E AttributeError: 'NoneType' object has no attribute 'query' CI: https://s3.amazonaws.com/clickhouse-test-reports/71406/8b3ce129456a1f85839a48538780639e2e3c3020/integration_tests__asan__old_analyzer__[6_6]//home/ubuntu/actions-runner/_work/_temp/test/output_dir/integration_run_parallel3_0.log Signed-off-by: Azat Khuzhin --- tests/integration/test_quorum_inserts/test.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index eefc4882e8e..66f96d61b3e 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -366,7 +366,7 @@ def test_insert_quorum_with_ttl(started_cluster): zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster") -def test_insert_quorum_with_keeper_loss_connection(): +def test_insert_quorum_with_keeper_loss_connection(started_cluster): zero.query( "DROP TABLE IF EXISTS test_insert_quorum_with_keeper_fail ON CLUSTER cluster" ) From 097b45bf5af2d32c4a816c9208c65dab60f2da18 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Mon, 4 Nov 2024 13:56:40 +0000 Subject: [PATCH 408/680] small refactoring --- tests/ci/libfuzzer_test_check.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/libfuzzer_test_check.py b/tests/ci/libfuzzer_test_check.py index d0936eb2323..2616fbe3f5d 100644 --- a/tests/ci/libfuzzer_test_check.py +++ b/tests/ci/libfuzzer_test_check.py @@ -21,8 +21,8 @@ from s3_helper import S3Helper from stopwatch import Stopwatch from tee_popen import TeePopen -NO_CHANGES_MSG = "Nothing to run" TIMEOUT = 60 +NO_CHANGES_MSG = "Nothing to run" s3 = S3Helper() From 978cf9a90525e7162f7841d794ec20d1096c84ed Mon Sep 17 00:00:00 2001 From: alesapin Date: Mon, 4 Nov 2024 15:32:55 +0100 Subject: [PATCH 409/680] Add per host dashboards to advanced dashboard --- .../System/StorageSystemDashboards.cpp | 490 +++++++++++++++++- 1 file changed, 489 insertions(+), 1 deletion(-) diff --git a/src/Storages/System/StorageSystemDashboards.cpp b/src/Storages/System/StorageSystemDashboards.cpp index 96ba7e59cf2..340117d1494 100644 --- a/src/Storages/System/StorageSystemDashboards.cpp +++ b/src/Storages/System/StorageSystemDashboards.cpp @@ -227,6 +227,194 @@ FROM merge('system', '^metric_log') WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + /// Default per host dashboard for self-managed ClickHouse + { + { "dashboard", "Overview (host)" }, + { "title", "Queries/second" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(ProfileEvent_Query) +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "CPU Usage (cores)" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(ProfileEvent_OSCPUVirtualTimeMicroseconds) / 1000000 +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "Queries Running" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(CurrentMetric_Query) +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "Merges Running" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(CurrentMetric_Merge) +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "Selected Bytes/second" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(ProfileEvent_SelectedBytes) +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "IO Wait" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(ProfileEvent_OSIOWaitMicroseconds) / 1000000 +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "CPU Wait" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(ProfileEvent_OSCPUWaitMicroseconds) / 1000000 +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "OS CPU Usage (Userspace)" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(value) +FROM merge('system', '^asynchronous_metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} AND metric = 'OSUserTimeNormalized' +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "OS CPU Usage (Kernel)" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(value) +FROM merge('system', '^asynchronous_metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} AND metric = 'OSSystemTimeNormalized' +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "Read From Disk" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(ProfileEvent_OSReadBytes) +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "Read From Filesystem" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(ProfileEvent_OSReadChars) +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "Memory (tracked)" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(CurrentMetric_MemoryTracking) +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "Load Average (15 minutes)" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(value) +FROM merge('system', '^asynchronous_metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} AND metric = 'LoadAverage15' +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "Selected Rows/second" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(ProfileEvent_SelectedRows) +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "Inserted Rows/second" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(ProfileEvent_InsertedRows) +FROM merge('system', '^metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "Total MergeTree Parts" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, avg(value) +FROM merge('system', '^asynchronous_metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} AND metric = 'TotalPartsOfMergeTreeTables' +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} +)EOQ") } + }, + { + { "dashboard", "Overview (host)" }, + { "title", "Max Parts For Partition" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t, hostname, max(value) +FROM merge('system', '^asynchronous_metric_log') +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} AND metric = 'MaxPartCountForPartition' +GROUP BY t, hostname +ORDER BY t WITH FILL STEP {rounding:UInt32} )EOQ") } }, /// Default dashboard for ClickHouse Cloud @@ -369,7 +557,307 @@ ORDER BY t WITH FILL STEP {rounding:UInt32} { "dashboard", "Cloud overview" }, { "title", "Concurrent network connections" }, { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, max(TCP_Connections), max(MySQL_Connections), max(HTTP_Connections) FROM (SELECT event_time, sum(CurrentMetric_TCPConnection) AS TCP_Connections, sum(CurrentMetric_MySQLConnection) AS MySQL_Connections, sum(CurrentMetric_HTTPConnection) AS HTTP_Connections FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - } + }, + /// Default per host dashboard for ClickHouse Cloud + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Queries/second" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_Query) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "CPU Usage (cores)" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, hostname, avg(metric) / 1000000\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_OSCPUVirtualTimeMicroseconds) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32} GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Queries Running" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(CurrentMetric_Query) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Merges Running" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(CurrentMetric_Merge) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Selected Bytes/second" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_SelectedBytes) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "IO Wait (local fs)" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_OSIOWaitMicroseconds) / 1000000 AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "S3 read wait" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_ReadBufferFromS3Microseconds) / 1000000 AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "S3 read errors/sec" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_ReadBufferFromS3RequestsErrors) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "CPU Wait" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_OSCPUWaitMicroseconds) / 1000000 AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "OS CPU Usage (Userspace, normalized)" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, hostname, avg(value)\nFROM clusterAllReplicas(default, merge('system', '^asynchronous_metric_log'))\nWHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32}\nAND metric = 'OSUserTimeNormalized'\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "OS CPU Usage (Kernel, normalized)" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, hostname, avg(value)\nFROM clusterAllReplicas(default, merge('system', '^asynchronous_metric_log'))\nWHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32}\nAND metric = 'OSSystemTimeNormalized'\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Read From Disk (bytes/sec)" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_OSReadBytes) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Read From Filesystem (bytes/sec)" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_OSReadChars) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Memory (tracked, bytes)" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(CurrentMetric_MemoryTracking) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Load Average (15 minutes)" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, hostname, avg(value)\nFROM (\n SELECT event_time, hostname, sum(value) AS value\n FROM clusterAllReplicas(default, merge('system', '^asynchronous_metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n AND metric = 'LoadAverage15'\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Selected Rows/sec" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_SelectedRows) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Inserted Rows/sec" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_InsertedRows) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Total MergeTree Parts" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, hostname, max(value)\nFROM clusterAllReplicas(default, merge('system', '^asynchronous_metric_log'))\nWHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32}\nAND metric = 'TotalPartsOfMergeTreeTables'\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Max Parts For Partition" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, hostname, max(value)\nFROM clusterAllReplicas(default, merge('system', '^asynchronous_metric_log'))\nWHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32}\nAND metric = 'MaxPartCountForPartition'\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Read From S3 (bytes/sec)" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_ReadBufferFromS3Bytes) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Filesystem Cache Size" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(CurrentMetric_FilesystemCacheSize) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Disk S3 write req/sec" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT as t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_DiskS3PutObject + ProfileEvent_DiskS3UploadPart + ProfileEvent_DiskS3CreateMultipartUpload + ProfileEvent_DiskS3CompleteMultipartUpload) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\n GROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Disk S3 read req/sec" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_DiskS3GetObject + ProfileEvent_DiskS3HeadObject + ProfileEvent_DiskS3ListObjects) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\nGROUP BY t, hostname\nORDER BY t\nWITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "FS cache hit rate" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, sum(ProfileEvent_CachedReadBufferReadFromCacheBytes) / (sum(ProfileEvent_CachedReadBufferReadFromCacheBytes) + sum(ProfileEvent_CachedReadBufferReadFromSourceBytes)) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\nGROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Page cache hit rate" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t,\n hostname,\n avg(metric)\nFROM (\n SELECT event_time, hostname, greatest(0, (sum(ProfileEvent_OSReadChars) - sum(ProfileEvent_OSReadBytes)) / (sum(ProfileEvent_OSReadChars) + sum(ProfileEvent_ReadBufferFromS3Bytes))) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time, hostname)\nGROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Network receive bytes/sec" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, hostname, avg(value)\nFROM (\n SELECT event_time, hostname, sum(value) AS value\n FROM clusterAllReplicas(default, merge('system', '^asynchronous_metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n AND metric LIKE 'NetworkReceiveBytes%'\n GROUP BY event_time, hostname)\nGROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Cloud overview (host)" }, + { "title", "Network send bytes/sec" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, hostname, avg(value)\nFROM (\n SELECT event_time, hostname, sum(value) AS value\n FROM clusterAllReplicas(default, merge('system', '^asynchronous_metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n AND metric LIKE 'NetworkSendBytes%'\n GROUP BY event_time, hostname)\nGROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + /// Distributed cache client metrics start + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Read from Distributed Cache (bytes/sec)" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCacheReadBytesFromCache) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Read from Distributed Cache fallback buffer (bytes/sec)" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCacheReadBytesFromFallbackBuffer) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Read From Filesystem (no Distributed Cache) (bytes/sec)" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t,\n avg(metric)\nFROM (\n SELECT event_time, sum(ProfileEvent_OSReadChars) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time)\nGROUP BY t\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Read From S3 (no Distributed Cache) (bytes/sec)" }, + { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t,\n avg(metric)\nFROM (\n SELECT event_time, sum(ProfileEvent_ReadBufferFromS3Bytes) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time)\nGROUP BY t\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Distributed Cache read requests" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(CurrentMetric_DistrCacheReadRequests) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Distributed Cache write requests" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(CurrentMetric_DistrCacheWriteRequests) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Distributed Cache open connections" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(CurrentMetric_DistrCacheOpenedConnections) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Distributed Cache registered servers" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(CurrentMetric_DistrCacheRegisteredServersCurrentAZ) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Distributed Cache read errors" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCacheReadErrors) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Distributed Cache make request errors" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) +FROM (SELECT event_time, sum(ProfileEvent_DistrCacheMakeRequestErrors) AS metric FROM clusterAllReplicas(default, system.metric_log) +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY event_time) +GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 +)EOQ") } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Distributed Cache receive response errors" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) +FROM (SELECT event_time, sum(ProfileEvent_DistrCacheReceiveResponseErrors) AS metric FROM clusterAllReplicas(default, system.metric_log) +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY event_time) +GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 +)EOQ") } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Distributed Cache registry updates" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, max(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCacheHashRingRebuilds) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Distributed Cache packets" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCachePackets) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + { + { "dashboard", "Distributed cache client overview" }, + { "title", "Distributed Cache unused packets" }, + { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCacheUnusedPackets) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } + }, + /// Distributed cache client metrics end + /// + /// Distributed cache server metrics start + { + { "dashboard", "Distributed cache server overview" }, + { "title", "Distributed Cache open connections" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) +FROM (SELECT event_time, sum(CurrentMetric_DistrCacheServerConnections) AS metric FROM clusterAllReplicas(default, system.metric_log) +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY event_time) +GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 +)EOQ") } + }, + { + { "dashboard", "Distributed cache server overview" }, + { "title", "Distributed Cache StartRequest packets" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) +FROM (SELECT event_time, sum(ProfileEvent_DistrCacheServerStartRequestPackets) AS metric FROM clusterAllReplicas(default, system.metric_log) +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY event_time) +GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 +)EOQ") } + }, + { + { "dashboard", "Distributed cache server overview" }, + { "title", "Distributed Cache ContinueRequest packets" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) +FROM (SELECT event_time, sum(ProfileEvent_DistrCacheServerContinueRequestPackets) AS metric FROM clusterAllReplicas(default, system.metric_log) +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY event_time) +GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 +)EOQ") } + }, + { + { "dashboard", "Distributed cache server overview" }, + { "title", "Distributed Cache EndRequest packets" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) +FROM (SELECT event_time, sum(ProfileEvent_DistrCacheServerEndRequestPackets) AS metric FROM clusterAllReplicas(default, system.metric_log) +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY event_time) +GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 +)EOQ") } + }, + { + { "dashboard", "Distributed cache server overview" }, + { "title", "Distributed Cache AckRequest packets" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) +FROM (SELECT event_time, sum(ProfileEvent_DistrCacheServerAckRequestPackets) AS metric FROM clusterAllReplicas(default, system.metric_log) +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY event_time) +GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 +)EOQ") } + }, + { + { "dashboard", "Distributed cache server overview" }, + { "title", "Distributed Cache reused s3 clients" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) +FROM (SELECT event_time, sum(ProfileEvent_DistrCacheServerReusedS3CachedClients) AS metric FROM clusterAllReplicas(default, system.metric_log) +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY event_time) +GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 +)EOQ") } + }, + { + { "dashboard", "Distributed cache server overview" }, + { "title", "Distributed Cache new s3 clients" }, + { "query", trim(R"EOQ( +SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) +FROM (SELECT event_time, sum(ProfileEvent_DistrCacheNewS3CachedClients) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) +WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} +GROUP BY event_time) +GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 +)EOQ") } + }, + /// Distributed cache server metrics end }; auto add_dashboards = [&](const auto & dashboards) From 1976c399ca8f58283ac97d1a47749cb5e6072649 Mon Sep 17 00:00:00 2001 From: alesapin Date: Mon, 4 Nov 2024 15:34:30 +0100 Subject: [PATCH 410/680] Remove redundant changes --- .../System/StorageSystemDashboards.cpp | 164 ------------------ 1 file changed, 164 deletions(-) diff --git a/src/Storages/System/StorageSystemDashboards.cpp b/src/Storages/System/StorageSystemDashboards.cpp index 340117d1494..27579da4bfe 100644 --- a/src/Storages/System/StorageSystemDashboards.cpp +++ b/src/Storages/System/StorageSystemDashboards.cpp @@ -694,170 +694,6 @@ ORDER BY t WITH FILL STEP {rounding:UInt32} { "title", "Network send bytes/sec" }, { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, hostname, avg(value)\nFROM (\n SELECT event_time, hostname, sum(value) AS value\n FROM clusterAllReplicas(default, merge('system', '^asynchronous_metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n AND metric LIKE 'NetworkSendBytes%'\n GROUP BY event_time, hostname)\nGROUP BY t, hostname\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } }, - /// Distributed cache client metrics start - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Read from Distributed Cache (bytes/sec)" }, - { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCacheReadBytesFromCache) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Read from Distributed Cache fallback buffer (bytes/sec)" }, - { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCacheReadBytesFromFallbackBuffer) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Read From Filesystem (no Distributed Cache) (bytes/sec)" }, - { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t,\n avg(metric)\nFROM (\n SELECT event_time, sum(ProfileEvent_OSReadChars) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time)\nGROUP BY t\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Read From S3 (no Distributed Cache) (bytes/sec)" }, - { "query", "SELECT \n toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t,\n avg(metric)\nFROM (\n SELECT event_time, sum(ProfileEvent_ReadBufferFromS3Bytes) AS metric \n FROM clusterAllReplicas(default, merge('system', '^metric_log'))\n WHERE event_date >= toDate(now() - {seconds:UInt32})\n AND event_time >= now() - {seconds:UInt32}\n GROUP BY event_time)\nGROUP BY t\nORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Distributed Cache read requests" }, - { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(CurrentMetric_DistrCacheReadRequests) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Distributed Cache write requests" }, - { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(CurrentMetric_DistrCacheWriteRequests) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Distributed Cache open connections" }, - { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(CurrentMetric_DistrCacheOpenedConnections) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Distributed Cache registered servers" }, - { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(CurrentMetric_DistrCacheRegisteredServersCurrentAZ) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Distributed Cache read errors" }, - { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCacheReadErrors) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Distributed Cache make request errors" }, - { "query", trim(R"EOQ( -SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) -FROM (SELECT event_time, sum(ProfileEvent_DistrCacheMakeRequestErrors) AS metric FROM clusterAllReplicas(default, system.metric_log) -WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} -GROUP BY event_time) -GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 -)EOQ") } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Distributed Cache receive response errors" }, - { "query", trim(R"EOQ( -SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) -FROM (SELECT event_time, sum(ProfileEvent_DistrCacheReceiveResponseErrors) AS metric FROM clusterAllReplicas(default, system.metric_log) -WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} -GROUP BY event_time) -GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 -)EOQ") } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Distributed Cache registry updates" }, - { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, max(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCacheHashRingRebuilds) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Distributed Cache packets" }, - { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCachePackets) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - { - { "dashboard", "Distributed cache client overview" }, - { "title", "Distributed Cache unused packets" }, - { "query", "SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) FROM (SELECT event_time, sum(ProfileEvent_DistrCacheUnusedPackets) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} GROUP BY event_time) GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1" } - }, - /// Distributed cache client metrics end - /// - /// Distributed cache server metrics start - { - { "dashboard", "Distributed cache server overview" }, - { "title", "Distributed Cache open connections" }, - { "query", trim(R"EOQ( -SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) -FROM (SELECT event_time, sum(CurrentMetric_DistrCacheServerConnections) AS metric FROM clusterAllReplicas(default, system.metric_log) -WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} -GROUP BY event_time) -GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 -)EOQ") } - }, - { - { "dashboard", "Distributed cache server overview" }, - { "title", "Distributed Cache StartRequest packets" }, - { "query", trim(R"EOQ( -SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) -FROM (SELECT event_time, sum(ProfileEvent_DistrCacheServerStartRequestPackets) AS metric FROM clusterAllReplicas(default, system.metric_log) -WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} -GROUP BY event_time) -GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 -)EOQ") } - }, - { - { "dashboard", "Distributed cache server overview" }, - { "title", "Distributed Cache ContinueRequest packets" }, - { "query", trim(R"EOQ( -SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) -FROM (SELECT event_time, sum(ProfileEvent_DistrCacheServerContinueRequestPackets) AS metric FROM clusterAllReplicas(default, system.metric_log) -WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} -GROUP BY event_time) -GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 -)EOQ") } - }, - { - { "dashboard", "Distributed cache server overview" }, - { "title", "Distributed Cache EndRequest packets" }, - { "query", trim(R"EOQ( -SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) -FROM (SELECT event_time, sum(ProfileEvent_DistrCacheServerEndRequestPackets) AS metric FROM clusterAllReplicas(default, system.metric_log) -WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} -GROUP BY event_time) -GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 -)EOQ") } - }, - { - { "dashboard", "Distributed cache server overview" }, - { "title", "Distributed Cache AckRequest packets" }, - { "query", trim(R"EOQ( -SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) -FROM (SELECT event_time, sum(ProfileEvent_DistrCacheServerAckRequestPackets) AS metric FROM clusterAllReplicas(default, system.metric_log) -WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} -GROUP BY event_time) -GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 -)EOQ") } - }, - { - { "dashboard", "Distributed cache server overview" }, - { "title", "Distributed Cache reused s3 clients" }, - { "query", trim(R"EOQ( -SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) -FROM (SELECT event_time, sum(ProfileEvent_DistrCacheServerReusedS3CachedClients) AS metric FROM clusterAllReplicas(default, system.metric_log) -WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} -GROUP BY event_time) -GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 -)EOQ") } - }, - { - { "dashboard", "Distributed cache server overview" }, - { "title", "Distributed Cache new s3 clients" }, - { "query", trim(R"EOQ( -SELECT toStartOfInterval(event_time, INTERVAL {rounding:UInt32} SECOND)::INT AS t, avg(metric) -FROM (SELECT event_time, sum(ProfileEvent_DistrCacheNewS3CachedClients) AS metric FROM clusterAllReplicas(default, merge('system', '^metric_log')) -WHERE event_date >= toDate(now() - {seconds:UInt32}) AND event_time >= now() - {seconds:UInt32} -GROUP BY event_time) -GROUP BY t ORDER BY t WITH FILL STEP {rounding:UInt32} SETTINGS skip_unavailable_shards = 1 -)EOQ") } - }, - /// Distributed cache server metrics end }; auto add_dashboards = [&](const auto & dashboards) From 47b1b2c1584babf53eadf062a8421e1ce481580c Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Mon, 4 Nov 2024 14:51:43 +0000 Subject: [PATCH 411/680] Try fix integration test - second attempt --- tests/integration/test_quorum_inserts/test.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index 66f96d61b3e..f64864185c5 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -368,7 +368,7 @@ def test_insert_quorum_with_ttl(started_cluster): def test_insert_quorum_with_keeper_loss_connection(started_cluster): zero.query( - "DROP TABLE IF EXISTS test_insert_quorum_with_keeper_fail ON CLUSTER cluster" + "DROP TABLE IF EXISTS test_insert_quorum_with_keeper_loss ON CLUSTER cluster" ) create_query = ( "CREATE TABLE test_insert_quorum_with_keeper_loss" From 1d888bc1ebc762faf1136d6910fef8641216fb6e Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Mon, 4 Nov 2024 16:40:26 +0100 Subject: [PATCH 412/680] Fix wrong change --- src/Interpreters/Cache/FileSegment.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Interpreters/Cache/FileSegment.cpp b/src/Interpreters/Cache/FileSegment.cpp index 080b54feb06..9c8f041fabf 100644 --- a/src/Interpreters/Cache/FileSegment.cpp +++ b/src/Interpreters/Cache/FileSegment.cpp @@ -139,7 +139,7 @@ FileSegmentGuard::Lock FileSegment::lock() const void FileSegment::setDownloadState(State state, const FileSegmentGuard::Lock & lock) { - if (isCompleted(false)) + if (isCompleted(false) && state != State::DETACHED) { throw Exception( ErrorCodes::LOGICAL_ERROR, From 929da1411e5357d7a99210a4b6f617a2f66f933e Mon Sep 17 00:00:00 2001 From: vdimir Date: Mon, 4 Nov 2024 16:06:20 +0000 Subject: [PATCH 413/680] Fix crash in mongodb table function --- src/TableFunctions/TableFunctionMongoDB.cpp | 10 +++++++--- .../TableFunctionMongoDBPocoLegacy.cpp | 8 +++++--- .../03261_mongodb_argumetns_crash.reference | 0 .../0_stateless/03261_mongodb_argumetns_crash.sql | 13 +++++++++++++ 4 files changed, 25 insertions(+), 6 deletions(-) create mode 100644 tests/queries/0_stateless/03261_mongodb_argumetns_crash.reference create mode 100644 tests/queries/0_stateless/03261_mongodb_argumetns_crash.sql diff --git a/src/TableFunctions/TableFunctionMongoDB.cpp b/src/TableFunctions/TableFunctionMongoDB.cpp index e13427c1557..966ce858875 100644 --- a/src/TableFunctions/TableFunctionMongoDB.cpp +++ b/src/TableFunctions/TableFunctionMongoDB.cpp @@ -118,14 +118,18 @@ void TableFunctionMongoDB::parseArguments(const ASTPtr & ast_function, ContextPt if (const auto * ast_func = typeid_cast(args[i].get())) { const auto * args_expr = assert_cast(ast_func->arguments.get()); - auto function_args = args_expr->children; - if (function_args.size() != 2) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument"); + const auto & function_args = args_expr->children; + if (function_args.size() != 2 || ast_func->name != "equals" || function_args[0]->as()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument, got {}", ast_func->formatForErrorMessage()); auto arg_name = function_args[0]->as()->name(); if (arg_name == "structure") structure = checkAndGetLiteralArgument(function_args[1], "structure"); + else if (arg_name == "options") + main_arguments.push_back(function_args[1]); + else + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument, got {}", ast_func->formatForErrorMessage()); } else if (i == 2) { diff --git a/src/TableFunctions/TableFunctionMongoDBPocoLegacy.cpp b/src/TableFunctions/TableFunctionMongoDBPocoLegacy.cpp index dc1df7fcad8..70b28ddfaf0 100644 --- a/src/TableFunctions/TableFunctionMongoDBPocoLegacy.cpp +++ b/src/TableFunctions/TableFunctionMongoDBPocoLegacy.cpp @@ -98,9 +98,9 @@ void TableFunctionMongoDBPocoLegacy::parseArguments(const ASTPtr & ast_function, if (const auto * ast_func = typeid_cast(args[i].get())) { const auto * args_expr = assert_cast(ast_func->arguments.get()); - auto function_args = args_expr->children; - if (function_args.size() != 2) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument"); + const auto & function_args = args_expr->children; + if (function_args.size() != 2 || ast_func->name != "equals" || function_args[0]->as()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument, got {}", ast_func->formatForErrorMessage()); auto arg_name = function_args[0]->as()->name(); @@ -108,6 +108,8 @@ void TableFunctionMongoDBPocoLegacy::parseArguments(const ASTPtr & ast_function, structure = checkAndGetLiteralArgument(function_args[1], "structure"); else if (arg_name == "options") main_arguments.push_back(function_args[1]); + else + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument, got {}", ast_func->formatForErrorMessage()); } else if (i == 5) { diff --git a/tests/queries/0_stateless/03261_mongodb_argumetns_crash.reference b/tests/queries/0_stateless/03261_mongodb_argumetns_crash.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03261_mongodb_argumetns_crash.sql b/tests/queries/0_stateless/03261_mongodb_argumetns_crash.sql new file mode 100644 index 00000000000..830d3995bd5 --- /dev/null +++ b/tests/queries/0_stateless/03261_mongodb_argumetns_crash.sql @@ -0,0 +1,13 @@ +-- Tags: no-fasttest + +SELECT * FROM mongodb('mongodb://some-cluster:27017/?retryWrites=false', NULL, 'my_collection', 'test_user', 'password', 'x Int32'); -- { serverError BAD_ARGUMENTS } +SELECT * FROM mongodb('mongodb://some-cluster:27017/?retryWrites=false', 'test', NULL, 'test_user', 'password', 'x Int32'); -- { serverError BAD_ARGUMENTS } +SELECT * FROM mongodb('mongodb://some-cluster:27017/?retryWrites=false', 'test', 'my_collection', NULL, 'password', 'x Int32'); -- { serverError BAD_ARGUMENTS } +SELECT * FROM mongodb('mongodb://some-cluster:27017/?retryWrites=false', 'test', 'my_collection', 'test_user', NULL, 'x Int32'); -- { serverError BAD_ARGUMENTS } +SELECT * FROM mongodb('mongodb://some-cluster:27017/?retryWrites=false', 'test', 'my_collection', 'test_user', 'password', NULL); -- { serverError BAD_ARGUMENTS } +SELECT * FROM mongodb('mongodb://some-cluster:27017/?retryWrites=false', 'test', 'my_collection', 'test_user', 'password', materialize(1) + 1); -- { serverError BAD_ARGUMENTS } +SELECT * FROM mongodb('mongodb://some-cluster:27017/?retryWrites=false', 'test', 'my_collection', 'test_user', 'password', 'x Int32', NULL); -- { serverError BAD_ARGUMENTS } +SELECT * FROM mongodb('mongodb://some-cluster:27017/?retryWrites=false', 'test', 'my_collection', 'test_user', 'password', NULL, 'x Int32'); -- { serverError BAD_ARGUMENTS } +SELECT * FROM mongodb('mongodb://some-cluster:27017/?retryWrites=false', 'test', 'my_collection', 'test_user', 'password', NULL, 'x Int32'); -- { serverError BAD_ARGUMENTS } +SELECT * FROM mongodb(NULL, 'test', 'my_collection', 'test_user', 'password', 'x Int32'); -- { serverError BAD_ARGUMENTS } + From 24017bb7add084f38022c2cf1a1fa9a96788ebc9 Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Mon, 4 Nov 2024 17:31:39 +0100 Subject: [PATCH 414/680] add parallel_replicas_prefer_local_join --- ...eplicas_join_algo_and_analyzer_4.reference | 58 +++++++++++++++++++ ...allel_replicas_join_algo_and_analyzer_4.sh | 34 ++++++----- 2 files changed, 77 insertions(+), 15 deletions(-) diff --git a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference index 8464317f7e6..52c4e872f84 100644 --- a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference +++ b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference @@ -56,3 +56,61 @@ SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` GROUP 500030000 500040000 SELECT sum(`__table1`.`item_id`) AS `sum(item_id)` FROM (SELECT `__table2`.`item_id` AS `item_id`, `__table2`.`price_sold` AS `price_sold` FROM `default`.`t` AS `__table2`) AS `__table1` ALL LEFT JOIN (SELECT `__table4`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table4`) AS `__table3` ON `__table1`.`item_id` = `__table3`.`item_id` GROUP BY `__table1`.`price_sold` ORDER BY `__table1`.`price_sold` ASC +4999950000 +4999950000 +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t` AS `__table1` GROUP BY `__table1`.`item_id` +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` +4999950000 +4999950000 +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t` AS `__table1` +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` GROUP BY `__table1`.`item_id` +499950000 +499960000 +499970000 +499980000 +499990000 +500000000 +500010000 +500020000 +500030000 +500040000 +499950000 +499960000 +499970000 +499980000 +499990000 +500000000 +500010000 +500020000 +500030000 +500040000 +SELECT sum(`__table1`.`item_id`) AS `sum(item_id)` FROM (SELECT `__table2`.`item_id` AS `item_id`, `__table2`.`price_sold` AS `price_sold` FROM `default`.`t` AS `__table2`) AS `__table1` GLOBAL ALL LEFT JOIN `_data_4551627371769371400_3093038500622465792` AS `__table3` ON `__table1`.`item_id` = `__table3`.`item_id` GROUP BY `__table1`.`price_sold` ORDER BY `__table1`.`price_sold` ASC +4999950000 +4999950000 +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t` AS `__table1` GROUP BY `__table1`.`item_id` +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` +4999950000 +4999950000 +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t` AS `__table1` +SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` GROUP BY `__table1`.`item_id` +499950000 +499960000 +499970000 +499980000 +499990000 +500000000 +500010000 +500020000 +500030000 +500040000 +499950000 +499960000 +499970000 +499980000 +499990000 +500000000 +500010000 +500020000 +500030000 +500040000 +SELECT sum(`__table1`.`item_id`) AS `sum(item_id)` FROM (SELECT `__table2`.`item_id` AS `item_id`, `__table2`.`price_sold` AS `price_sold` FROM `default`.`t` AS `__table2`) AS `__table1` GLOBAL ALL LEFT JOIN `_data_4551627371769371400_3093038500622465792` AS `__table3` ON `__table1`.`item_id` = `__table3`.`item_id` GROUP BY `__table1`.`price_sold` ORDER BY `__table1`.`price_sold` ASC diff --git a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh index 0e1f07b6ac5..18a2fbd317b 100755 --- a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh +++ b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh @@ -75,23 +75,27 @@ query3=" ORDER BY price_sold " -for prefer_local_plan in {0..1}; do - for query in "${query1}" "${query2}" "${query3}"; do - for enable_parallel_replicas in {0..1}; do - ${CLICKHOUSE_CLIENT} --query=" - set enable_analyzer=1; - set parallel_replicas_local_plan=${prefer_local_plan}; - set allow_experimental_parallel_reading_from_replicas=${enable_parallel_replicas}, cluster_for_parallel_replicas='parallel_replicas', max_parallel_replicas=100, parallel_replicas_for_non_replicated_merge_tree=1; +for parallel_replicas_prefer_local_join in 1 0; do + for prefer_local_plan in {0..1}; do + for query in "${query1}" "${query2}" "${query3}"; do + for enable_parallel_replicas in {0..1}; do + ${CLICKHOUSE_CLIENT} --query=" + set enable_analyzer=1; + set parallel_replicas_prefer_local_join=${parallel_replicas_prefer_local_join}; + set parallel_replicas_local_plan=${prefer_local_plan}; + set allow_experimental_parallel_reading_from_replicas=${enable_parallel_replicas}, cluster_for_parallel_replicas='parallel_replicas', max_parallel_replicas=100, parallel_replicas_for_non_replicated_merge_tree=1; - ${query}; + --SELECT '----- enable_parallel_replicas=$enable_parallel_replicas prefer_local_plan=$prefer_local_plan parallel_replicas_prefer_local_join=$parallel_replicas_prefer_local_join -----'; + ${query}; - SELECT replaceRegexpAll(explain, '.*Query: (.*) Replicas:.*', '\\1') - FROM - ( - EXPLAIN actions=1 ${query} - ) - WHERE explain LIKE '%ParallelReplicas%'; - " + SELECT replaceRegexpAll(explain, '.*Query: (.*) Replicas:.*', '\\1') + FROM + ( + EXPLAIN actions=1 ${query} + ) + WHERE explain LIKE '%ParallelReplicas%'; + " + done done done done From 6b4d44be2894bf99897fca011817c9d77bfbabdf Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Mon, 4 Nov 2024 16:42:06 +0000 Subject: [PATCH 415/680] Update version_date.tsv and changelogs after v24.8.6.70-lts --- SECURITY.md | 3 +- docker/keeper/Dockerfile | 2 +- docker/server/Dockerfile.alpine | 2 +- docker/server/Dockerfile.ubuntu | 2 +- docs/changelogs/v24.8.6.70-lts.md | 50 ++++++++++++++++++++++++++++ utils/list-versions/version_date.tsv | 2 ++ 6 files changed, 57 insertions(+), 4 deletions(-) create mode 100644 docs/changelogs/v24.8.6.70-lts.md diff --git a/SECURITY.md b/SECURITY.md index db302da8ecd..1b0648dc489 100644 --- a/SECURITY.md +++ b/SECURITY.md @@ -14,9 +14,10 @@ The following versions of ClickHouse server are currently supported with securit | Version | Supported | |:-|:-| +| 24.10 | ✔️ | | 24.9 | ✔️ | | 24.8 | ✔️ | -| 24.7 | ✔️ | +| 24.7 | ❌ | | 24.6 | ❌ | | 24.5 | ❌ | | 24.4 | ❌ | diff --git a/docker/keeper/Dockerfile b/docker/keeper/Dockerfile index dfe6a420260..bc76bdbb619 100644 --- a/docker/keeper/Dockerfile +++ b/docker/keeper/Dockerfile @@ -34,7 +34,7 @@ RUN arch=${TARGETARCH:-amd64} \ # lts / testing / prestable / etc ARG REPO_CHANNEL="stable" ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}" -ARG VERSION="24.9.2.42" +ARG VERSION="24.10.1.2812" ARG PACKAGES="clickhouse-keeper" ARG DIRECT_DOWNLOAD_URLS="" diff --git a/docker/server/Dockerfile.alpine b/docker/server/Dockerfile.alpine index 991c25ad142..93acf1a5773 100644 --- a/docker/server/Dockerfile.alpine +++ b/docker/server/Dockerfile.alpine @@ -35,7 +35,7 @@ RUN arch=${TARGETARCH:-amd64} \ # lts / testing / prestable / etc ARG REPO_CHANNEL="stable" ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}" -ARG VERSION="24.9.2.42" +ARG VERSION="24.10.1.2812" ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static" ARG DIRECT_DOWNLOAD_URLS="" diff --git a/docker/server/Dockerfile.ubuntu b/docker/server/Dockerfile.ubuntu index 5dc88b49e31..506a627b11c 100644 --- a/docker/server/Dockerfile.ubuntu +++ b/docker/server/Dockerfile.ubuntu @@ -28,7 +28,7 @@ RUN sed -i "s|http://archive.ubuntu.com|${apt_archive}|g" /etc/apt/sources.list ARG REPO_CHANNEL="stable" ARG REPOSITORY="deb [signed-by=/usr/share/keyrings/clickhouse-keyring.gpg] https://packages.clickhouse.com/deb ${REPO_CHANNEL} main" -ARG VERSION="24.9.2.42" +ARG VERSION="24.10.1.2812" ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static" #docker-official-library:off diff --git a/docs/changelogs/v24.8.6.70-lts.md b/docs/changelogs/v24.8.6.70-lts.md new file mode 100644 index 00000000000..81fa4db1458 --- /dev/null +++ b/docs/changelogs/v24.8.6.70-lts.md @@ -0,0 +1,50 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.8.6.70-lts (ddb8c219771) FIXME as compared to v24.8.5.115-lts (8c4cb00a384) + +#### Backward Incompatible Change +* Backported in [#71359](https://github.com/ClickHouse/ClickHouse/issues/71359): Fix possible error `No such file or directory` due to unescaped special symbols in files for JSON subcolumns. [#71182](https://github.com/ClickHouse/ClickHouse/pull/71182) ([Pavel Kruglov](https://github.com/Avogar)). + +#### Improvement +* Backported in [#70680](https://github.com/ClickHouse/ClickHouse/issues/70680): Don't do validation when synchronizing user_directories from keeper. [#70644](https://github.com/ClickHouse/ClickHouse/pull/70644) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#71395](https://github.com/ClickHouse/ClickHouse/issues/71395): Do not call the object storage API when listing directories, as this may be cost-inefficient. Instead, store the list of filenames in the memory. The trade-offs are increased initial load time and memory required to store filenames. [#70823](https://github.com/ClickHouse/ClickHouse/pull/70823) ([Julia Kartseva](https://github.com/jkartseva)). +* Backported in [#71287](https://github.com/ClickHouse/ClickHouse/issues/71287): Reduce the number of object storage HEAD API requests in the plain_rewritable disk. [#70915](https://github.com/ClickHouse/ClickHouse/pull/70915) ([Julia Kartseva](https://github.com/jkartseva)). + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#70934](https://github.com/ClickHouse/ClickHouse/issues/70934): Fix incorrect JOIN ON section optimization in case of `IS NULL` check under any other function (like `NOT`) that may lead to wrong results. Closes [#67915](https://github.com/ClickHouse/ClickHouse/issues/67915). [#68049](https://github.com/ClickHouse/ClickHouse/pull/68049) ([Vladimir Cherkasov](https://github.com/vdimir)). +* Backported in [#70735](https://github.com/ClickHouse/ClickHouse/issues/70735): Fix unexpected exception when passing empty tuple in array. This fixes [#68618](https://github.com/ClickHouse/ClickHouse/issues/68618). [#68848](https://github.com/ClickHouse/ClickHouse/pull/68848) ([Amos Bird](https://github.com/amosbird)). +* Backported in [#71138](https://github.com/ClickHouse/ClickHouse/issues/71138): Fix propogating structure argument in s3Cluster. Previously the `DEFAULT` expression of the column could be lost when sending the query to the replicas in s3Cluster. [#69147](https://github.com/ClickHouse/ClickHouse/pull/69147) ([Pavel Kruglov](https://github.com/Avogar)). +* Backported in [#70561](https://github.com/ClickHouse/ClickHouse/issues/70561): Fix `getSubcolumn` with `LowCardinality` columns by overriding `useDefaultImplementationForLowCardinalityColumns` to return `true`. [#69831](https://github.com/ClickHouse/ClickHouse/pull/69831) ([Miсhael Stetsyuk](https://github.com/mstetsyuk)). +* Backported in [#70903](https://github.com/ClickHouse/ClickHouse/issues/70903): Avoid reusing columns among different named tuples when evaluating `tuple` functions. This fixes [#70022](https://github.com/ClickHouse/ClickHouse/issues/70022). [#70103](https://github.com/ClickHouse/ClickHouse/pull/70103) ([Amos Bird](https://github.com/amosbird)). +* Backported in [#70623](https://github.com/ClickHouse/ClickHouse/issues/70623): Fix server segfault on creating a materialized view with two selects and an `INTERSECT`, e.g. `CREATE MATERIALIZED VIEW v0 AS (SELECT 1) INTERSECT (SELECT 1);`. [#70264](https://github.com/ClickHouse/ClickHouse/pull/70264) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Backported in [#70688](https://github.com/ClickHouse/ClickHouse/issues/70688): Fix possible use-after-free in `SYSTEM DROP FORMAT SCHEMA CACHE FOR Protobuf`. [#70358](https://github.com/ClickHouse/ClickHouse/pull/70358) ([Azat Khuzhin](https://github.com/azat)). +* Backported in [#70494](https://github.com/ClickHouse/ClickHouse/issues/70494): Fix crash during GROUP BY JSON sub-object subcolumn. [#70374](https://github.com/ClickHouse/ClickHouse/pull/70374) ([Pavel Kruglov](https://github.com/Avogar)). +* Backported in [#70482](https://github.com/ClickHouse/ClickHouse/issues/70482): Don't prefetch parts for vertical merges if part has no rows. [#70452](https://github.com/ClickHouse/ClickHouse/pull/70452) ([Antonio Andelic](https://github.com/antonio2368)). +* Backported in [#70556](https://github.com/ClickHouse/ClickHouse/issues/70556): Fix crash in WHERE with lambda functions. [#70464](https://github.com/ClickHouse/ClickHouse/pull/70464) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#70878](https://github.com/ClickHouse/ClickHouse/issues/70878): Fix table creation with `CREATE ... AS table_function()` with database `Replicated` and unavailable table function source on secondary replica. [#70511](https://github.com/ClickHouse/ClickHouse/pull/70511) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#70575](https://github.com/ClickHouse/ClickHouse/issues/70575): Ignore all output on async insert with `wait_for_async_insert=1`. Closes [#62644](https://github.com/ClickHouse/ClickHouse/issues/62644). [#70530](https://github.com/ClickHouse/ClickHouse/pull/70530) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Backported in [#71052](https://github.com/ClickHouse/ClickHouse/issues/71052): Ignore frozen_metadata.txt while traversing shadow directory from system.remote_data_paths. [#70590](https://github.com/ClickHouse/ClickHouse/pull/70590) ([Aleksei Filatov](https://github.com/aalexfvk)). +* Backported in [#70651](https://github.com/ClickHouse/ClickHouse/issues/70651): Fix creation of stateful window functions on misaligned memory. [#70631](https://github.com/ClickHouse/ClickHouse/pull/70631) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#70757](https://github.com/ClickHouse/ClickHouse/issues/70757): Fixed rare crashes in `SELECT`-s and merges after adding a column of `Array` type with non-empty default expression. [#70695](https://github.com/ClickHouse/ClickHouse/pull/70695) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#70763](https://github.com/ClickHouse/ClickHouse/issues/70763): Fix infinite recursion when infering a proto schema with skip unsupported fields enabled. [#70697](https://github.com/ClickHouse/ClickHouse/pull/70697) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#71118](https://github.com/ClickHouse/ClickHouse/issues/71118): `GroupArraySortedData` uses a PODArray with non-POD elements, manually calling constructors and destructors for the elements as needed. But it wasn't careful enough: in two places it forgot to call destructor, in one place it left elements uninitialized if an exception is thrown when deserializing previous elements. Then `GroupArraySortedData`'s destructor called destructors on uninitialized elements and crashed: ``` 2024.10.17 22:58:23.523790 [ 5233 ] {} BaseDaemon: ########## Short fault info ############ 2024.10.17 22:58:23.523834 [ 5233 ] {} BaseDaemon: (version 24.6.1.4609 (official build), build id: 5423339A6571004018D55BBE05D464AFA35E6718, git hash: fa6cdfda8a94890eb19bc7f22f8b0b56292f7a26) (from thread 682) Received signal 11 2024.10.17 22:58:23.523862 [ 5233 ] {} BaseDaemon: Signal description: Segmentation fault 2024.10.17 22:58:23.523883 [ 5233 ] {} BaseDaemon: Address: 0x8f. Access: . Address not mapped to object. 2024.10.17 22:58:23.523908 [ 5233 ] {} BaseDaemon: Stack trace: 0x0000aaaac4b78308 0x0000ffffb7701850 0x0000aaaac0104855 0x0000aaaac01048a0 0x0000aaaac501e84c 0x0000aaaac7c510d0 0x0000aaaac7c4ba20 0x0000aaaac968bbfc 0x0000aaaac968fab0 0x0000aaaac969bf50 0x0000aaaac9b7520c 0x0000aaaac9b74c74 0x0000aaaac9b8a150 0x0000aaaac9b809f0 0x0000aaaac9b80574 0x0000aaaac9b8e364 0x0000aaaac9b8e4fc 0x0000aaaac94f4328 0x0000aaaac94f428c 0x0000aaaac94f7df0 0x0000aaaac98b5a3c 0x0000aaaac950b234 0x0000aaaac49ae264 0x0000aaaac49b1dd0 0x0000aaaac49b0a80 0x0000ffffb755d5c8 0x0000ffffb75c5edc 2024.10.17 22:58:23.523936 [ 5233 ] {} BaseDaemon: ######################################## 2024.10.17 22:58:23.523959 [ 5233 ] {} BaseDaemon: (version 24.6.1.4609 (official build), build id: 5423339A6571004018D55BBE05D464AFA35E6718, git hash: fa6cdfda8a94890eb19bc7f22f8b0b56292f7a26) (from thread 682) (query_id: 6c8a33a2-f45a-4a3b-bd71-ded6a1c9ccd3::202410_534066_534078_2) (query: ) Received signal Segmentation fault (11) 2024.10.17 22:58:23.523977 [ 5233 ] {} BaseDaemon: Address: 0x8f. Access: . Address not mapped to object. 2024.10.17 22:58:23.523993 [ 5233 ] {} BaseDaemon: Stack trace: 0x0000aaaac4b78308 0x0000ffffb7701850 0x0000aaaac0104855 0x0000aaaac01048a0 0x0000aaaac501e84c 0x0000aaaac7c510d0 0x0000aaaac7c4ba20 0x0000aaaac968bbfc 0x0000aaaac968fab0 0x0000aaaac969bf50 0x0000aaaac9b7520c 0x0000aaaac9b74c74 0x0000aaaac9b8a150 0x0000aaaac9b809f0 0x0000aaaac9b80574 0x0000aaaac9b8e364 0x0000aaaac9b8e4fc 0x0000aaaac94f4328 0x0000aaaac94f428c 0x0000aaaac94f7df0 0x0000aaaac98b5a3c 0x0000aaaac950b234 0x0000aaaac49ae264 0x0000aaaac49b1dd0 0x0000aaaac49b0a80 0x0000ffffb755d5c8 0x0000ffffb75c5edc 2024.10.17 22:58:23.524817 [ 5233 ] {} BaseDaemon: 0. signalHandler(int, siginfo_t*, void*) @ 0x000000000c6f8308 2024.10.17 22:58:23.524917 [ 5233 ] {} BaseDaemon: 1. ? @ 0x0000ffffb7701850 2024.10.17 22:58:23.524962 [ 5233 ] {} BaseDaemon: 2. DB::Field::~Field() @ 0x0000000007c84855 2024.10.17 22:58:23.525012 [ 5233 ] {} BaseDaemon: 3. DB::Field::~Field() @ 0x0000000007c848a0 2024.10.17 22:58:23.526626 [ 5233 ] {} BaseDaemon: 4. DB::IAggregateFunctionDataHelper, DB::(anonymous namespace)::GroupArraySorted, DB::Field>>::destroy(char*) const (.5a6a451027f732f9fd91c13f4a13200c) @ 0x000000000cb9e84c 2024.10.17 22:58:23.527322 [ 5233 ] {} BaseDaemon: 5. DB::SerializationAggregateFunction::deserializeBinaryBulk(DB::IColumn&, DB::ReadBuffer&, unsigned long, double) const @ 0x000000000f7d10d0 2024.10.17 22:58:23.528470 [ 5233 ] {} BaseDaemon: 6. DB::ISerialization::deserializeBinaryBulkWithMultipleStreams(COW::immutable_ptr&, unsigned long, DB::ISerialization::DeserializeBinaryBulkSettings&, std::shared_ptr&, std::unordered_map::immutable_ptr, std::hash, std::equal_to, std::allocator::immutable_ptr>>>*) const @ 0x000000000f7cba20 2024.10.17 22:58:23.529213 [ 5233 ] {} BaseDaemon: 7. DB::MergeTreeReaderCompact::readData(DB::NameAndTypePair const&, COW::immutable_ptr&, unsigned long, std::function const&) @ 0x000000001120bbfc 2024.10.17 22:58:23.529277 [ 5233 ] {} BaseDaemon: 8. DB::MergeTreeReaderCompactSingleBuffer::readRows(unsigned long, unsigned long, bool, unsigned long, std::vector::immutable_ptr, std::allocator::immutable_ptr>>&) @ 0x000000001120fab0 2024.10.17 22:58:23.529319 [ 5233 ] {} BaseDaemon: 9. DB::MergeTreeSequentialSource::generate() @ 0x000000001121bf50 2024.10.17 22:58:23.529346 [ 5233 ] {} BaseDaemon: 10. DB::ISource::tryGenerate() @ 0x00000000116f520c 2024.10.17 22:58:23.529653 [ 5233 ] {} BaseDaemon: 11. DB::ISource::work() @ 0x00000000116f4c74 2024.10.17 22:58:23.529679 [ 5233 ] {} BaseDaemon: 12. DB::ExecutionThreadContext::executeTask() @ 0x000000001170a150 2024.10.17 22:58:23.529733 [ 5233 ] {} BaseDaemon: 13. DB::PipelineExecutor::executeStepImpl(unsigned long, std::atomic*) @ 0x00000000117009f0 2024.10.17 22:58:23.529763 [ 5233 ] {} BaseDaemon: 14. DB::PipelineExecutor::executeStep(std::atomic*) @ 0x0000000011700574 2024.10.17 22:58:23.530089 [ 5233 ] {} BaseDaemon: 15. DB::PullingPipelineExecutor::pull(DB::Chunk&) @ 0x000000001170e364 2024.10.17 22:58:23.530277 [ 5233 ] {} BaseDaemon: 16. DB::PullingPipelineExecutor::pull(DB::Block&) @ 0x000000001170e4fc 2024.10.17 22:58:23.530295 [ 5233 ] {} BaseDaemon: 17. DB::MergeTask::ExecuteAndFinalizeHorizontalPart::executeImpl() @ 0x0000000011074328 2024.10.17 22:58:23.530318 [ 5233 ] {} BaseDaemon: 18. DB::MergeTask::ExecuteAndFinalizeHorizontalPart::execute() @ 0x000000001107428c 2024.10.17 22:58:23.530339 [ 5233 ] {} BaseDaemon: 19. DB::MergeTask::execute() @ 0x0000000011077df0 2024.10.17 22:58:23.530362 [ 5233 ] {} BaseDaemon: 20. DB::SharedMergeMutateTaskBase::executeStep() @ 0x0000000011435a3c 2024.10.17 22:58:23.530384 [ 5233 ] {} BaseDaemon: 21. DB::MergeTreeBackgroundExecutor::threadFunction() @ 0x000000001108b234 2024.10.17 22:58:23.530410 [ 5233 ] {} BaseDaemon: 22. ThreadPoolImpl>::worker(std::__list_iterator, void*>) @ 0x000000000c52e264 2024.10.17 22:58:23.530448 [ 5233 ] {} BaseDaemon: 23. void std::__function::__policy_invoker::__call_impl::ThreadFromGlobalPoolImpl>::scheduleImpl(std::function, Priority, std::optional, bool)::'lambda0'()>(void&&)::'lambda'(), void ()>>(std::__function::__policy_storage const*) @ 0x000000000c531dd0 2024.10.17 22:58:23.530476 [ 5233 ] {} BaseDaemon: 24. void* std::__thread_proxy[abi:v15000]>, void ThreadPoolImpl::scheduleImpl(std::function, Priority, std::optional, bool)::'lambda0'()>>(void*) @ 0x000000000c530a80 2024.10.17 22:58:23.530514 [ 5233 ] {} BaseDaemon: 25. ? @ 0x000000000007d5c8 2024.10.17 22:58:23.530534 [ 5233 ] {} BaseDaemon: 26. ? @ 0x00000000000e5edc 2024.10.17 22:58:23.530551 [ 5233 ] {} BaseDaemon: Integrity check of the executable skipped because the reference checksum could not be read. 2024.10.17 22:58:23.531083 [ 5233 ] {} BaseDaemon: Report this error to https://github.com/ClickHouse/ClickHouse/issues 2024.10.17 22:58:23.531294 [ 5233 ] {} BaseDaemon: Changed settings: max_insert_threads = 4, max_threads = 42, use_hedged_requests = false, distributed_foreground_insert = true, alter_sync = 0, enable_memory_bound_merging_of_aggregation_results = true, cluster_for_parallel_replicas = 'default', do_not_merge_across_partitions_select_final = false, log_queries = true, log_queries_probability = 1., max_http_get_redirects = 10, enable_deflate_qpl_codec = false, enable_zstd_qat_codec = false, query_profiler_real_time_period_ns = 0, query_profiler_cpu_time_period_ns = 0, max_bytes_before_external_group_by = 90194313216, max_bytes_before_external_sort = 90194313216, max_memory_usage = 180388626432, backup_restore_keeper_retry_max_backoff_ms = 60000, cancel_http_readonly_queries_on_client_close = true, max_table_size_to_drop = 1000000000000, max_partition_size_to_drop = 1000000000000, default_table_engine = 'ReplicatedMergeTree', mutations_sync = 0, optimize_trivial_insert_select = false, database_replicated_allow_only_replicated_engine = true, cloud_mode = true, cloud_mode_engine = 2, distributed_ddl_output_mode = 'none_only_active', distributed_ddl_entry_format_version = 6, async_insert_max_data_size = 10485760, async_insert_busy_timeout_max_ms = 1000, enable_filesystem_cache_on_write_operations = true, load_marks_asynchronously = true, allow_prefetched_read_pool_for_remote_filesystem = true, filesystem_prefetch_max_memory_usage = 18038862643, filesystem_prefetches_limit = 200, compatibility = '24.6', insert_keeper_max_retries = 20, allow_experimental_materialized_postgresql_table = false, date_time_input_format = 'best_effort' ```. [#70820](https://github.com/ClickHouse/ClickHouse/pull/70820) ([Michael Kolupaev](https://github.com/al13n321)). +* Backported in [#70896](https://github.com/ClickHouse/ClickHouse/issues/70896): Disable enable_named_columns_in_function_tuple by default. [#70833](https://github.com/ClickHouse/ClickHouse/pull/70833) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#70994](https://github.com/ClickHouse/ClickHouse/issues/70994): Fix a logical error due to negative zeros in the two-level hash table. This closes [#70973](https://github.com/ClickHouse/ClickHouse/issues/70973). [#70979](https://github.com/ClickHouse/ClickHouse/pull/70979) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#71210](https://github.com/ClickHouse/ClickHouse/issues/71210): Fix logical error in `StorageS3Queue` "Cannot create a persistent node in /processed since it already exists". [#70984](https://github.com/ClickHouse/ClickHouse/pull/70984) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#71248](https://github.com/ClickHouse/ClickHouse/issues/71248): Fixed named sessions not being closed and hanging on forever under certain circumstances. [#70998](https://github.com/ClickHouse/ClickHouse/pull/70998) ([Márcio Martins](https://github.com/marcio-absmartly)). +* Backported in [#71375](https://github.com/ClickHouse/ClickHouse/issues/71375): Add try/catch to data parts destructors to avoid terminate. [#71364](https://github.com/ClickHouse/ClickHouse/pull/71364) ([alesapin](https://github.com/alesapin)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Backported in [#71026](https://github.com/ClickHouse/ClickHouse/issues/71026): Fix dropping of file cache in CHECK query in case of enabled transactions. [#69256](https://github.com/ClickHouse/ClickHouse/pull/69256) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#70388](https://github.com/ClickHouse/ClickHouse/issues/70388): CI: Enable Integration Tests for backport PRs. [#70329](https://github.com/ClickHouse/ClickHouse/pull/70329) ([Max Kainov](https://github.com/maxknv)). +* Backported in [#70701](https://github.com/ClickHouse/ClickHouse/issues/70701): Fix order in 03249_dynamic_alter_consistency. [#70453](https://github.com/ClickHouse/ClickHouse/pull/70453) ([Alexander Gololobov](https://github.com/davenger)). +* Backported in [#70542](https://github.com/ClickHouse/ClickHouse/issues/70542): Remove slow poll() logs in keeper. [#70508](https://github.com/ClickHouse/ClickHouse/pull/70508) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#70804](https://github.com/ClickHouse/ClickHouse/issues/70804): When the `PR Check` status is set, it's a valid RunConfig job failure. [#70643](https://github.com/ClickHouse/ClickHouse/pull/70643) ([Mikhail f. Shiryaev](https://github.com/Felixoid)). +* Backported in [#71229](https://github.com/ClickHouse/ClickHouse/issues/71229): Maybe not GWPAsan by default. [#71174](https://github.com/ClickHouse/ClickHouse/pull/71174) ([Antonio Andelic](https://github.com/antonio2368)). + diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index 10c55aa4bf5..cf28db5d49a 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -1,5 +1,7 @@ +v24.10.1.2812-stable 2024-11-01 v24.9.2.42-stable 2024-10-03 v24.9.1.3278-stable 2024-09-26 +v24.8.6.70-lts 2024-11-04 v24.8.5.115-lts 2024-10-08 v24.8.4.13-lts 2024-09-06 v24.8.3.59-lts 2024-09-03 From de751c7e4d3e6445348cd6e5d92a09dc7c41e0ab Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Mon, 4 Nov 2024 18:25:27 +0100 Subject: [PATCH 416/680] Update AccessControl.cpp --- src/Access/AccessControl.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Access/AccessControl.cpp b/src/Access/AccessControl.cpp index 647fb238d48..9b3b8d2a977 100644 --- a/src/Access/AccessControl.cpp +++ b/src/Access/AccessControl.cpp @@ -608,7 +608,7 @@ AuthResult AccessControl::authenticate(const Credentials & credentials, const Po } catch (...) { - tryLogCurrentException(getLogger(), "from: " + address.toString() + ", user: " + credentials.getUserName() + ": Authentication failed", LogsLevel::debug); + tryLogCurrentException(getLogger(), "from: " + address.toString() + ", user: " + credentials.getUserName() + ": Authentication failed", LogsLevel::information); WriteBufferFromOwnString message; message << credentials.getUserName() << ": Authentication failed: password is incorrect, or there is no user with such name."; From a612e9248c44bd41db761eb88e152a7d2ce6218c Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Mon, 4 Nov 2024 18:26:02 +0100 Subject: [PATCH 417/680] Update TCPHandler.cpp --- src/Server/TCPHandler.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Server/TCPHandler.cpp b/src/Server/TCPHandler.cpp index ea5507c3155..4f54918445f 100644 --- a/src/Server/TCPHandler.cpp +++ b/src/Server/TCPHandler.cpp @@ -1614,7 +1614,7 @@ void TCPHandler::receiveHello() if (e.code() != DB::ErrorCodes::AUTHENTICATION_FAILED) throw; - tryLogCurrentException(log, "SSL authentication failed, falling back to password authentication", LogsLevel::debug); + tryLogCurrentException(log, "SSL authentication failed, falling back to password authentication", LogsLevel::information); /// ^^ Log at debug level instead of default error level as authentication failures are not an unusual event. } } From 876158672c07361f54574c2eefabff5de9e0a48f Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Mon, 4 Nov 2024 17:53:48 +0000 Subject: [PATCH 418/680] Fix integration test: Sync all drop table calls --- tests/integration/test_quorum_inserts/test.py | 24 +++++++++---------- 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index f64864185c5..350da822c80 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -46,7 +46,7 @@ def started_cluster(): def test_simple_add_replica(started_cluster): - zero.query("DROP TABLE IF EXISTS test_simple ON CLUSTER cluster") + zero.query("DROP TABLE IF EXISTS test_simple ON CLUSTER cluster SYNC") create_query = ( "CREATE TABLE test_simple " @@ -82,12 +82,12 @@ def test_simple_add_replica(started_cluster): assert "1\t2011-01-01\n" == first.query("SELECT * from test_simple") assert "1\t2011-01-01\n" == second.query("SELECT * from test_simple") - zero.query("DROP TABLE IF EXISTS test_simple ON CLUSTER cluster") + zero.query("DROP TABLE IF EXISTS test_simple ON CLUSTER cluster SYNC") def test_drop_replica_and_achieve_quorum(started_cluster): zero.query( - "DROP TABLE IF EXISTS test_drop_replica_and_achieve_quorum ON CLUSTER cluster" + "DROP TABLE IF EXISTS test_drop_replica_and_achieve_quorum ON CLUSTER cluster SYNC" ) create_query = ( @@ -156,7 +156,7 @@ def test_insert_quorum_with_drop_partition(started_cluster, add_new_data): if add_new_data else "test_quorum_insert_with_drop_partition" ) - zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") + zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster SYNC") create_query = ( f"CREATE TABLE {table_name} ON CLUSTER cluster " @@ -208,7 +208,7 @@ def test_insert_quorum_with_drop_partition(started_cluster, add_new_data): assert TSV("") == TSV(zero.query(f"SELECT * FROM {table_name}")) assert TSV("") == TSV(second.query(f"SELECT * FROM {table_name}")) - zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") + zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster SYNC") @pytest.mark.parametrize(("add_new_data"), [False, True]) @@ -224,8 +224,8 @@ def test_insert_quorum_with_move_partition(started_cluster, add_new_data): if add_new_data else "test_insert_quorum_with_move_partition_destination" ) - zero.query(f"DROP TABLE IF EXISTS {source_table_name} ON CLUSTER cluster") - zero.query(f"DROP TABLE IF EXISTS {destination_table_name} ON CLUSTER cluster") + zero.query(f"DROP TABLE IF EXISTS {source_table_name} ON CLUSTER cluster SYNC") + zero.query(f"DROP TABLE IF EXISTS {destination_table_name} ON CLUSTER cluster SYNC") create_source = ( f"CREATE TABLE {source_table_name} ON CLUSTER cluster " @@ -291,12 +291,12 @@ def test_insert_quorum_with_move_partition(started_cluster, add_new_data): assert TSV("") == TSV(zero.query(f"SELECT * FROM {source_table_name}")) assert TSV("") == TSV(second.query(f"SELECT * FROM {source_table_name}")) - zero.query(f"DROP TABLE IF EXISTS {source_table_name} ON CLUSTER cluster") - zero.query(f"DROP TABLE IF EXISTS {destination_table_name} ON CLUSTER cluster") + zero.query(f"DROP TABLE IF EXISTS {source_table_name} ON CLUSTER cluster SYNC") + zero.query(f"DROP TABLE IF EXISTS {destination_table_name} ON CLUSTER cluster SYNC") def test_insert_quorum_with_ttl(started_cluster): - zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster") + zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster SYNC") create_query = ( "CREATE TABLE test_insert_quorum_with_ttl " @@ -363,12 +363,12 @@ def test_insert_quorum_with_ttl(started_cluster): ) ) - zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster") + zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster SYNC") def test_insert_quorum_with_keeper_loss_connection(started_cluster): zero.query( - "DROP TABLE IF EXISTS test_insert_quorum_with_keeper_loss ON CLUSTER cluster" + "DROP TABLE IF EXISTS test_insert_quorum_with_keeper_loss ON CLUSTER cluster SYNC" ) create_query = ( "CREATE TABLE test_insert_quorum_with_keeper_loss" From 64fbc9eb8d328db7013525fd6bb34fe0939b7c68 Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Mon, 4 Nov 2024 18:06:08 +0000 Subject: [PATCH 419/680] Style --- tests/integration/test_quorum_inserts/test.py | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index 350da822c80..0809d2c003f 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -296,7 +296,9 @@ def test_insert_quorum_with_move_partition(started_cluster, add_new_data): def test_insert_quorum_with_ttl(started_cluster): - zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster SYNC") + zero.query( + "DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster SYNC" + ) create_query = ( "CREATE TABLE test_insert_quorum_with_ttl " @@ -363,7 +365,9 @@ def test_insert_quorum_with_ttl(started_cluster): ) ) - zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster SYNC") + zero.query( + "DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster SYNC" + ) def test_insert_quorum_with_keeper_loss_connection(started_cluster): From c1ce74f52f9b5b53db7bcf43aa0a1a47c9dd9859 Mon Sep 17 00:00:00 2001 From: MikhailBurdukov <102754618+MikhailBurdukov@users.noreply.github.com> Date: Mon, 4 Nov 2024 21:40:59 +0300 Subject: [PATCH 420/680] Update tests/integration/test_named_collections/test.py Co-authored-by: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> --- tests/integration/test_named_collections/test.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/integration/test_named_collections/test.py b/tests/integration/test_named_collections/test.py index bd04bb9e3c8..e2fa776a8f0 100644 --- a/tests/integration/test_named_collections/test.py +++ b/tests/integration/test_named_collections/test.py @@ -803,7 +803,7 @@ def test_keeper_storage_remove_on_cluster(cluster, ignore, expected_raise): def test_name_escaping(cluster, instance_name): node = cluster.instances[instance_name] - node.query("DROP NAMED COLLECTION IF EXISTS test;") + node.query("DROP NAMED COLLECTION IF EXISTS `test_!strange/symbols!`;") node.query("CREATE NAMED COLLECTION `test_!strange/symbols!` AS key1=1, key2=2") node.restart_clickhouse() From 157e1695d5f8d8dd0962f89a782317a5249ad8eb Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Mon, 4 Nov 2024 20:02:57 +0100 Subject: [PATCH 421/680] Fix ExecuteScalarSubqueriesMatcher visiting join elements --- src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp b/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp index d4da038c089..c80852e9ae7 100644 --- a/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp +++ b/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp @@ -63,7 +63,7 @@ bool ExecuteScalarSubqueriesMatcher::needChildVisit(ASTPtr & node, const ASTPtr if (node->as()) { /// Do not go to FROM, JOIN, UNION. - if (child->as() || child->as()) + if (child->as() || child->as() || child->as()) return false; } From b4a3f6d3709b87f5b1a30316b60f042fc1c0f2ad Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Mon, 4 Nov 2024 20:11:33 +0100 Subject: [PATCH 422/680] Make sure to update table_join children properly --- src/Analyzer/JoinNode.cpp | 10 ++++++++-- src/Interpreters/QueryNormalizer.cpp | 6 ++++++ .../TimeSeries/PrometheusRemoteReadProtocol.cpp | 1 + 3 files changed, 15 insertions(+), 2 deletions(-) diff --git a/src/Analyzer/JoinNode.cpp b/src/Analyzer/JoinNode.cpp index bf99c014826..722c1e19b7e 100644 --- a/src/Analyzer/JoinNode.cpp +++ b/src/Analyzer/JoinNode.cpp @@ -48,9 +48,15 @@ ASTPtr JoinNode::toASTTableJoin() const auto join_expression_ast = children[join_expression_child_index]->toAST(); if (is_using_join_expression) - join_ast->using_expression_list = std::move(join_expression_ast); + { + join_ast->using_expression_list = join_expression_ast; + join_ast->children.push_back(join_ast->using_expression_list); + } else - join_ast->on_expression = std::move(join_expression_ast); + { + join_ast->on_expression = join_expression_ast; + join_ast->children.push_back(join_ast->on_expression); + } } return join_ast; diff --git a/src/Interpreters/QueryNormalizer.cpp b/src/Interpreters/QueryNormalizer.cpp index a8639906aad..bba30fb5194 100644 --- a/src/Interpreters/QueryNormalizer.cpp +++ b/src/Interpreters/QueryNormalizer.cpp @@ -161,7 +161,13 @@ void QueryNormalizer::visit(ASTTablesInSelectQueryElement & node, const ASTPtr & { auto & join = node.table_join->as(); if (join.on_expression) + { + ASTPtr original_on_expression = join.on_expression; visit(join.on_expression, data); + if (join.on_expression != original_on_expression) + join.children = { join.on_expression }; + } + } } diff --git a/src/Storages/TimeSeries/PrometheusRemoteReadProtocol.cpp b/src/Storages/TimeSeries/PrometheusRemoteReadProtocol.cpp index df0f6b8bc5c..b8a3b2911b9 100644 --- a/src/Storages/TimeSeries/PrometheusRemoteReadProtocol.cpp +++ b/src/Storages/TimeSeries/PrometheusRemoteReadProtocol.cpp @@ -245,6 +245,7 @@ namespace table_join->strictness = JoinStrictness::Semi; table_join->on_expression = makeASTFunction("equals", makeASTColumn(data_table_id, TimeSeriesColumnNames::ID), makeASTColumn(tags_table_id, TimeSeriesColumnNames::ID)); + table_join->children.push_back(table_join->on_expression); table->table_join = table_join; auto table_exp = std::make_shared(); From 35a0d08a32302247b3689e887e9a3b72bb9152e3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Mon, 4 Nov 2024 20:12:34 +0100 Subject: [PATCH 423/680] RewriteArrayExistsFunctionVisitor: Assert proper child on join expression --- .../RewriteArrayExistsFunctionVisitor.cpp | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/src/Interpreters/RewriteArrayExistsFunctionVisitor.cpp b/src/Interpreters/RewriteArrayExistsFunctionVisitor.cpp index 22ce91d8c67..60bac2fb7a3 100644 --- a/src/Interpreters/RewriteArrayExistsFunctionVisitor.cpp +++ b/src/Interpreters/RewriteArrayExistsFunctionVisitor.cpp @@ -20,21 +20,21 @@ void RewriteArrayExistsFunctionMatcher::visit(ASTPtr & ast, Data & data) if (join->using_expression_list) { auto * it = std::find(join->children.begin(), join->children.end(), join->using_expression_list); + if (it == join->children.end()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Could not find join->using_expression_list in '{}'", join->formatForLogging()); visit(join->using_expression_list, data); - - if (it && *it != join->using_expression_list) - *it = join->using_expression_list; + *it = join->using_expression_list; } if (join->on_expression) { auto * it = std::find(join->children.begin(), join->children.end(), join->on_expression); + if (it == join->children.end()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Could not find join->on_expression in '{}'", join->formatForLogging()); visit(join->on_expression, data); - - if (it && *it != join->on_expression) - *it = join->on_expression; + *it = join->on_expression; } } } From 389fdd80d36b5073698b87b7a7d24dcc4c6560bf Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Mon, 4 Nov 2024 20:15:29 +0100 Subject: [PATCH 424/680] Add test for crasher --- ...ptimize_rewrite_array_exists_to_has_crash.reference | 0 ...3261_optimize_rewrite_array_exists_to_has_crash.sql | 10 ++++++++++ 2 files changed, 10 insertions(+) create mode 100644 tests/queries/0_stateless/03261_optimize_rewrite_array_exists_to_has_crash.reference create mode 100644 tests/queries/0_stateless/03261_optimize_rewrite_array_exists_to_has_crash.sql diff --git a/tests/queries/0_stateless/03261_optimize_rewrite_array_exists_to_has_crash.reference b/tests/queries/0_stateless/03261_optimize_rewrite_array_exists_to_has_crash.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03261_optimize_rewrite_array_exists_to_has_crash.sql b/tests/queries/0_stateless/03261_optimize_rewrite_array_exists_to_has_crash.sql new file mode 100644 index 00000000000..5a54d86f339 --- /dev/null +++ b/tests/queries/0_stateless/03261_optimize_rewrite_array_exists_to_has_crash.sql @@ -0,0 +1,10 @@ +-- https://github.com/ClickHouse/ClickHouse/issues/71382 +DROP TABLE IF EXISTS rewrite; +CREATE TABLE rewrite (c0 Int) ENGINE = Memory(); +SELECT 1 +FROM rewrite +INNER JOIN rewrite AS y ON ( + SELECT 1 +) +INNER JOIN rewrite AS z ON 1 +SETTINGS allow_experimental_analyzer=0, optimize_rewrite_array_exists_to_has=1; From 19422e75b0fbe7fbbe68bef98f10f22ee046db4b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Mon, 4 Nov 2024 20:24:06 +0100 Subject: [PATCH 425/680] Style --- src/Interpreters/RewriteArrayExistsFunctionVisitor.cpp | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/src/Interpreters/RewriteArrayExistsFunctionVisitor.cpp b/src/Interpreters/RewriteArrayExistsFunctionVisitor.cpp index 60bac2fb7a3..2890357494d 100644 --- a/src/Interpreters/RewriteArrayExistsFunctionVisitor.cpp +++ b/src/Interpreters/RewriteArrayExistsFunctionVisitor.cpp @@ -6,6 +6,12 @@ namespace DB { + +namespace ErrorCode +{ +extern const int LOGICAL_ERROR; +} + void RewriteArrayExistsFunctionMatcher::visit(ASTPtr & ast, Data & data) { if (auto * func = ast->as()) From f9f1870a0e91f029849fa7897c74b9d3355f7f6a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Mon, 4 Nov 2024 21:10:44 +0100 Subject: [PATCH 426/680] Fix upgrade check (24.11) --- src/Core/SettingsChangesHistory.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 157054e5627..b95dc5f85ed 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -71,6 +71,7 @@ static std::initializer_list Date: Tue, 5 Nov 2024 01:58:23 +0000 Subject: [PATCH 427/680] attempt to fix irrelevant test --- tests/integration/test_quorum_inserts/test.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index eefc4882e8e..66f96d61b3e 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -366,7 +366,7 @@ def test_insert_quorum_with_ttl(started_cluster): zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster") -def test_insert_quorum_with_keeper_loss_connection(): +def test_insert_quorum_with_keeper_loss_connection(started_cluster): zero.query( "DROP TABLE IF EXISTS test_insert_quorum_with_keeper_fail ON CLUSTER cluster" ) From a35cc85a68c9356f5697fa22e057bf74a28ee5bb Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy Date: Tue, 5 Nov 2024 04:07:09 +0000 Subject: [PATCH 428/680] remove irrelevant changes --- tests/integration/test_quorum_inserts/test.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index 66f96d61b3e..eefc4882e8e 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -366,7 +366,7 @@ def test_insert_quorum_with_ttl(started_cluster): zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster") -def test_insert_quorum_with_keeper_loss_connection(started_cluster): +def test_insert_quorum_with_keeper_loss_connection(): zero.query( "DROP TABLE IF EXISTS test_insert_quorum_with_keeper_fail ON CLUSTER cluster" ) From 3491c0c0e83c5f76c5d5de5097ce513436b4d010 Mon Sep 17 00:00:00 2001 From: Max Kainov Date: Fri, 1 Nov 2024 09:42:02 +0100 Subject: [PATCH 429/680] CI: Remove deprecated release script --- tests/ci/mark_release_ready.py | 3 +- tests/ci/release.py | 693 --------------------------------- 2 files changed, 2 insertions(+), 694 deletions(-) delete mode 100755 tests/ci/release.py diff --git a/tests/ci/mark_release_ready.py b/tests/ci/mark_release_ready.py index 7ffb3c9a89b..838961bd89f 100755 --- a/tests/ci/mark_release_ready.py +++ b/tests/ci/mark_release_ready.py @@ -9,9 +9,10 @@ from get_robot_token import get_best_robot_token from git_helper import commit as commit_arg from github_helper import GitHub from pr_info import PRInfo -from release import RELEASE_READY_STATUS from report import SUCCESS +RELEASE_READY_STATUS = "Ready for release" + def main(): parser = argparse.ArgumentParser( diff --git a/tests/ci/release.py b/tests/ci/release.py deleted file mode 100755 index ed9d60a5cad..00000000000 --- a/tests/ci/release.py +++ /dev/null @@ -1,693 +0,0 @@ -#!/usr/bin/env python3 - -""" -script to create releases for ClickHouse - -The `gh` CLI preferred over the PyGithub to have an easy way to rollback bad -release in command line by simple execution giving rollback commands - -On another hand, PyGithub is used for convenient getting commit's status from API - -To run this script on a freshly installed Ubuntu 22.04 system, it is enough to do the following commands: - -sudo apt install pip -pip install requests boto3 github PyGithub -sudo snap install gh -gh auth login -""" - - -import argparse -import json -import logging -import subprocess -from contextlib import contextmanager -from typing import Any, Final, Iterator, List, Optional, Tuple - -from ci_config import Labels -from git_helper import Git, commit, release_branch -from report import SUCCESS -from version_helper import ( - FILE_WITH_VERSION_PATH, - GENERATED_CONTRIBUTORS, - ClickHouseVersion, - VersionType, - get_abs_path, - get_version_from_repo, - update_cmake_version, - update_contributors, -) - -RELEASE_READY_STATUS = "Ready for release" - - -class Repo: - VALID = ("ssh", "https", "origin") - - def __init__(self, repo: str, protocol: str): - self._repo = repo - self._url = "" - self.url = protocol - - @property - def url(self) -> str: - return self._url - - @url.setter - def url(self, protocol: str) -> None: - if protocol == "ssh": - self._url = f"git@github.com:{self}.git" - elif protocol == "https": - self._url = f"https://github.com/{self}.git" - elif protocol == "origin": - self._url = protocol - else: - raise ValueError(f"protocol must be in {self.VALID}") - - def __str__(self): - return self._repo - - -class Release: - NEW = "new" # type: Final - PATCH = "patch" # type: Final - VALID_TYPE = (NEW, PATCH) # type: Final[Tuple[str, str]] - CMAKE_PATH = get_abs_path(FILE_WITH_VERSION_PATH) - CONTRIBUTORS_PATH = get_abs_path(GENERATED_CONTRIBUTORS) - - def __init__( - self, - repo: Repo, - release_commit: str, - release_type: str, - dry_run: bool, - with_stderr: bool, - ): - self.repo = repo - self._release_commit = "" - self.release_commit = release_commit - self.dry_run = dry_run - self.with_stderr = with_stderr - assert release_type in self.VALID_TYPE - self.release_type = release_type - self._git = Git() - self._version = get_version_from_repo(git=self._git) - self.release_version = self.version - self._release_branch = "" - self._version_new_tag = None # type: Optional[ClickHouseVersion] - self._rollback_stack = [] # type: List[str] - - def run( - self, cmd: str, cwd: Optional[str] = None, dry_run: bool = False, **kwargs: Any - ) -> str: - cwd_text = "" - if cwd: - cwd_text = f" (CWD='{cwd}')" - if dry_run: - logging.info("Would run command%s:\n %s", cwd_text, cmd) - return "" - if not self.with_stderr: - kwargs["stderr"] = subprocess.DEVNULL - - logging.info("Running command%s:\n %s", cwd_text, cmd) - return self._git.run(cmd, cwd, **kwargs) - - def set_release_info(self): - # Fetch release commit and tags in case they don't exist locally - self.run( - f"git fetch {self.repo.url} {self.release_commit} --no-recurse-submodules" - ) - self.run(f"git fetch {self.repo.url} --tags --no-recurse-submodules") - - # Get the actual version for the commit before check - with self._checkout(self.release_commit, True): - self.release_branch = f"{self.version.major}.{self.version.minor}" - self.release_version = get_version_from_repo(git=self._git) - self.release_version.with_description(self.get_stable_release_type()) - - self.read_version() - - def read_version(self): - self._git.update() - self.version = get_version_from_repo(git=self._git) - - def get_stable_release_type(self) -> str: - if self.version.is_lts: - return VersionType.LTS - return VersionType.STABLE - - def check_commit_release_ready(self): - per_page = 100 - page = 1 - while True: - statuses = json.loads( - self.run( - f"gh api 'repos/{self.repo}/commits/{self.release_commit}" - f"/statuses?per_page={per_page}&page={page}'" - ) - ) - - if not statuses: - break - - for status in statuses: - if status["context"] == RELEASE_READY_STATUS: - if not status["state"] == SUCCESS: - raise ValueError( - f"the status {RELEASE_READY_STATUS} is {status['state']}" - ", not success" - ) - - return - - page += 1 - - raise KeyError( - f"the status {RELEASE_READY_STATUS} " - f"is not found for commit {self.release_commit}" - ) - - def check_prerequisites(self): - """ - Check tooling installed in the system, `git` is checked by Git() init - """ - try: - self.run("gh auth status") - except subprocess.SubprocessError: - logging.error( - "The github-cli either not installed or not setup, please follow " - "the instructions on https://github.com/cli/cli#installation and " - "https://cli.github.com/manual/" - ) - raise - - if self.release_type == self.PATCH: - self.check_commit_release_ready() - - def do( - self, check_dirty: bool, check_run_from_master: bool, check_branch: bool - ) -> None: - self.check_prerequisites() - - if check_dirty: - logging.info("Checking if repo is clean") - try: - self.run("git diff HEAD --exit-code") - except subprocess.CalledProcessError: - logging.fatal("Repo contains uncommitted changes") - raise - - if check_run_from_master and self._git.branch != "master": - raise RuntimeError("the script must be launched only from master") - - self.set_release_info() - - if check_branch: - self.check_branch() - - if self.release_type == self.NEW: - with self._checkout(self.release_commit, True): - # Checkout to the commit, it will provide the correct current version - with self.new_release(): - with self.create_release_branch(): - logging.info( - "Publishing release %s from commit %s is done", - self.release_version.describe, - self.release_commit, - ) - - elif self.release_type == self.PATCH: - with self._checkout(self.release_commit, True): - with self.patch_release(): - logging.info( - "Publishing release %s from commit %s is done", - self.release_version.describe, - self.release_commit, - ) - - if self.dry_run: - logging.info("Dry running, clean out possible changes") - rollback = self._rollback_stack.copy() - rollback.reverse() - for cmd in rollback: - self.run(cmd) - return - - self.log_post_workflows() - self.log_rollback() - - def check_no_tags_after(self): - tags_after_commit = self.run(f"git tag --contains={self.release_commit}") - if tags_after_commit: - raise RuntimeError( - f"Commit {self.release_commit} belongs to following tags:\n" - f"{tags_after_commit}\nChoose another commit" - ) - - def check_branch(self): - branch = self.release_branch - if self.release_type == self.NEW: - # Commit to spin up the release must belong to a main branch - branch = "master" - elif self.release_type != self.PATCH: - raise ( - ValueError(f"release_type {self.release_type} not in {self.VALID_TYPE}") - ) - - # Prefetch the branch to have it updated - if self._git.branch == branch: - self.run("git pull --no-recurse-submodules") - else: - self.run( - f"git fetch {self.repo.url} {branch}:{branch} --no-recurse-submodules" - ) - output = self.run(f"git branch --contains={self.release_commit} {branch}") - if branch not in output: - raise RuntimeError( - f"commit {self.release_commit} must belong to {branch} " - f"for {self.release_type} release" - ) - - def _update_cmake_contributors( - self, version: ClickHouseVersion, reset_tweak: bool = True - ) -> None: - if reset_tweak: - desc = version.description - version = version.reset_tweak() - version.with_description(desc) - update_cmake_version(version) - update_contributors(raise_error=True) - if self.dry_run: - logging.info( - "Dry running, resetting the following changes in the repo:\n%s", - self.run(f"git diff '{self.CMAKE_PATH}' '{self.CONTRIBUTORS_PATH}'"), - ) - self.run(f"git checkout '{self.CMAKE_PATH}' '{self.CONTRIBUTORS_PATH}'") - - def _commit_cmake_contributors( - self, version: ClickHouseVersion, reset_tweak: bool = True - ) -> None: - if reset_tweak: - version = version.reset_tweak() - self.run( - f"git commit '{self.CMAKE_PATH}' '{self.CONTRIBUTORS_PATH}' " - f"-m 'Update autogenerated version to {version.string} and contributors'", - dry_run=self.dry_run, - ) - - @property - def bump_part(self) -> ClickHouseVersion.PART_TYPE: - if self.release_type == Release.NEW: - if self._version.minor >= 12: - return "major" - return "minor" - return "patch" - - @property - def has_rollback(self) -> bool: - return bool(self._rollback_stack) - - def log_rollback(self): - if self.has_rollback: - rollback = self._rollback_stack.copy() - rollback.reverse() - logging.info( - "To rollback the action run the following commands:\n %s", - "\n ".join(rollback), - ) - - def log_post_workflows(self): - logging.info( - "To verify all actions are running good visit the following links:\n %s", - "\n ".join( - f"https://github.com/{self.repo}/actions/workflows/{action}.yml" - for action in ("release", "tags_stable") - ), - ) - - @contextmanager - def create_release_branch(self): - self.check_no_tags_after() - # Create release branch - self.read_version() - assert self._version_new_tag is not None - with self._create_tag( - self._version_new_tag.describe, - self.release_commit, - f"Initial commit for release {self._version_new_tag.major}.{self._version_new_tag.minor}", - ): - with self._create_branch(self.release_branch, self.release_commit): - with self._checkout(self.release_branch, True): - with self._bump_release_branch(): - yield - - @contextmanager - def patch_release(self): - self.check_no_tags_after() - self.read_version() - version_type = self.get_stable_release_type() - self.version.with_description(version_type) - with self._create_gh_release(False): - self.version = self.version.update(self.bump_part) - self.version.with_description(version_type) - self._update_cmake_contributors(self.version) - # Checking out the commit of the branch and not the branch itself, - # then we are able to skip rollback - with self._checkout(f"{self.release_branch}^0", False): - current_commit = self.run("git rev-parse HEAD") - self._commit_cmake_contributors(self.version) - with self._push( - "HEAD", with_rollback_on_fail=False, remote_ref=self.release_branch - ): - # DO NOT PUT ANYTHING ELSE HERE - # The push must be the last action and mean the successful release - self._rollback_stack.append( - f"{self.dry_run_prefix}git push {self.repo.url} " - f"+{current_commit}:{self.release_branch}" - ) - yield - - @contextmanager - def new_release(self): - # Create branch for a version bump - self.read_version() - self.version = self.version.update(self.bump_part) - helper_branch = f"{self.version.major}.{self.version.minor}-prepare" - with self._create_branch(helper_branch, self.release_commit): - with self._checkout(helper_branch, True): - with self._bump_version_in_master(helper_branch): - yield - - @property - def version(self) -> ClickHouseVersion: - return self._version - - @version.setter - def version(self, version: ClickHouseVersion) -> None: - if not isinstance(version, ClickHouseVersion): - raise ValueError(f"version must be ClickHouseVersion, not {type(version)}") - self._version = version - - @property - def release_branch(self) -> str: - return self._release_branch - - @release_branch.setter - def release_branch(self, branch: str) -> None: - self._release_branch = release_branch(branch) - - @property - def release_commit(self) -> str: - return self._release_commit - - @release_commit.setter - def release_commit(self, release_commit: str) -> None: - self._release_commit = commit(release_commit) - - @property - def dry_run_prefix(self) -> str: - if self.dry_run: - return "# " - return "" - - @contextmanager - def _bump_release_branch(self): - # Update only git, original version stays the same - self._git.update() - new_version = self.version.copy() - version_type = self.get_stable_release_type() - pr_labels = f"--label {Labels.RELEASE}" - if version_type == VersionType.LTS: - pr_labels += f" --label {Labels.RELEASE_LTS}" - new_version.with_description(version_type) - self._update_cmake_contributors(new_version) - self._commit_cmake_contributors(new_version) - with self._push(self.release_branch): - with self._create_gh_label( - f"v{self.release_branch}-must-backport", "10dbed" - ): - with self._create_gh_label( - f"v{self.release_branch}-affected", "c2bfff" - ): - # The following command is rolled back by deleting branch - # in self._push - self.run( - f"gh pr create --repo {self.repo} --title " - f"'Release pull request for branch {self.release_branch}' " - f"--head {self.release_branch} {pr_labels} " - "--body 'This PullRequest is a part of ClickHouse release " - "cycle. It is used by CI system only. Do not perform any " - "changes with it.'", - dry_run=self.dry_run, - ) - # Here the release branch part is done. - # We don't create a release itself automatically to have a - # safe window to backport possible bug fixes. - yield - - @contextmanager - def _bump_version_in_master(self, helper_branch: str) -> Iterator[None]: - self.read_version() - self.version = self.version.update(self.bump_part) - self.version.with_description(VersionType.TESTING) - self._update_cmake_contributors(self.version) - self._commit_cmake_contributors(self.version) - # Create a version-new tag - self._version_new_tag = self.version.copy() - self._version_new_tag.tweak = 1 - self._version_new_tag.with_description(VersionType.NEW) - - with self._push(helper_branch): - body_file = get_abs_path(".github/PULL_REQUEST_TEMPLATE.md") - # The following command is rolled back by deleting branch in self._push - self.run( - f"gh pr create --repo {self.repo} --title 'Update version after " - f"release' --head {helper_branch} --body-file '{body_file}' " - "--label 'do not test' --assignee @me", - dry_run=self.dry_run, - ) - # Here the new release part is done - yield - - @contextmanager - def _checkout(self, ref: str, with_checkout_back: bool = False) -> Iterator[None]: - self._git.update() - orig_ref = self._git.branch or self._git.sha - rollback_cmd = "" - if ref not in (self._git.branch, self._git.sha): - self.run(f"git checkout {ref}") - # checkout is not put into rollback_stack intentionally - rollback_cmd = f"git checkout {orig_ref}" - # always update version and git after checked out ref - self.read_version() - try: - yield - except (Exception, KeyboardInterrupt): - logging.warning("Rolling back checked out %s for %s", ref, orig_ref) - self.run(f"git reset --hard; git checkout -f {orig_ref}") - raise - # Normal flow when we need to checkout back - if with_checkout_back and rollback_cmd: - self.run(rollback_cmd) - - @contextmanager - def _create_branch(self, name: str, start_point: str = "") -> Iterator[None]: - self.run(f"git branch {name} {start_point}") - - rollback_cmd = f"git branch -D {name}" - self._rollback_stack.append(rollback_cmd) - try: - yield - except (Exception, KeyboardInterrupt): - logging.warning("Rolling back created branch %s", name) - self.run(rollback_cmd) - raise - - @contextmanager - def _create_gh_label(self, label: str, color_hex: str) -> Iterator[None]: - # API call, https://docs.github.com/en/rest/reference/issues#create-a-label - self.run( - f"gh api repos/{self.repo}/labels -f name={label} -f color={color_hex}", - dry_run=self.dry_run, - ) - rollback_cmd = ( - f"{self.dry_run_prefix}gh api repos/{self.repo}/labels/{label} -X DELETE" - ) - self._rollback_stack.append(rollback_cmd) - try: - yield - except (Exception, KeyboardInterrupt): - logging.warning("Rolling back label %s", label) - self.run(rollback_cmd) - raise - - @contextmanager - def _create_gh_release(self, as_prerelease: bool) -> Iterator[None]: - tag = self.release_version.describe - with self._create_tag(tag, self.release_commit): - # Preserve tag if version is changed - prerelease = "" - if as_prerelease: - prerelease = "--prerelease" - self.run( - f"gh release create {prerelease} --repo {self.repo} " - f"--title 'Release {tag}' '{tag}'", - dry_run=self.dry_run, - ) - rollback_cmd = ( - f"{self.dry_run_prefix}gh release delete --yes " - f"--repo {self.repo} '{tag}'" - ) - self._rollback_stack.append(rollback_cmd) - try: - yield - except (Exception, KeyboardInterrupt): - logging.warning("Rolling back release publishing") - self.run(rollback_cmd) - raise - - @contextmanager - def _create_tag( - self, tag: str, commit: str, tag_message: str = "" - ) -> Iterator[None]: - tag_message = tag_message or f"Release {tag}" - # Create tag even in dry-run - self.run(f"git tag -a -m '{tag_message}' '{tag}' {commit}") - rollback_cmd = f"git tag -d '{tag}'" - self._rollback_stack.append(rollback_cmd) - try: - with self._push(tag): - yield - except (Exception, KeyboardInterrupt): - logging.warning("Rolling back tag %s", tag) - self.run(rollback_cmd) - raise - - @contextmanager - def _push( - self, ref: str, with_rollback_on_fail: bool = True, remote_ref: str = "" - ) -> Iterator[None]: - if remote_ref == "": - remote_ref = ref - - self.run(f"git push {self.repo.url} {ref}:{remote_ref}", dry_run=self.dry_run) - if with_rollback_on_fail: - rollback_cmd = ( - f"{self.dry_run_prefix}git push -d {self.repo.url} {remote_ref}" - ) - self._rollback_stack.append(rollback_cmd) - - try: - yield - except (Exception, KeyboardInterrupt): - if with_rollback_on_fail: - logging.warning("Rolling back pushed ref %s", ref) - self.run(rollback_cmd) - - raise - - -def parse_args() -> argparse.Namespace: - parser = argparse.ArgumentParser( - formatter_class=argparse.ArgumentDefaultsHelpFormatter, - description="Script to release a new ClickHouse version, requires `git` and " - "`gh` (github-cli) commands " - "!!! LAUNCH IT ONLY FROM THE MASTER BRANCH !!!", - ) - - parser.add_argument( - "--commit", - required=True, - type=commit, - help="commit create a release", - ) - parser.add_argument( - "--repo", - default="ClickHouse/ClickHouse", - help="repository to create the release", - ) - parser.add_argument( - "--remote-protocol", - "-p", - default="ssh", - choices=Repo.VALID, - help="repo protocol for git commands remote, 'origin' is a special case and " - "uses 'origin' as a remote", - ) - parser.add_argument( - "--type", - required=True, - choices=Release.VALID_TYPE, - dest="release_type", - help="a release type to bump the major.minor.patch version part, " - "new branch is created only for the value 'new'", - ) - parser.add_argument("--with-release-branch", default=True, help=argparse.SUPPRESS) - parser.add_argument("--check-dirty", default=True, help=argparse.SUPPRESS) - parser.add_argument( - "--no-check-dirty", - dest="check_dirty", - action="store_false", - default=argparse.SUPPRESS, - help="(dangerous) if set, skip check repository for uncommitted changes", - ) - parser.add_argument("--check-run-from-master", default=True, help=argparse.SUPPRESS) - parser.add_argument( - "--no-run-from-master", - dest="check_run_from_master", - action="store_false", - default=argparse.SUPPRESS, - help="(for development) if set, the script could run from non-master branch", - ) - parser.add_argument("--check-branch", default=True, help=argparse.SUPPRESS) - parser.add_argument( - "--no-check-branch", - dest="check_branch", - action="store_false", - default=argparse.SUPPRESS, - help="(debug or development only, dangerous) if set, skip the branch check for " - "a run. By default, 'new' type work only for master, and 'patch' " - "works only for a release branches, that name " - "should be the same as '$MAJOR.$MINOR' version, e.g. 22.2", - ) - parser.add_argument( - "--dry-run", - action="store_true", - help="do not make any actual changes in the repo, just show what will be done", - ) - parser.add_argument( - "--with-stderr", - action="store_true", - help="if set, the stderr of all subprocess commands will be printed as well", - ) - - return parser.parse_args() - - -def main(): - logging.basicConfig(level=logging.INFO) - args = parse_args() - repo = Repo(args.repo, args.remote_protocol) - release = Release( - repo, args.commit, args.release_type, args.dry_run, args.with_stderr - ) - - try: - release.do(args.check_dirty, args.check_run_from_master, args.check_branch) - except: - if release.has_rollback: - logging.error( - "!!The release process finished with error, read the output carefully!!" - ) - logging.error( - "Probably, rollback finished with error. " - "If you don't see any of the following commands in the output, " - "execute them manually:" - ) - release.log_rollback() - raise - - -if __name__ == "__main__": - assert False, "Script Deprecated, ask ci team for help" - main() From 1abfa41b890d4cdcb09d06579b8e9b7f14d4f4f5 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Tue, 5 Nov 2024 11:18:11 +0100 Subject: [PATCH 430/680] Update CMakeLists.txt --- contrib/usearch-cmake/CMakeLists.txt | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/contrib/usearch-cmake/CMakeLists.txt b/contrib/usearch-cmake/CMakeLists.txt index 69a986de192..fda061bf467 100644 --- a/contrib/usearch-cmake/CMakeLists.txt +++ b/contrib/usearch-cmake/CMakeLists.txt @@ -19,7 +19,8 @@ endif () add_library(ch_contrib::usearch ALIAS _usearch) - +# Cf. https://github.com/llvm/llvm-project/issues/107810 (though it is not 100% the same stack) +# # LLVM ERROR: Cannot select: 0x7996e7a73150: f32,ch = load<(load (s16) from %ir.22, !tbaa !54231), anyext from bf16> 0x79961cb737c0, 0x7996e7a1a500, undef:i64, ./contrib/SimSIMD/include/simsimd/dot.h:215:1 # 0x7996e7a1a500: i64 = add 0x79961e770d00, Constant:i64<-16>, ./contrib/SimSIMD/include/simsimd/dot.h:215:1 # 0x79961e770d00: i64,ch = CopyFromReg 0x79961cb737c0, Register:i64 %4, ./contrib/SimSIMD/include/simsimd/dot.h:215:1 From 087a886bc9f312c4cc4fc6cba1d1ea5d1681c137 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Tue, 5 Nov 2024 11:18:21 +0100 Subject: [PATCH 431/680] Update src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp Co-authored-by: Nikita Taranov --- src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp | 2 -- 1 file changed, 2 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp index 0b5ffa659dc..5a725922e14 100644 --- a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp @@ -118,8 +118,6 @@ USearchIndexWithSerialization::USearchIndexWithSerialization( if (!result) throw Exception(ErrorCodes::INCORRECT_DATA, "Could not create vector similarity index. Error: {}", String(result.error.release())); swap(result.index); - - /// LOG_TRACE(getLogger("XXX"), "{}", simsimd_uses_dynamic_dispatch()); } void USearchIndexWithSerialization::serialize(WriteBuffer & ostr) const From 0cc8626279fefc6ceae0a806b4e326ea0a354476 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Tue, 5 Nov 2024 11:31:27 +0000 Subject: [PATCH 432/680] Fix assert during insert into vector similarity index in presence of other skipping indexes --- .../MergeTreeIndexVectorSimilarity.cpp | 79 ++++++++++--------- .../02354_vector_search_bugs.reference | 1 + .../0_stateless/02354_vector_search_bugs.sql | 15 ++++ 3 files changed, 58 insertions(+), 37 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp index 5a725922e14..498d0131d5a 100644 --- a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp @@ -347,53 +347,58 @@ void MergeTreeIndexAggregatorVectorSimilarity::update(const Block & block, size_ if (index_sample_block.columns() > 1) throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected block with single column"); - const String & index_column_name = index_sample_block.getByPosition(0).name; - const ColumnPtr & index_column = block.getByName(index_column_name).column; - ColumnPtr column_cut = index_column->cut(*pos, rows_read); + for (size_t i = 0; i < index_sample_block.columns(); ++i) + { + const auto & index_column_with_type_and_name = index_sample_block.getByPosition(i); - const auto * column_array = typeid_cast(column_cut.get()); - if (!column_array) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected Array(Float*) column"); + const auto & index_column_name = index_column_with_type_and_name.name; + const auto & index_column = block.getByName(index_column_name).column; + ColumnPtr column_cut = index_column->cut(*pos, rows_read); - if (column_array->empty()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Array is unexpectedly empty"); + const auto * column_array = typeid_cast(column_cut.get()); + if (!column_array) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected Array(Float*) column"); - /// The vector similarity algorithm naturally assumes that the indexed vectors have dimension >= 1. This condition is violated if empty arrays - /// are INSERTed into an vector-similarity-indexed column or if no value was specified at all in which case the arrays take on their default - /// values which is also empty. - if (column_array->isDefaultAt(0)) - throw Exception(ErrorCodes::INCORRECT_DATA, "The arrays in column '{}' must not be empty. Did you try to INSERT default values?", index_column_name); + if (column_array->empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Array is unexpectedly empty"); - const size_t rows = column_array->size(); + /// The vector similarity algorithm naturally assumes that the indexed vectors have dimension >= 1. This condition is violated if empty arrays + /// are INSERTed into an vector-similarity-indexed column or if no value was specified at all in which case the arrays take on their default + /// values which is also empty. + if (column_array->isDefaultAt(0)) + throw Exception(ErrorCodes::INCORRECT_DATA, "The arrays in column '{}' must not be empty. Did you try to INSERT default values?", index_column_name); - const auto & column_array_offsets = column_array->getOffsets(); - const size_t dimensions = column_array_offsets[0]; + const size_t rows = column_array->size(); - if (!index) - index = std::make_shared(dimensions, metric_kind, scalar_kind, usearch_hnsw_params); + const auto & column_array_offsets = column_array->getOffsets(); + const size_t dimensions = column_array_offsets[0]; - /// Also check that previously inserted blocks have the same size as this block. - /// Note that this guarantees consistency of dimension only within parts. We are unable to detect inconsistent dimensions across - /// parts - for this, a little help from the user is needed, e.g. CONSTRAINT cnstr CHECK length(array) = 42. - if (index->dimensions() != dimensions) - throw Exception(ErrorCodes::INCORRECT_DATA, "All arrays in column with vector similarity index must have equal length"); + if (!index) + index = std::make_shared(dimensions, metric_kind, scalar_kind, usearch_hnsw_params); - /// We use Usearch's index_dense_t as index type which supports only 4 bio entries according to https://github.com/unum-cloud/usearch/tree/main/cpp - if (index->size() + rows > std::numeric_limits::max()) - throw Exception(ErrorCodes::INCORRECT_DATA, "Size of vector similarity index would exceed 4 billion entries"); + /// Also check that previously inserted blocks have the same size as this block. + /// Note that this guarantees consistency of dimension only within parts. We are unable to detect inconsistent dimensions across + /// parts - for this, a little help from the user is needed, e.g. CONSTRAINT cnstr CHECK length(array) = 42. + if (index->dimensions() != dimensions) + throw Exception(ErrorCodes::INCORRECT_DATA, "All arrays in column with vector similarity index must have equal length"); - DataTypePtr data_type = block.getDataTypes()[0]; - const auto * data_type_array = typeid_cast(data_type.get()); - if (!data_type_array) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected data type Array(Float*)"); - const TypeIndex nested_type_index = data_type_array->getNestedType()->getTypeId(); + /// We use Usearch's index_dense_t as index type which supports only 4 bio entries according to https://github.com/unum-cloud/usearch/tree/main/cpp + if (index->size() + rows > std::numeric_limits::max()) + throw Exception(ErrorCodes::INCORRECT_DATA, "Size of vector similarity index would exceed 4 billion entries"); - if (WhichDataType(nested_type_index).isFloat32()) - updateImpl(column_array, column_array_offsets, index, dimensions, rows); - else if (WhichDataType(nested_type_index).isFloat64()) - updateImpl(column_array, column_array_offsets, index, dimensions, rows); - else - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected data type Array(Float*)"); + DataTypePtr data_type = index_column_with_type_and_name.type; + const auto * data_type_array = typeid_cast(data_type.get()); + if (!data_type_array) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected data type Array(Float*)"); + const TypeIndex nested_type_index = data_type_array->getNestedType()->getTypeId(); + + if (WhichDataType(nested_type_index).isFloat32()) + updateImpl(column_array, column_array_offsets, index, dimensions, rows); + else if (WhichDataType(nested_type_index).isFloat64()) + updateImpl(column_array, column_array_offsets, index, dimensions, rows); + else + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected data type Array(Float*)"); + } *pos += rows_read; diff --git a/tests/queries/0_stateless/02354_vector_search_bugs.reference b/tests/queries/0_stateless/02354_vector_search_bugs.reference index 9b610cf543a..dec921cf586 100644 --- a/tests/queries/0_stateless/02354_vector_search_bugs.reference +++ b/tests/queries/0_stateless/02354_vector_search_bugs.reference @@ -41,3 +41,4 @@ Expression (Projection) Parts: 1/1 Granules: 4/4 index_granularity_bytes = 0 is disallowed +Issue #71381: Vector similarity index and other skipping indexes used on the same table diff --git a/tests/queries/0_stateless/02354_vector_search_bugs.sql b/tests/queries/0_stateless/02354_vector_search_bugs.sql index d55bdb88a76..6bcb0f78e75 100644 --- a/tests/queries/0_stateless/02354_vector_search_bugs.sql +++ b/tests/queries/0_stateless/02354_vector_search_bugs.sql @@ -117,3 +117,18 @@ CREATE TABLE tab(id Int32, vec Array(Float32)) ENGINE = MergeTree ORDER BY id SE ALTER TABLE tab ADD INDEX vec_idx1(vec) TYPE vector_similarity('hnsw', 'cosineDistance'); -- { serverError INVALID_SETTING_VALUE } DROP TABLE tab; + +SELECT 'Issue #71381: Vector similarity index and other skipping indexes used on the same table'; + +CREATE TABLE tab( + val String, + vec Array(Float32), + INDEX ann_idx vec TYPE vector_similarity('hnsw', 'cosineDistance'), + INDEX set_idx val TYPE set(100) GRANULARITY 100 +) +ENGINE = MergeTree() +ORDER BY tuple(); + +INSERT INTO tab VALUES ('hello world', [0.0]); + +DROP TABLE tab; From 8c2d1ec7f8ef625c7bfb914a551af183520a3119 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Tue, 5 Nov 2024 12:35:23 +0100 Subject: [PATCH 433/680] Allow ExecuteScalarSubqueriesVisitor on ARRAY JOIN --- .../ExecuteScalarSubqueriesVisitor.cpp | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp b/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp index c80852e9ae7..943febf4b0e 100644 --- a/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp +++ b/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp @@ -63,10 +63,22 @@ bool ExecuteScalarSubqueriesMatcher::needChildVisit(ASTPtr & node, const ASTPtr if (node->as()) { /// Do not go to FROM, JOIN, UNION. - if (child->as() || child->as() || child->as()) + if (child->as() || child->as()) return false; } + if (auto tables = node->as()) + { + /// Contrary to what's said in the code block above, ARRAY JOIN needs to resolve the subquery if possible + /// and assign an alias for 02367_optimize_trivial_count_with_array_join to pass. Otherwise it will fail in + /// ArrayJoinedColumnsVisitor (`No alias for non-trivial value in ARRAY JOIN: _a`) + /// This looks 100% as a incomplete code working on top of a bug, but this code has already been made obsolete + /// by the new analyzer, so it's an inconvenience we can live with until we deprecate it. + if (child == tables->array_join) + return true; + return false; + } + return true; } From 996773b205121f55d6f066826dab95b38b49dbbf Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Tue, 5 Nov 2024 12:39:36 +0100 Subject: [PATCH 434/680] Test with both analyzers --- .../03261_optimize_rewrite_array_exists_to_has_crash.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03261_optimize_rewrite_array_exists_to_has_crash.sql b/tests/queries/0_stateless/03261_optimize_rewrite_array_exists_to_has_crash.sql index 5a54d86f339..e0018632be4 100644 --- a/tests/queries/0_stateless/03261_optimize_rewrite_array_exists_to_has_crash.sql +++ b/tests/queries/0_stateless/03261_optimize_rewrite_array_exists_to_has_crash.sql @@ -7,4 +7,4 @@ INNER JOIN rewrite AS y ON ( SELECT 1 ) INNER JOIN rewrite AS z ON 1 -SETTINGS allow_experimental_analyzer=0, optimize_rewrite_array_exists_to_has=1; +SETTINGS optimize_rewrite_array_exists_to_has=1; From bbe28d45bff0bd721685c812706f113e1412ed6b Mon Sep 17 00:00:00 2001 From: vdimir Date: Tue, 5 Nov 2024 12:33:25 +0000 Subject: [PATCH 435/680] fix --- src/Parsers/ASTFunction.cpp | 5 ++- src/TableFunctions/TableFunctionMongoDB.cpp | 42 +++++++++---------- src/TableFunctions/TableFunctionMongoDB.h | 15 +++++++ .../TableFunctionMongoDBPocoLegacy.cpp | 15 ++----- .../03261_mongodb_argumetns_crash.sql | 1 + 5 files changed, 45 insertions(+), 33 deletions(-) create mode 100644 src/TableFunctions/TableFunctionMongoDB.h diff --git a/src/Parsers/ASTFunction.cpp b/src/Parsers/ASTFunction.cpp index 53d44e2f325..11cfe2e584e 100644 --- a/src/Parsers/ASTFunction.cpp +++ b/src/Parsers/ASTFunction.cpp @@ -724,7 +724,10 @@ void ASTFunction::formatImplWithoutAlias(const FormatSettings & settings, Format { if (secret_arguments.are_named) { - assert_cast(argument.get())->arguments->children[0]->formatImpl(settings, state, nested_dont_need_parens); + if (const auto * func_ast = typeid_cast(argument.get())) + func_ast->arguments->children[0]->formatImpl(settings, state, nested_dont_need_parens); + else + argument->formatImpl(settings, state, nested_dont_need_parens); settings.ostr << (settings.hilite ? hilite_operator : "") << " = " << (settings.hilite ? hilite_none : ""); } if (!secret_arguments.replacement.empty()) diff --git a/src/TableFunctions/TableFunctionMongoDB.cpp b/src/TableFunctions/TableFunctionMongoDB.cpp index 966ce858875..9f91839fb33 100644 --- a/src/TableFunctions/TableFunctionMongoDB.cpp +++ b/src/TableFunctions/TableFunctionMongoDB.cpp @@ -15,7 +15,7 @@ #include #include #include - +#include namespace DB { @@ -85,17 +85,11 @@ void TableFunctionMongoDB::parseArguments(const ASTPtr & ast_function, ContextPt { if (const auto * ast_func = typeid_cast(args[i].get())) { - const auto * args_expr = assert_cast(ast_func->arguments.get()); - auto function_args = args_expr->children; - if (function_args.size() != 2) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument"); - - auto arg_name = function_args[0]->as()->name(); - + const auto & [arg_name, arg_value] = getKeyValueMongoDBArgument(ast_func); if (arg_name == "structure") - structure = checkAndGetLiteralArgument(function_args[1], "structure"); + structure = checkAndGetLiteralArgument(arg_value, arg_name); else if (arg_name == "options") - main_arguments.push_back(function_args[1]); + main_arguments.push_back(arg_value); } else if (i == 5) { @@ -117,19 +111,11 @@ void TableFunctionMongoDB::parseArguments(const ASTPtr & ast_function, ContextPt { if (const auto * ast_func = typeid_cast(args[i].get())) { - const auto * args_expr = assert_cast(ast_func->arguments.get()); - const auto & function_args = args_expr->children; - if (function_args.size() != 2 || ast_func->name != "equals" || function_args[0]->as()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument, got {}", ast_func->formatForErrorMessage()); - - auto arg_name = function_args[0]->as()->name(); - + const auto & [arg_name, arg_value] = getKeyValueMongoDBArgument(ast_func); if (arg_name == "structure") - structure = checkAndGetLiteralArgument(function_args[1], "structure"); + structure = checkAndGetLiteralArgument(arg_value, arg_name); else if (arg_name == "options") - main_arguments.push_back(function_args[1]); - else - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument, got {}", ast_func->formatForErrorMessage()); + main_arguments.push_back(arg_value); } else if (i == 2) { @@ -149,6 +135,20 @@ void TableFunctionMongoDB::parseArguments(const ASTPtr & ast_function, ContextPt } +std::pair getKeyValueMongoDBArgument(const ASTFunction * ast_func) +{ + const auto * args_expr = assert_cast(ast_func->arguments.get()); + const auto & function_args = args_expr->children; + if (function_args.size() != 2 || ast_func->name != "equals" || !function_args[0]->as()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument, got {}", ast_func->formatForErrorMessage()); + + const auto & arg_name = function_args[0]->as()->name(); + if (arg_name == "structure" || arg_name == "options") + return std::make_pair(arg_name, function_args[1]); + + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument, got {}", ast_func->formatForErrorMessage()); +} + void registerTableFunctionMongoDB(TableFunctionFactory & factory) { factory.registerFunction( diff --git a/src/TableFunctions/TableFunctionMongoDB.h b/src/TableFunctions/TableFunctionMongoDB.h new file mode 100644 index 00000000000..2b75fda1675 --- /dev/null +++ b/src/TableFunctions/TableFunctionMongoDB.h @@ -0,0 +1,15 @@ + +#include + +#include +#include +#include + + +namespace DB +{ + +std::pair getKeyValueMongoDBArgument(const ASTFunction * ast_func); + +} + diff --git a/src/TableFunctions/TableFunctionMongoDBPocoLegacy.cpp b/src/TableFunctions/TableFunctionMongoDBPocoLegacy.cpp index 70b28ddfaf0..4e27fd35e12 100644 --- a/src/TableFunctions/TableFunctionMongoDBPocoLegacy.cpp +++ b/src/TableFunctions/TableFunctionMongoDBPocoLegacy.cpp @@ -15,6 +15,7 @@ #include #include #include +#include namespace DB @@ -97,19 +98,11 @@ void TableFunctionMongoDBPocoLegacy::parseArguments(const ASTPtr & ast_function, { if (const auto * ast_func = typeid_cast(args[i].get())) { - const auto * args_expr = assert_cast(ast_func->arguments.get()); - const auto & function_args = args_expr->children; - if (function_args.size() != 2 || ast_func->name != "equals" || function_args[0]->as()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument, got {}", ast_func->formatForErrorMessage()); - - auto arg_name = function_args[0]->as()->name(); - + const auto & [arg_name, arg_value] = getKeyValueMongoDBArgument(ast_func); if (arg_name == "structure") - structure = checkAndGetLiteralArgument(function_args[1], "structure"); + structure = checkAndGetLiteralArgument(arg_value, "structure"); else if (arg_name == "options") - main_arguments.push_back(function_args[1]); - else - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Expected key-value defined argument, got {}", ast_func->formatForErrorMessage()); + main_arguments.push_back(arg_value); } else if (i == 5) { diff --git a/tests/queries/0_stateless/03261_mongodb_argumetns_crash.sql b/tests/queries/0_stateless/03261_mongodb_argumetns_crash.sql index 830d3995bd5..ca558ac6bc6 100644 --- a/tests/queries/0_stateless/03261_mongodb_argumetns_crash.sql +++ b/tests/queries/0_stateless/03261_mongodb_argumetns_crash.sql @@ -11,3 +11,4 @@ SELECT * FROM mongodb('mongodb://some-cluster:27017/?retryWrites=false', 'test', SELECT * FROM mongodb('mongodb://some-cluster:27017/?retryWrites=false', 'test', 'my_collection', 'test_user', 'password', NULL, 'x Int32'); -- { serverError BAD_ARGUMENTS } SELECT * FROM mongodb(NULL, 'test', 'my_collection', 'test_user', 'password', 'x Int32'); -- { serverError BAD_ARGUMENTS } +CREATE TABLE IF NOT EXISTS store_version ( `_id` String ) ENGINE = MongoDB(`localhost:27017`, mongodb, storeinfo, adminUser, adminUser); -- { serverError NAMED_COLLECTION_DOESNT_EXIST } From d7977f0b916ccdcc240de8d413015532d492f668 Mon Sep 17 00:00:00 2001 From: kssenii Date: Tue, 5 Nov 2024 13:36:27 +0100 Subject: [PATCH 436/680] More correct assertion --- src/Interpreters/Cache/EvictionCandidates.cpp | 3 ++- src/Interpreters/Cache/FileSegment.cpp | 7 ++++--- src/Interpreters/Cache/FileSegment.h | 7 +++++-- 3 files changed, 11 insertions(+), 6 deletions(-) diff --git a/src/Interpreters/Cache/EvictionCandidates.cpp b/src/Interpreters/Cache/EvictionCandidates.cpp index 08776ad5aee..f5d5fdec6ba 100644 --- a/src/Interpreters/Cache/EvictionCandidates.cpp +++ b/src/Interpreters/Cache/EvictionCandidates.cpp @@ -83,7 +83,8 @@ void EvictionCandidates::removeQueueEntries(const CachePriorityGuard::Lock & loc queue_iterator->invalidate(); chassert(candidate->releasable()); - candidate->file_segment->resetQueueIterator(); + candidate->file_segment->markDelayedRemovalAndResetQueueIterator(); + /// We need to set removed flag in file segment metadata, /// because in dynamic cache resize we first remove queue entries, /// then evict which also removes file segment metadata, diff --git a/src/Interpreters/Cache/FileSegment.cpp b/src/Interpreters/Cache/FileSegment.cpp index 080b54feb06..307d9c8afe1 100644 --- a/src/Interpreters/Cache/FileSegment.cpp +++ b/src/Interpreters/Cache/FileSegment.cpp @@ -171,10 +171,11 @@ void FileSegment::setQueueIterator(Priority::IteratorPtr iterator) queue_iterator = iterator; } -void FileSegment::resetQueueIterator() +void FileSegment::markDelayedRemovalAndResetQueueIterator() { auto lk = lock(); - queue_iterator.reset(); + on_delayed_removal = true; + queue_iterator = {}; } size_t FileSegment::getCurrentWriteOffset() const @@ -861,7 +862,7 @@ bool FileSegment::assertCorrectnessUnlocked(const FileSegmentGuard::Lock & lock) chassert(downloaded_size > 0); chassert(fs::file_size(getPath()) > 0); - chassert(queue_iterator); + chassert(queue_iterator || on_delayed_removal); check_iterator(queue_iterator); break; } diff --git a/src/Interpreters/Cache/FileSegment.h b/src/Interpreters/Cache/FileSegment.h index 79adc342329..6946d70b764 100644 --- a/src/Interpreters/Cache/FileSegment.h +++ b/src/Interpreters/Cache/FileSegment.h @@ -177,7 +177,7 @@ public: void setQueueIterator(Priority::IteratorPtr iterator); - void resetQueueIterator(); + void markDelayedRemovalAndResetQueueIterator(); KeyMetadataPtr tryGetKeyMetadata() const; @@ -249,11 +249,12 @@ private: String tryGetPath() const; - Key file_key; + const Key file_key; Range segment_range; const FileSegmentKind segment_kind; /// Size of the segment is not known until it is downloaded and /// can be bigger than max_file_segment_size. + /// is_unbound == true for temporary data in cache. const bool is_unbound; const bool background_download_enabled; @@ -279,6 +280,8 @@ private: std::atomic hits_count = 0; /// cache hits. std::atomic ref_count = 0; /// Used for getting snapshot state + bool on_delayed_removal = false; + CurrentMetrics::Increment metric_increment{CurrentMetrics::CacheFileSegments}; }; From ead7630d04b5aab7ff28a0e99710a8b6ce17800c Mon Sep 17 00:00:00 2001 From: maxvostrikov Date: Tue, 5 Nov 2024 14:28:28 +0100 Subject: [PATCH 437/680] Missing tests in several tests in 24.10 Added corner cases for tests for: to_utc_timestamp and from_utc_timestamp (more timezones, spetial timezones, epoch corners does not look right, raising a bug over that) arrayUnion (empty and big arrays) quantilesExactWeightedInterpolated (more data types) --- .../02812_from_to_utc_timestamp.reference | 5 +++ .../02812_from_to_utc_timestamp.sh | 8 +++- .../0_stateless/03224_arrayUnion.reference | 10 +++++ .../queries/0_stateless/03224_arrayUnion.sql | 21 ++++++++- ...tile_exact_weighted_interpolated.reference | 13 +++--- ...0_quantile_exact_weighted_interpolated.sql | 45 ++++++++++++++++--- 6 files changed, 88 insertions(+), 14 deletions(-) diff --git a/tests/queries/0_stateless/02812_from_to_utc_timestamp.reference b/tests/queries/0_stateless/02812_from_to_utc_timestamp.reference index 4da8a9784dd..bdce849e069 100644 --- a/tests/queries/0_stateless/02812_from_to_utc_timestamp.reference +++ b/tests/queries/0_stateless/02812_from_to_utc_timestamp.reference @@ -3,3 +3,8 @@ 3 2023-03-16 12:22:33 2023-03-16 10:22:33.000 2023-03-16 03:22:33 2023-03-16 19:22:33.123 2024-02-24 10:22:33 2024-02-24 12:22:33 2024-10-24 09:22:33 2024-10-24 13:22:33 +2024-10-24 16:22:33 2024-10-24 06:22:33 +leap year: 2024-02-29 16:22:33 2024-02-29 06:22:33 +non-leap year: 2023-03-01 16:22:33 2023-03-01 06:22:33 +timezone with half-hour offset: 2024-02-29 00:52:33 2024-02-29 21:52:33 +jump over a year: 2024-01-01 04:01:01 2023-12-31 20:01:01 diff --git a/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh b/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh index 835dab8af57..441fc254256 100755 --- a/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh +++ b/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh @@ -15,4 +15,10 @@ $CLICKHOUSE_CLIENT -q "select x, to_utc_timestamp(toDateTime('2023-03-16 11:22:3 # timestamp convert between DST timezone and UTC $CLICKHOUSE_CLIENT -q "select to_utc_timestamp(toDateTime('2024-02-24 11:22:33'), 'Europe/Madrid'), from_utc_timestamp(toDateTime('2024-02-24 11:22:33'), 'Europe/Madrid')" $CLICKHOUSE_CLIENT -q "select to_utc_timestamp(toDateTime('2024-10-24 11:22:33'), 'Europe/Madrid'), from_utc_timestamp(toDateTime('2024-10-24 11:22:33'), 'Europe/Madrid')" -$CLICKHOUSE_CLIENT -q "drop table test_tbl" \ No newline at end of file +$CLICKHOUSE_CLIENT -q "select to_utc_timestamp(toDateTime('2024-10-24 11:22:33'), 'EST'), from_utc_timestamp(toDateTime('2024-10-24 11:22:33'), 'EST')" +$CLICKHOUSE_CLIENT -q "select 'leap year:', to_utc_timestamp(toDateTime('2024-02-29 11:22:33'), 'EST'), from_utc_timestamp(toDateTime('2024-02-29 11:22:33'), 'EST')" +$CLICKHOUSE_CLIENT -q "select 'non-leap year:', to_utc_timestamp(toDateTime('2023-02-29 11:22:33'), 'EST'), from_utc_timestamp(toDateTime('2023-02-29 11:22:33'), 'EST')" +$CLICKHOUSE_CLIENT -q "select 'timezone with half-hour offset:', to_utc_timestamp(toDateTime('2024-02-29 11:22:33'), 'Australia/Adelaide'), from_utc_timestamp(toDateTime('2024-02-29 11:22:33'), 'Australia/Adelaide')" +$CLICKHOUSE_CLIENT -q "select 'jump over a year:', to_utc_timestamp(toDateTime('2023-12-31 23:01:01'), 'EST'), from_utc_timestamp(toDateTime('2024-01-01 01:01:01'), 'EST')" + +$CLICKHOUSE_CLIENT -q "drop table test_tbl" diff --git a/tests/queries/0_stateless/03224_arrayUnion.reference b/tests/queries/0_stateless/03224_arrayUnion.reference index b900b6cdb0a..9b871234d27 100644 --- a/tests/queries/0_stateless/03224_arrayUnion.reference +++ b/tests/queries/0_stateless/03224_arrayUnion.reference @@ -41,3 +41,13 @@ [1,2,3,4,5,10,20] ------- [1,2,3] +------- +[10,-2,1] ['hello','hi'] [3,2,1,NULL] +------- +------- +[1] +------- +[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199,200,201,202,203,204,205,206,207,208,209,210,211,212,213,214,215,216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231,232,233,234,235,236,237,238,239,240,241,242,243,244,245,246,247,248,249,250,251,252,253,254,255,256] +199999 +------- +[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19] diff --git a/tests/queries/0_stateless/03224_arrayUnion.sql b/tests/queries/0_stateless/03224_arrayUnion.sql index dedbacad906..14a9192f39a 100644 --- a/tests/queries/0_stateless/03224_arrayUnion.sql +++ b/tests/queries/0_stateless/03224_arrayUnion.sql @@ -35,4 +35,23 @@ SELECT arraySort(arrayUnion([NULL, NULL, NULL, 1], [1, NULL, NULL], [1, 2, 3, NU select '-------'; SELECT arraySort(arrayUnion([1, 1, 1, 2, 3], [2, 2, 4], [5, 10, 20])); select '-------'; -SELECT arraySort(arrayUnion([1, 2], [1, 3], [])), +SELECT arraySort(arrayUnion([1, 2], [1, 3], [])); +select '-------'; +-- example from docs +SELECT + arrayUnion([-2, 1], [10, 1], [-2], []) as num_example, + arrayUnion(['hi'], [], ['hello', 'hi']) as str_example, + arrayUnion([1, 3, NULL], [2, 3, NULL]) as null_example; +select '-------'; +--mix of types +SELECT arrayUnion([1], [-2], [1.1, 'hi'], [NULL, 'hello', []]); -- {serverError NO_COMMON_TYPE} +select '-------'; +SELECT arrayUnion([1]); +SELECT arrayUnion(); -- {serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH} +select '-------'; +--bigger arrays +SELECT arraySort(arrayUnion(range(1, 256), range(2, 257))); +SELECT length(arrayUnion(range(1, 100000), range(9999, 200000))); +select '-------'; +--bigger number of arguments +SELECT arraySort(arrayUnion([1, 2], [1, 3], [1, 4], [1, 5], [1, 6], [1, 7], [1, 8], [1, 9], [1, 10], [1, 11], [1, 12], [1, 13], [1, 14], [1, 15], [1, 16], [1, 17], [1, 18], [1, 19])); diff --git a/tests/queries/0_stateless/03240_quantile_exact_weighted_interpolated.reference b/tests/queries/0_stateless/03240_quantile_exact_weighted_interpolated.reference index 23cbe2bfdec..ccb315b8305 100644 --- a/tests/queries/0_stateless/03240_quantile_exact_weighted_interpolated.reference +++ b/tests/queries/0_stateless/03240_quantile_exact_weighted_interpolated.reference @@ -1,6 +1,6 @@ quantileExactWeightedInterpolated -0 0 0 Decimal(38, 8) --25.5 -8.49999999 -5.1 Decimal(38, 8) +0 0 0 25 2024-02-20 Decimal(38, 8) +-25.5 -8.49999999 -5.1 12.25 2024-01-25 Decimal(38, 8) 0 0 0 10 3.33333333 2 20 6.66666666 4 @@ -10,11 +10,14 @@ quantileExactWeightedInterpolated [-50,-40,-30,-20,-10,0,10,20,30,40,50] [-16.66666666,-13.33333333,-10,-6.66666666,-3.33333333,0,3.33333333,6.66666666,10,13.33333333,16.66666666] [-10,-8,-6,-4,-2,0,2,4,6,8,10] +[0,5,10,15,20,25,30,35,40,45,50] +['2024-01-01','2024-01-11','2024-01-21','2024-01-31','2024-02-10','2024-02-20','2024-03-01','2024-03-11','2024-03-21','2024-03-31','2024-04-10'] quantileExactWeightedInterpolatedState [10000.6,20000.2,29999.8,39999.4] Test with filter that returns no rows -0 0 0 +0 0 0 nan 1970-01-01 +0 0 0 nan 1970-01-01 Test with dynamic weights -21 7 4.2 +21 7 4.2 35.5 2024-03-12 Test with all weights set to 0 -0 0 0 +0 0 0 nan 1970-01-01 diff --git a/tests/queries/0_stateless/03240_quantile_exact_weighted_interpolated.sql b/tests/queries/0_stateless/03240_quantile_exact_weighted_interpolated.sql index dba16eae22a..a64b46e751b 100644 --- a/tests/queries/0_stateless/03240_quantile_exact_weighted_interpolated.sql +++ b/tests/queries/0_stateless/03240_quantile_exact_weighted_interpolated.sql @@ -5,16 +5,28 @@ CREATE TABLE decimal a Decimal32(4), b Decimal64(8), c Decimal128(8), + f Float64, + d Date, w UInt64 ) ENGINE = Memory; -INSERT INTO decimal (a, b, c, w) -SELECT toDecimal32(number - 50, 4), toDecimal64(number - 50, 8) / 3, toDecimal128(number - 50, 8) / 5, number +INSERT INTO decimal (a, b, c, f, d, w) +SELECT toDecimal32(number - 50, 4), toDecimal64(number - 50, 8) / 3, toDecimal128(number - 50, 8) / 5, number/2, addDays(toDate('2024-01-01'), number), number FROM system.numbers LIMIT 101; SELECT 'quantileExactWeightedInterpolated'; -SELECT medianExactWeightedInterpolated(a, 1), medianExactWeightedInterpolated(b, 2), medianExactWeightedInterpolated(c, 3) as x, toTypeName(x) FROM decimal; -SELECT quantileExactWeightedInterpolated(a, 1), quantileExactWeightedInterpolated(b, 2), quantileExactWeightedInterpolated(c, 3) as x, toTypeName(x) FROM decimal WHERE a < 0; +SELECT medianExactWeightedInterpolated(a, 1), + medianExactWeightedInterpolated(b, 2), + medianExactWeightedInterpolated(c, 3) as x, + medianExactWeightedInterpolated(f, 4), + medianExactWeightedInterpolated(d, 5), + toTypeName(x) FROM decimal; +SELECT quantileExactWeightedInterpolated(a, 1), + quantileExactWeightedInterpolated(b, 2), + quantileExactWeightedInterpolated(c, 3) as x, + quantileExactWeightedInterpolated(f, 4), + quantileExactWeightedInterpolated(d, 5), + toTypeName(x) FROM decimal WHERE a < 0; SELECT quantileExactWeightedInterpolated(0.0)(a, 1), quantileExactWeightedInterpolated(0.0)(b, 2), quantileExactWeightedInterpolated(0.0)(c, 3) FROM decimal WHERE a >= 0; SELECT quantileExactWeightedInterpolated(0.2)(a, 1), quantileExactWeightedInterpolated(0.2)(b, 2), quantileExactWeightedInterpolated(0.2)(c, 3) FROM decimal WHERE a >= 0; SELECT quantileExactWeightedInterpolated(0.4)(a, 1), quantileExactWeightedInterpolated(0.4)(b, 2), quantileExactWeightedInterpolated(0.4)(c, 3) FROM decimal WHERE a >= 0; @@ -24,6 +36,8 @@ SELECT quantileExactWeightedInterpolated(1.0)(a, 1), quantileExactWeightedInterp SELECT quantilesExactWeightedInterpolated(0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0)(a, 1) FROM decimal; SELECT quantilesExactWeightedInterpolated(0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0)(b, 2) FROM decimal; SELECT quantilesExactWeightedInterpolated(0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0)(c, 3) FROM decimal; +SELECT quantilesExactWeightedInterpolated(0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0)(f, 4) FROM decimal; +SELECT quantilesExactWeightedInterpolated(0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0)(d, 5) FROM decimal; SELECT 'quantileExactWeightedInterpolatedState'; SELECT quantilesExactWeightedInterpolatedMerge(0.2, 0.4, 0.6, 0.8)(x) @@ -34,12 +48,29 @@ FROM ); SELECT 'Test with filter that returns no rows'; -SELECT medianExactWeightedInterpolated(a, 1), medianExactWeightedInterpolated(b, 2), medianExactWeightedInterpolated(c, 3) FROM decimal WHERE a > 1000; +SELECT medianExactWeightedInterpolated(a, 1), + medianExactWeightedInterpolated(b, 2), + medianExactWeightedInterpolated(c, 3), + medianExactWeightedInterpolated(f, 4), + medianExactWeightedInterpolated(d, 5) FROM decimal WHERE a > 1000; +SELECT quantileExactWeightedInterpolated(a, 1), + quantileExactWeightedInterpolated(b, 2), + quantileExactWeightedInterpolated(c, 3), + quantileExactWeightedInterpolated(f, 4), + quantileExactWeightedInterpolated(d, 5) FROM decimal WHERE d < toDate('2024-01-01'); SELECT 'Test with dynamic weights'; -SELECT medianExactWeightedInterpolated(a, w), medianExactWeightedInterpolated(b, w), medianExactWeightedInterpolated(c, w) FROM decimal; +SELECT medianExactWeightedInterpolated(a, w), + medianExactWeightedInterpolated(b, w), + medianExactWeightedInterpolated(c, w), + medianExactWeightedInterpolated(f, w), + medianExactWeightedInterpolated(d, w) FROM decimal; SELECT 'Test with all weights set to 0'; -SELECT medianExactWeightedInterpolated(a, 0), medianExactWeightedInterpolated(b, 0), medianExactWeightedInterpolated(c, 0) FROM decimal; +SELECT medianExactWeightedInterpolated(a, 0), + medianExactWeightedInterpolated(b, 0), + medianExactWeightedInterpolated(c, 0), + medianExactWeightedInterpolated(f, 0), + medianExactWeightedInterpolated(d, 0) FROM decimal; DROP TABLE IF EXISTS decimal; From 5152984bb170e5c63144db3dd238099534353378 Mon Sep 17 00:00:00 2001 From: vdimir Date: Tue, 5 Nov 2024 13:52:14 +0000 Subject: [PATCH 438/680] upd src/TableFunctions/TableFunctionMongoDB.h --- src/TableFunctions/TableFunctionMongoDB.h | 1 + 1 file changed, 1 insertion(+) diff --git a/src/TableFunctions/TableFunctionMongoDB.h b/src/TableFunctions/TableFunctionMongoDB.h index 2b75fda1675..2ab8ee9479f 100644 --- a/src/TableFunctions/TableFunctionMongoDB.h +++ b/src/TableFunctions/TableFunctionMongoDB.h @@ -1,3 +1,4 @@ +#pragma once #include From c16e1f021b7c24250ebf3bef1c764ba7c218de0d Mon Sep 17 00:00:00 2001 From: Anton Popov Date: Tue, 5 Nov 2024 14:57:35 +0000 Subject: [PATCH 439/680] fix memory usage in inserts with delayed streams --- .../MergeTree/IMergeTreeDataPartWriter.h | 2 ++ .../MergeTree/IMergedBlockOutputStream.h | 5 +++++ .../MergeTreeDataPartWriterCompact.h | 2 ++ .../MergeTree/MergeTreeDataPartWriterWide.h | 2 ++ src/Storages/MergeTree/MergeTreeSink.cpp | 12 +++++++---- .../MergeTree/ReplicatedMergeTreeSink.cpp | 13 ++++++++---- .../03261_delayed_streams_memory.reference | 1 + .../03261_delayed_streams_memory.sql | 20 +++++++++++++++++++ 8 files changed, 49 insertions(+), 8 deletions(-) create mode 100644 tests/queries/0_stateless/03261_delayed_streams_memory.reference create mode 100644 tests/queries/0_stateless/03261_delayed_streams_memory.sql diff --git a/src/Storages/MergeTree/IMergeTreeDataPartWriter.h b/src/Storages/MergeTree/IMergeTreeDataPartWriter.h index b8ac14b1750..d1c76505d7c 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPartWriter.h +++ b/src/Storages/MergeTree/IMergeTreeDataPartWriter.h @@ -46,6 +46,8 @@ public: virtual void finish(bool sync) = 0; + virtual size_t getNumberOfOpenStreams() const = 0; + Columns releaseIndexColumns(); PlainMarksByName releaseCachedMarks(); diff --git a/src/Storages/MergeTree/IMergedBlockOutputStream.h b/src/Storages/MergeTree/IMergedBlockOutputStream.h index a901b03c115..7dd6d720170 100644 --- a/src/Storages/MergeTree/IMergedBlockOutputStream.h +++ b/src/Storages/MergeTree/IMergedBlockOutputStream.h @@ -39,6 +39,11 @@ public: return writer->releaseCachedMarks(); } + size_t getNumberOfOpenStreams() const + { + return writer->getNumberOfOpenStreams(); + } + protected: /// Remove all columns marked expired in data_part. Also, clears checksums diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h index b440a37222d..20c47fb8314 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h @@ -32,6 +32,8 @@ public: void fillChecksums(MergeTreeDataPartChecksums & checksums, NameSet & checksums_to_remove) override; void finish(bool sync) override; + size_t getNumberOfOpenStreams() const override { return 1; } + private: /// Finish serialization of the data. Flush rows in buffer to disk, compute checksums. void fillDataChecksums(MergeTreeDataPartChecksums & checksums); diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h index 68f016a7421..b594b2d79bb 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h @@ -43,6 +43,8 @@ public: void finish(bool sync) final; + size_t getNumberOfOpenStreams() const override { return column_streams.size(); } + private: /// Finish serialization of data: write final mark if required and compute checksums /// Also validate written data in debug mode diff --git a/src/Storages/MergeTree/MergeTreeSink.cpp b/src/Storages/MergeTree/MergeTreeSink.cpp index 604112c26ea..99852309c77 100644 --- a/src/Storages/MergeTree/MergeTreeSink.cpp +++ b/src/Storages/MergeTree/MergeTreeSink.cpp @@ -94,7 +94,7 @@ void MergeTreeSink::consume(Chunk & chunk) DelayedPartitions partitions; const Settings & settings = context->getSettingsRef(); - size_t streams = 0; + size_t total_streams = 0; bool support_parallel_write = false; auto token_info = chunk.getChunkInfos().get(); @@ -153,16 +153,18 @@ void MergeTreeSink::consume(Chunk & chunk) max_insert_delayed_streams_for_parallel_write = 0; /// In case of too much columns/parts in block, flush explicitly. - streams += temp_part.streams.size(); + size_t current_streams = 0; + for (const auto & stream : temp_part.streams) + current_streams += stream.stream->getNumberOfOpenStreams(); - if (streams > max_insert_delayed_streams_for_parallel_write) + if (total_streams + current_streams > max_insert_delayed_streams_for_parallel_write) { finishDelayedChunk(); delayed_chunk = std::make_unique(); delayed_chunk->partitions = std::move(partitions); finishDelayedChunk(); - streams = 0; + total_streams = 0; support_parallel_write = false; partitions = DelayedPartitions{}; } @@ -174,6 +176,8 @@ void MergeTreeSink::consume(Chunk & chunk) .block_dedup_token = block_dedup_token, .part_counters = std::move(part_counters), }); + + total_streams += current_streams; } if (need_to_define_dedup_token) diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp index f1b0e5ec385..f3ae6e77ac3 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp @@ -341,7 +341,7 @@ void ReplicatedMergeTreeSinkImpl::consume(Chunk & chunk) using DelayedPartitions = std::vector; DelayedPartitions partitions; - size_t streams = 0; + size_t total_streams = 0; bool support_parallel_write = false; for (auto & current_block : part_blocks) @@ -418,15 +418,18 @@ void ReplicatedMergeTreeSinkImpl::consume(Chunk & chunk) max_insert_delayed_streams_for_parallel_write = 0; /// In case of too much columns/parts in block, flush explicitly. - streams += temp_part.streams.size(); - if (streams > max_insert_delayed_streams_for_parallel_write) + size_t current_streams = 0; + for (const auto & stream : temp_part.streams) + current_streams += stream.stream->getNumberOfOpenStreams(); + + if (total_streams + current_streams > max_insert_delayed_streams_for_parallel_write) { finishDelayedChunk(zookeeper); delayed_chunk = std::make_unique::DelayedChunk>(replicas_num); delayed_chunk->partitions = std::move(partitions); finishDelayedChunk(zookeeper); - streams = 0; + total_streams = 0; support_parallel_write = false; partitions = DelayedPartitions{}; } @@ -447,6 +450,8 @@ void ReplicatedMergeTreeSinkImpl::consume(Chunk & chunk) std::move(unmerged_block), std::move(part_counters) /// profile_events_scope must be reset here. )); + + total_streams += current_streams; } if (need_to_define_dedup_token) diff --git a/tests/queries/0_stateless/03261_delayed_streams_memory.reference b/tests/queries/0_stateless/03261_delayed_streams_memory.reference new file mode 100644 index 00000000000..7326d960397 --- /dev/null +++ b/tests/queries/0_stateless/03261_delayed_streams_memory.reference @@ -0,0 +1 @@ +Ok diff --git a/tests/queries/0_stateless/03261_delayed_streams_memory.sql b/tests/queries/0_stateless/03261_delayed_streams_memory.sql new file mode 100644 index 00000000000..863644a0dff --- /dev/null +++ b/tests/queries/0_stateless/03261_delayed_streams_memory.sql @@ -0,0 +1,20 @@ +-- Tags: long, no-debug, no-asan, no-tsan, no-msan, no-ubsan, no-random-settings, no-random-merge-tree-settings + +DROP TABLE IF EXISTS t_100_columns; + +CREATE TABLE t_100_columns (id UInt64, c0 String, c1 String, c2 String, c3 String, c4 String, c5 String, c6 String, c7 String, c8 String, c9 String, c10 String, c11 String, c12 String, c13 String, c14 String, c15 String, c16 String, c17 String, c18 String, c19 String, c20 String, c21 String, c22 String, c23 String, c24 String, c25 String, c26 String, c27 String, c28 String, c29 String, c30 String, c31 String, c32 String, c33 String, c34 String, c35 String, c36 String, c37 String, c38 String, c39 String, c40 String, c41 String, c42 String, c43 String, c44 String, c45 String, c46 String, c47 String, c48 String, c49 String, c50 String) +ENGINE = MergeTree +ORDER BY id PARTITION BY id % 50 +SETTINGS min_bytes_for_wide_part = 0, ratio_of_defaults_for_sparse_serialization = 1.0, max_compress_block_size = '1M', storage_policy = 's3_cache'; + +SET max_insert_delayed_streams_for_parallel_write = 55; + +INSERT INTO t_100_columns (id) SELECT number FROM numbers(100); + +SYSTEM FLUSH LOGS; + +SELECT if (memory_usage < 300000000, 'Ok', format('Fail: memory usage {}', formatReadableSize(memory_usage))) +FROM system.query_log +WHERE current_database = currentDatabase() AND query LIKE 'INSERT INTO t_100_columns%' AND type = 'QueryFinish'; + +DROP TABLE t_100_columns; From 6c63587f7747cc05e5df4aad259cee40c34ac7c6 Mon Sep 17 00:00:00 2001 From: Vladimir Cherkasov Date: Fri, 1 Nov 2024 13:27:09 +0100 Subject: [PATCH 440/680] More info in TOO_SLOW exception --- src/QueryPipeline/ExecutionSpeedLimits.cpp | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/src/QueryPipeline/ExecutionSpeedLimits.cpp b/src/QueryPipeline/ExecutionSpeedLimits.cpp index 05fd394db77..fc0e86781f0 100644 --- a/src/QueryPipeline/ExecutionSpeedLimits.cpp +++ b/src/QueryPipeline/ExecutionSpeedLimits.cpp @@ -86,10 +86,12 @@ void ExecutionSpeedLimits::throttle( if (timeout_overflow_mode == OverflowMode::THROW && estimated_execution_time_seconds > max_estimated_execution_time.totalSeconds()) throw Exception( ErrorCodes::TOO_SLOW, - "Estimated query execution time ({} seconds) is too long. Maximum: {}. Estimated rows to process: {}", + "Estimated query execution time ({:.5f} seconds) is too long. Maximum: {}. Estimated rows to process: {} ({} read in {:.5f} seconds).", estimated_execution_time_seconds, max_estimated_execution_time.totalSeconds(), - total_rows_to_read); + total_rows_to_read, + read_rows, + elapsed_seconds); } if (max_execution_rps && rows_per_second >= max_execution_rps) From 6ecc673f7d4a9890004a24c16d8c6b9b5a857d93 Mon Sep 17 00:00:00 2001 From: divanik Date: Tue, 5 Nov 2024 16:02:40 +0000 Subject: [PATCH 441/680] Fix quorum inserts tests --- tests/integration/test_quorum_inserts/test.py | 114 +++++++++--------- 1 file changed, 54 insertions(+), 60 deletions(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index eefc4882e8e..de437fc3206 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -2,6 +2,7 @@ import concurrent import time import pytest +import uuid from helpers.cluster import ClickHouseCluster from helpers.network import PartitionManager @@ -46,10 +47,11 @@ def started_cluster(): def test_simple_add_replica(started_cluster): - zero.query("DROP TABLE IF EXISTS test_simple ON CLUSTER cluster") + table_name = "test_simple_" + uuid.uuid4().hex + zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") create_query = ( - "CREATE TABLE test_simple " + f"CREATE TABLE {table_name} " "(a Int8, d Date) " "Engine = ReplicatedMergeTree('/clickhouse/tables/{shard}/{table}', '{replica}') " "PARTITION BY d ORDER BY a" @@ -58,91 +60,81 @@ def test_simple_add_replica(started_cluster): zero.query(create_query) first.query(create_query) - first.query("SYSTEM STOP FETCHES test_simple") + first.query(f"SYSTEM STOP FETCHES {table_name}") zero.query( - "INSERT INTO test_simple VALUES (1, '2011-01-01')", + f"INSERT INTO {table_name} VALUES (1, '2011-01-01')", settings={"insert_quorum": 1}, ) - assert "1\t2011-01-01\n" == zero.query("SELECT * from test_simple") - assert "" == first.query("SELECT * from test_simple") + assert "1\t2011-01-01\n" == zero.query(f"SELECT * from {table_name}") + assert "" == first.query(f"SELECT * from {table_name}") - first.query("SYSTEM START FETCHES test_simple") + first.query(f"SYSTEM START FETCHES {table_name}") - first.query("SYSTEM SYNC REPLICA test_simple", timeout=20) + first.query(f"SYSTEM SYNC REPLICA {table_name}", timeout=20) - assert "1\t2011-01-01\n" == zero.query("SELECT * from test_simple") - assert "1\t2011-01-01\n" == first.query("SELECT * from test_simple") + assert "1\t2011-01-01\n" == zero.query(f"SELECT * from {table_name}") + assert "1\t2011-01-01\n" == first.query(f"SELECT * from {table_name}") second.query(create_query) - second.query("SYSTEM SYNC REPLICA test_simple", timeout=20) + second.query(f"SYSTEM SYNC REPLICA {table_name}", timeout=20) - assert "1\t2011-01-01\n" == zero.query("SELECT * from test_simple") - assert "1\t2011-01-01\n" == first.query("SELECT * from test_simple") - assert "1\t2011-01-01\n" == second.query("SELECT * from test_simple") + assert "1\t2011-01-01\n" == zero.query(f"SELECT * from {table_name}") + assert "1\t2011-01-01\n" == first.query(f"SELECT * from {table_name}") + assert "1\t2011-01-01\n" == second.query(f"SELECT * from {table_name}") - zero.query("DROP TABLE IF EXISTS test_simple ON CLUSTER cluster") + zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") def test_drop_replica_and_achieve_quorum(started_cluster): + table_name = "test_drop_replica_and_achieve_quorum_" + uuid.uuid4().hex zero.query( - "DROP TABLE IF EXISTS test_drop_replica_and_achieve_quorum ON CLUSTER cluster" + f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster" ) - create_query = ( - "CREATE TABLE test_drop_replica_and_achieve_quorum " + f"CREATE TABLE {table_name} " "(a Int8, d Date) " "Engine = ReplicatedMergeTree('/clickhouse/tables/{shard}/{table}', '{replica}') " "PARTITION BY d ORDER BY a" ) - print("Create Replicated table with two replicas") zero.query(create_query) first.query(create_query) - print("Stop fetches on one replica. Since that, it will be isolated.") - first.query("SYSTEM STOP FETCHES test_drop_replica_and_achieve_quorum") - + first.query(f"SYSTEM STOP FETCHES {table_name}") print("Insert to other replica. This query will fail.") quorum_timeout = zero.query_and_get_error( - "INSERT INTO test_drop_replica_and_achieve_quorum(a,d) VALUES (1, '2011-01-01')", + f"INSERT INTO {table_name}(a,d) VALUES (1, '2011-01-01')", settings={"insert_quorum_timeout": 5000}, ) assert "Timeout while waiting for quorum" in quorum_timeout, "Query must fail." - assert TSV("1\t2011-01-01\n") == TSV( zero.query( - "SELECT * FROM test_drop_replica_and_achieve_quorum", + f"SELECT * FROM {table_name}", settings={"select_sequential_consistency": 0}, ) ) - assert TSV("") == TSV( zero.query( - "SELECT * FROM test_drop_replica_and_achieve_quorum", + f"SELECT * FROM {table_name}", settings={"select_sequential_consistency": 1}, ) ) - # TODO:(Mikhaylov) begin; maybe delete this lines. I want clickhouse to fetch parts and update quorum. print("START FETCHES first replica") - first.query("SYSTEM START FETCHES test_drop_replica_and_achieve_quorum") - + first.query(f"SYSTEM START FETCHES {table_name}") print("SYNC first replica") - first.query("SYSTEM SYNC REPLICA test_drop_replica_and_achieve_quorum", timeout=20) + first.query(f"SYSTEM SYNC REPLICA {table_name}", timeout=20) # TODO:(Mikhaylov) end - print("Add second replica") second.query(create_query) - print("SYNC second replica") - second.query("SYSTEM SYNC REPLICA test_drop_replica_and_achieve_quorum", timeout=20) - + second.query(f"SYSTEM SYNC REPLICA {table_name}", timeout=20) print("Quorum for previous insert achieved.") assert TSV("1\t2011-01-01\n") == TSV( second.query( - "SELECT * FROM test_drop_replica_and_achieve_quorum", + f"SELECT * FROM {table_name}", settings={"select_sequential_consistency": 1}, ) ) @@ -296,10 +288,11 @@ def test_insert_quorum_with_move_partition(started_cluster, add_new_data): def test_insert_quorum_with_ttl(started_cluster): - zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster") + table_name = "test_insert_quorum_with_ttl_" + uuid.uuid4().hex + zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") create_query = ( - "CREATE TABLE test_insert_quorum_with_ttl " + f"CREATE TABLE {table_name} " "(a Int8, d Date) " "Engine = ReplicatedMergeTree('/clickhouse/tables/{table}', '{replica}') " "PARTITION BY d ORDER BY a " @@ -311,12 +304,12 @@ def test_insert_quorum_with_ttl(started_cluster): zero.query(create_query) first.query(create_query) - print("Stop fetches for test_insert_quorum_with_ttl at first replica.") - first.query("SYSTEM STOP FETCHES test_insert_quorum_with_ttl") + print(f"Stop fetches for {table_name} at first replica.") + first.query(f"SYSTEM STOP FETCHES {table_name}") print("Insert should fail since it can not reach the quorum.") quorum_timeout = zero.query_and_get_error( - "INSERT INTO test_insert_quorum_with_ttl(a,d) VALUES(1, '2011-01-01')", + f"INSERT INTO {table_name}(a,d) VALUES(1, '2011-01-01')", settings={"insert_quorum_timeout": 5000}, ) assert "Timeout while waiting for quorum" in quorum_timeout, "Query must fail." @@ -327,51 +320,52 @@ def test_insert_quorum_with_ttl(started_cluster): time.sleep(10) assert TSV("1\t2011-01-01\n") == TSV( zero.query( - "SELECT * FROM test_insert_quorum_with_ttl", + f"SELECT * FROM {table_name}", settings={"select_sequential_consistency": 0}, ) ) - print("Resume fetches for test_insert_quorum_with_ttl at first replica.") - first.query("SYSTEM START FETCHES test_insert_quorum_with_ttl") + print(f"Resume fetches for {table_name} at first replica.") + first.query(f"SYSTEM START FETCHES {table_name}") print("Sync first replica.") - first.query("SYSTEM SYNC REPLICA test_insert_quorum_with_ttl") + first.query(f"SYSTEM SYNC REPLICA {table_name}") zero.query( - "INSERT INTO test_insert_quorum_with_ttl(a,d) VALUES(1, '2011-01-01')", + f"INSERT INTO {table_name}(a,d) VALUES(1, '2011-01-01')", settings={"insert_quorum_timeout": 5000}, ) print("Inserts should resume.") - zero.query("INSERT INTO test_insert_quorum_with_ttl(a, d) VALUES(2, '2012-02-02')") + zero.query(f"INSERT INTO {table_name}(a, d) VALUES(2, '2012-02-02')") - first.query("OPTIMIZE TABLE test_insert_quorum_with_ttl") - first.query("SYSTEM SYNC REPLICA test_insert_quorum_with_ttl") - zero.query("SYSTEM SYNC REPLICA test_insert_quorum_with_ttl") + first.query(f"OPTIMIZE TABLE {table_name}") + first.query(f"SYSTEM SYNC REPLICA {table_name}") + zero.query(f"SYSTEM SYNC REPLICA {table_name}") assert TSV("2\t2012-02-02\n") == TSV( first.query( - "SELECT * FROM test_insert_quorum_with_ttl", + f"SELECT * FROM {table_name}", settings={"select_sequential_consistency": 0}, ) ) assert TSV("2\t2012-02-02\n") == TSV( first.query( - "SELECT * FROM test_insert_quorum_with_ttl", + f"SELECT * FROM {table_name}", settings={"select_sequential_consistency": 1}, ) ) - zero.query("DROP TABLE IF EXISTS test_insert_quorum_with_ttl ON CLUSTER cluster") + zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") -def test_insert_quorum_with_keeper_loss_connection(): +def test_insert_quorum_with_keeper_loss_connection(started_cluster): + table_name = "test_insert_quorum_with_keeper_loss_" + uuid.uuid4().hex zero.query( - "DROP TABLE IF EXISTS test_insert_quorum_with_keeper_fail ON CLUSTER cluster" + f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster" ) create_query = ( - "CREATE TABLE test_insert_quorum_with_keeper_loss" + f"CREATE TABLE {table_name} " "(a Int8, d Date) " "Engine = ReplicatedMergeTree('/clickhouse/tables/{table}', '{replica}') " "ORDER BY a " @@ -380,7 +374,7 @@ def test_insert_quorum_with_keeper_loss_connection(): zero.query(create_query) first.query(create_query) - first.query("SYSTEM STOP FETCHES test_insert_quorum_with_keeper_loss") + first.query(f"SYSTEM STOP FETCHES {table_name}") zero.query("SYSTEM ENABLE FAILPOINT replicated_merge_tree_commit_zk_fail_after_op") zero.query("SYSTEM ENABLE FAILPOINT replicated_merge_tree_insert_retry_pause") @@ -388,7 +382,7 @@ def test_insert_quorum_with_keeper_loss_connection(): with concurrent.futures.ThreadPoolExecutor(max_workers=5) as executor: insert_future = executor.submit( lambda: zero.query( - "INSERT INTO test_insert_quorum_with_keeper_loss(a,d) VALUES(1, '2011-01-01')", + f"INSERT INTO {table_name}(a,d) VALUES(1, '2011-01-01')", settings={"insert_quorum_timeout": 150000}, ) ) @@ -401,7 +395,7 @@ def test_insert_quorum_with_keeper_loss_connection(): while True: if ( zk.exists( - "/clickhouse/tables/test_insert_quorum_with_keeper_loss/replicas/zero/is_active" + f"/clickhouse/tables/{table_name}/replicas/zero/is_active" ) is None ): @@ -418,7 +412,7 @@ def test_insert_quorum_with_keeper_loss_connection(): "SYSTEM WAIT FAILPOINT finish_set_quorum_failed_parts", timeout=300 ) ) - first.query("SYSTEM START FETCHES test_insert_quorum_with_keeper_loss") + first.query(f"SYSTEM START FETCHES {table_name}") concurrent.futures.wait([quorum_fail_future]) From 3eedc74c5943f23ed4e360533e6e3bb5a6238109 Mon Sep 17 00:00:00 2001 From: divanik Date: Tue, 5 Nov 2024 16:25:58 +0000 Subject: [PATCH 442/680] Reformatted because of style check --- tests/integration/test_quorum_inserts/test.py | 14 ++++---------- 1 file changed, 4 insertions(+), 10 deletions(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index de437fc3206..824cb371595 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -88,10 +88,8 @@ def test_simple_add_replica(started_cluster): def test_drop_replica_and_achieve_quorum(started_cluster): - table_name = "test_drop_replica_and_achieve_quorum_" + uuid.uuid4().hex - zero.query( - f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster" - ) + table_name = "test_drop_replica_and_achieve_quorum_" + uuid.uuid4().hex + zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") create_query = ( f"CREATE TABLE {table_name} " "(a Int8, d Date) " @@ -361,9 +359,7 @@ def test_insert_quorum_with_ttl(started_cluster): def test_insert_quorum_with_keeper_loss_connection(started_cluster): table_name = "test_insert_quorum_with_keeper_loss_" + uuid.uuid4().hex - zero.query( - f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster" - ) + zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") create_query = ( f"CREATE TABLE {table_name} " "(a Int8, d Date) " @@ -394,9 +390,7 @@ def test_insert_quorum_with_keeper_loss_connection(started_cluster): zk = cluster.get_kazoo_client("zoo1") while True: if ( - zk.exists( - f"/clickhouse/tables/{table_name}/replicas/zero/is_active" - ) + zk.exists(f"/clickhouse/tables/{table_name}/replicas/zero/is_active") is None ): break From 27153bfc27d45a9fddddf070bb82c7f1e164b455 Mon Sep 17 00:00:00 2001 From: divanik Date: Tue, 5 Nov 2024 16:58:21 +0000 Subject: [PATCH 443/680] Resolve issues --- tests/integration/test_quorum_inserts/test.py | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index 824cb371595..7adc51121b4 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -1,8 +1,8 @@ import concurrent import time +import uuid import pytest -import uuid from helpers.cluster import ClickHouseCluster from helpers.network import PartitionManager @@ -48,7 +48,6 @@ def started_cluster(): def test_simple_add_replica(started_cluster): table_name = "test_simple_" + uuid.uuid4().hex - zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") create_query = ( f"CREATE TABLE {table_name} " @@ -89,7 +88,6 @@ def test_simple_add_replica(started_cluster): def test_drop_replica_and_achieve_quorum(started_cluster): table_name = "test_drop_replica_and_achieve_quorum_" + uuid.uuid4().hex - zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") create_query = ( f"CREATE TABLE {table_name} " "(a Int8, d Date) " @@ -287,7 +285,6 @@ def test_insert_quorum_with_move_partition(started_cluster, add_new_data): def test_insert_quorum_with_ttl(started_cluster): table_name = "test_insert_quorum_with_ttl_" + uuid.uuid4().hex - zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") create_query = ( f"CREATE TABLE {table_name} " @@ -359,7 +356,6 @@ def test_insert_quorum_with_ttl(started_cluster): def test_insert_quorum_with_keeper_loss_connection(started_cluster): table_name = "test_insert_quorum_with_keeper_loss_" + uuid.uuid4().hex - zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") create_query = ( f"CREATE TABLE {table_name} " "(a Int8, d Date) " From 0687f7a83f1a64abd586c5046dbc5ddda427e00a Mon Sep 17 00:00:00 2001 From: divanik Date: Tue, 5 Nov 2024 17:09:03 +0000 Subject: [PATCH 444/680] Resolve issue --- tests/integration/test_quorum_inserts/test.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index 7adc51121b4..a646319c5f9 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -143,7 +143,7 @@ def test_insert_quorum_with_drop_partition(started_cluster, add_new_data): "test_quorum_insert_with_drop_partition_new_data" if add_new_data else "test_quorum_insert_with_drop_partition" - ) + ) + uuid.uuid4().hex zero.query(f"DROP TABLE IF EXISTS {table_name} ON CLUSTER cluster") create_query = ( @@ -206,12 +206,12 @@ def test_insert_quorum_with_move_partition(started_cluster, add_new_data): "test_insert_quorum_with_move_partition_source_new_data" if add_new_data else "test_insert_quorum_with_move_partition_source" - ) + ) + uuid.uuid4().hex destination_table_name = ( "test_insert_quorum_with_move_partition_destination_new_data" if add_new_data else "test_insert_quorum_with_move_partition_destination" - ) + ) + uuid.uuid4().hex zero.query(f"DROP TABLE IF EXISTS {source_table_name} ON CLUSTER cluster") zero.query(f"DROP TABLE IF EXISTS {destination_table_name} ON CLUSTER cluster") From 76683d021d96309bd3a19d2afde36f9ba802814f Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Tue, 5 Nov 2024 17:22:08 +0000 Subject: [PATCH 445/680] Fix constants in WHERE expression which could apparently contain Join. --- src/Interpreters/ExpressionAnalyzer.cpp | 8 +++++-- ...3258_old_analyzer_const_expr_bug.reference | 0 .../03258_old_analyzer_const_expr_bug.sql | 23 +++++++++++++++++++ 3 files changed, 29 insertions(+), 2 deletions(-) create mode 100644 tests/queries/0_stateless/03258_old_analyzer_const_expr_bug.reference create mode 100644 tests/queries/0_stateless/03258_old_analyzer_const_expr_bug.sql diff --git a/src/Interpreters/ExpressionAnalyzer.cpp b/src/Interpreters/ExpressionAnalyzer.cpp index 4e5cf7d2549..a89e8ca9b3c 100644 --- a/src/Interpreters/ExpressionAnalyzer.cpp +++ b/src/Interpreters/ExpressionAnalyzer.cpp @@ -1981,7 +1981,9 @@ ExpressionAnalysisResult::ExpressionAnalysisResult( Block before_prewhere_sample = source_header; if (sanitizeBlock(before_prewhere_sample)) { - before_prewhere_sample = prewhere_dag_and_flags->dag.updateHeader(before_prewhere_sample); + ExpressionActions( + prewhere_dag_and_flags->dag.clone(), + ExpressionActionsSettings::fromSettings(context->getSettingsRef())).execute(before_prewhere_sample); auto & column_elem = before_prewhere_sample.getByName(query.prewhere()->getColumnName()); /// If the filter column is a constant, record it. if (column_elem.column) @@ -2013,7 +2015,9 @@ ExpressionAnalysisResult::ExpressionAnalysisResult( before_where_sample = source_header; if (sanitizeBlock(before_where_sample)) { - before_where_sample = before_where->dag.updateHeader(before_where_sample); + ExpressionActions( + before_where->dag.clone(), + ExpressionActionsSettings::fromSettings(context->getSettingsRef())).execute(before_where_sample); auto & column_elem = before_where_sample.getByName(query.where()->getColumnName()); diff --git a/tests/queries/0_stateless/03258_old_analyzer_const_expr_bug.reference b/tests/queries/0_stateless/03258_old_analyzer_const_expr_bug.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03258_old_analyzer_const_expr_bug.sql b/tests/queries/0_stateless/03258_old_analyzer_const_expr_bug.sql new file mode 100644 index 00000000000..913de3b849c --- /dev/null +++ b/tests/queries/0_stateless/03258_old_analyzer_const_expr_bug.sql @@ -0,0 +1,23 @@ +WITH + multiIf('-1' = '-1', 10080, '-1' = '7', 60, '-1' = '1', 5, 1440) AS interval_start, -- noqa + multiIf('-1' = '-1', CEIL((today() - toDate('2017-06-22')) / 7)::UInt16, '-1' = '7', 168, '-1' = '1', 288, 90) AS days_run, -- noqa:L045 + block_time as (SELECT arrayJoin( + arrayMap( + i -> toDateTime(toStartOfInterval(now(), INTERVAL interval_start MINUTE) - interval_start * 60 * i, 'UTC'), + range(days_run) + ) + )), + +sales AS ( + SELECT + toDateTime(toStartOfInterval(now(), INTERVAL interval_start MINUTE), 'UTC') AS block_time + FROM + numbers(1) + GROUP BY + block_time + ORDER BY + block_time) + +SELECT + block_time +FROM sales where block_time >= (SELECT MIN(block_time) FROM sales) format Null; From 349010012e7f29ad38b159e99dce7f297f076f63 Mon Sep 17 00:00:00 2001 From: justindeguzman Date: Tue, 5 Nov 2024 09:41:01 -0800 Subject: [PATCH 446/680] [Docs] Add cloud not supported badge for EmbeddedRocksDB engine --- .../engines/table-engines/integrations/embedded-rocksdb.md | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/docs/en/engines/table-engines/integrations/embedded-rocksdb.md b/docs/en/engines/table-engines/integrations/embedded-rocksdb.md index 1958250ed73..41c4e8fc4a9 100644 --- a/docs/en/engines/table-engines/integrations/embedded-rocksdb.md +++ b/docs/en/engines/table-engines/integrations/embedded-rocksdb.md @@ -4,9 +4,13 @@ sidebar_position: 50 sidebar_label: EmbeddedRocksDB --- +import CloudNotSupportedBadge from '@theme/badges/CloudNotSupportedBadge'; + # EmbeddedRocksDB Engine -This engine allows integrating ClickHouse with [rocksdb](http://rocksdb.org/). + + +This engine allows integrating ClickHouse with [RocksDB](http://rocksdb.org/). ## Creating a Table {#creating-a-table} From 27efa296849e1aaa649adb51ef280410169d8018 Mon Sep 17 00:00:00 2001 From: Mikhail Artemenko Date: Tue, 5 Nov 2024 18:04:59 +0000 Subject: [PATCH 447/680] update docs --- .../statements/select/order-by.md | 61 ++++++++++++++++++- 1 file changed, 60 insertions(+), 1 deletion(-) diff --git a/docs/en/sql-reference/statements/select/order-by.md b/docs/en/sql-reference/statements/select/order-by.md index 512a58d7cd9..25d2e7123fd 100644 --- a/docs/en/sql-reference/statements/select/order-by.md +++ b/docs/en/sql-reference/statements/select/order-by.md @@ -291,7 +291,7 @@ All missed values of `expr` column will be filled sequentially and other columns To fill multiple columns, add `WITH FILL` modifier with optional parameters after each field name in `ORDER BY` section. ``` sql -ORDER BY expr [WITH FILL] [FROM const_expr] [TO const_expr] [STEP const_numeric_expr], ... exprN [WITH FILL] [FROM expr] [TO expr] [STEP numeric_expr] +ORDER BY expr [WITH FILL] [FROM const_expr] [TO const_expr] [STEP const_numeric_expr] [STALENESS const_numeric_expr], ... exprN [WITH FILL] [FROM expr] [TO expr] [STEP numeric_expr] [STALENESS numeric_expr] [INTERPOLATE [(col [AS expr], ... colN [AS exprN])]] ``` @@ -300,6 +300,7 @@ When `FROM const_expr` not defined sequence of filling use minimal `expr` field When `TO const_expr` not defined sequence of filling use maximum `expr` field value from `ORDER BY`. When `STEP const_numeric_expr` defined then `const_numeric_expr` interprets `as is` for numeric types, as `days` for Date type, as `seconds` for DateTime type. It also supports [INTERVAL](https://clickhouse.com/docs/en/sql-reference/data-types/special-data-types/interval/) data type representing time and date intervals. When `STEP const_numeric_expr` omitted then sequence of filling use `1.0` for numeric type, `1 day` for Date type and `1 second` for DateTime type. +When `STALENESS const_numeric_expr` is defined, the query will generate rows until the difference from the previous row in the original data exceeds `const_numeric_expr`. `INTERPOLATE` can be applied to columns not participating in `ORDER BY WITH FILL`. Such columns are filled based on previous fields values by applying `expr`. If `expr` is not present will repeat previous value. Omitted list will result in including all allowed columns. Example of a query without `WITH FILL`: @@ -497,6 +498,64 @@ Result: └────────────┴────────────┴──────────┘ ``` +Example of a query without `STALENESS`: + +``` sql +SELECT number as key, 5 * number value, 'original' AS source +FROM numbers(16) WHERE key % 5 == 0 +ORDER BY key WITH FILL; +``` + +Result: + +``` text + ┌─key─┬─value─┬─source───┐ + 1. │ 0 │ 0 │ original │ + 2. │ 1 │ 0 │ │ + 3. │ 2 │ 0 │ │ + 4. │ 3 │ 0 │ │ + 5. │ 4 │ 0 │ │ + 6. │ 5 │ 25 │ original │ + 7. │ 6 │ 0 │ │ + 8. │ 7 │ 0 │ │ + 9. │ 8 │ 0 │ │ +10. │ 9 │ 0 │ │ +11. │ 10 │ 50 │ original │ +12. │ 11 │ 0 │ │ +13. │ 12 │ 0 │ │ +14. │ 13 │ 0 │ │ +15. │ 14 │ 0 │ │ +16. │ 15 │ 75 │ original │ + └─────┴───────┴──────────┘ +``` + +Same query after applying `STALENESS 3`: + +``` sql +SELECT number as key, 5 * number value, 'original' AS source +FROM numbers(16) WHERE key % 5 == 0 +ORDER BY key WITH FILL STALENESS 3; +``` + +Result: + +``` text + ┌─key─┬─value─┬─source───┐ + 1. │ 0 │ 0 │ original │ + 2. │ 1 │ 0 │ │ + 3. │ 2 │ 0 │ │ + 4. │ 5 │ 25 │ original │ + 5. │ 6 │ 0 │ │ + 6. │ 7 │ 0 │ │ + 7. │ 10 │ 50 │ original │ + 8. │ 11 │ 0 │ │ + 9. │ 12 │ 0 │ │ +10. │ 15 │ 75 │ original │ +11. │ 16 │ 0 │ │ +12. │ 17 │ 0 │ │ + └─────┴───────┴──────────┘ +``` + Example of a query without `INTERPOLATE`: ``` sql From 9ec0dda6eeb52c482b4e1e5929b2e03f61672659 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Tue, 5 Nov 2024 20:40:32 +0100 Subject: [PATCH 448/680] Prevent crash in SortCursor with 0 columns --- src/Core/SortCursor.h | 19 +++++++++++---- .../IMergingAlgorithmWithDelayedChunk.cpp | 9 +++++-- .../IMergingAlgorithmWithSharedChunks.cpp | 5 ++-- .../Algorithms/MergingSortedAlgorithm.cpp | 4 ++-- .../Transforms/MergeJoinTransform.cpp | 2 +- .../Transforms/SortingTransform.cpp | 2 +- .../03261_sort_cursor_crash.reference | 4 ++++ .../0_stateless/03261_sort_cursor_crash.sql | 24 +++++++++++++++++++ 8 files changed, 57 insertions(+), 12 deletions(-) create mode 100644 tests/queries/0_stateless/03261_sort_cursor_crash.reference create mode 100644 tests/queries/0_stateless/03261_sort_cursor_crash.sql diff --git a/src/Core/SortCursor.h b/src/Core/SortCursor.h index 3d568be199c..6eb009fa259 100644 --- a/src/Core/SortCursor.h +++ b/src/Core/SortCursor.h @@ -35,6 +35,11 @@ namespace DB { +namespace ErrorCodes +{ +extern const int LOGICAL_ERROR; +} + /** Cursor allows to compare rows in different blocks (and parts). * Cursor moves inside single block. * It is used in priority queue. @@ -83,21 +88,27 @@ struct SortCursorImpl SortCursorImpl( const Block & header, const Columns & columns, + size_t num_rows, const SortDescription & desc_, size_t order_ = 0, IColumn::Permutation * perm = nullptr) : desc(desc_), sort_columns_size(desc.size()), order(order_), need_collation(desc.size()) { - reset(columns, header, perm); + reset(columns, header, num_rows, perm); } bool empty() const { return rows == 0; } /// Set the cursor to the beginning of the new block. - void reset(const Block & block, IColumn::Permutation * perm = nullptr) { reset(block.getColumns(), block, perm); } + void reset(const Block & block, IColumn::Permutation * perm = nullptr) + { + if (block.getColumns().empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Empty column list in block"); + reset(block.getColumns(), block, block.getColumns()[0]->size(), perm); + } /// Set the cursor to the beginning of the new block. - void reset(const Columns & columns, const Block & block, IColumn::Permutation * perm = nullptr) + void reset(const Columns & columns, const Block & block, UInt64 num_rows, IColumn::Permutation * perm = nullptr) { all_columns.clear(); sort_columns.clear(); @@ -125,7 +136,7 @@ struct SortCursorImpl } pos = 0; - rows = all_columns[0]->size(); + rows = num_rows; permutation = perm; } diff --git a/src/Processors/Merges/Algorithms/IMergingAlgorithmWithDelayedChunk.cpp b/src/Processors/Merges/Algorithms/IMergingAlgorithmWithDelayedChunk.cpp index cbad6813fbc..5e271e12943 100644 --- a/src/Processors/Merges/Algorithms/IMergingAlgorithmWithDelayedChunk.cpp +++ b/src/Processors/Merges/Algorithms/IMergingAlgorithmWithDelayedChunk.cpp @@ -24,7 +24,12 @@ void IMergingAlgorithmWithDelayedChunk::initializeQueue(Inputs inputs) continue; cursors[source_num] = SortCursorImpl( - header, current_inputs[source_num].chunk.getColumns(), description, source_num, current_inputs[source_num].permutation); + header, + current_inputs[source_num].chunk.getColumns(), + current_inputs[source_num].chunk.getNumRows(), + description, + source_num, + current_inputs[source_num].permutation); inputs_origin_merge_tree_part_level[source_num] = getPartLevelFromChunk(current_inputs[source_num].chunk); } @@ -41,7 +46,7 @@ void IMergingAlgorithmWithDelayedChunk::updateCursor(Input & input, size_t sourc last_chunk_sort_columns = std::move(cursors[source_num].sort_columns); current_input.swap(input); - cursors[source_num].reset(current_input.chunk.getColumns(), header, current_input.permutation); + cursors[source_num].reset(current_input.chunk.getColumns(), header, current_input.chunk.getNumRows(), current_input.permutation); inputs_origin_merge_tree_part_level[source_num] = getPartLevelFromChunk(current_input.chunk); diff --git a/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.cpp b/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.cpp index 47b7ddf38dc..f99f021286e 100644 --- a/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.cpp +++ b/src/Processors/Merges/Algorithms/IMergingAlgorithmWithSharedChunks.cpp @@ -31,7 +31,8 @@ void IMergingAlgorithmWithSharedChunks::initialize(Inputs inputs) source.skip_last_row = inputs[source_num].skip_last_row; source.chunk = chunk_allocator.alloc(inputs[source_num].chunk); - cursors[source_num] = SortCursorImpl(header, source.chunk->getColumns(), description, source_num, inputs[source_num].permutation); + cursors[source_num] = SortCursorImpl( + header, source.chunk->getColumns(), source.chunk->getNumRows(), description, source_num, inputs[source_num].permutation); source.chunk->all_columns = cursors[source_num].all_columns; source.chunk->sort_columns = cursors[source_num].sort_columns; @@ -49,7 +50,7 @@ void IMergingAlgorithmWithSharedChunks::consume(Input & input, size_t source_num auto & source = sources[source_num]; source.skip_last_row = input.skip_last_row; source.chunk = chunk_allocator.alloc(input.chunk); - cursors[source_num].reset(source.chunk->getColumns(), header, input.permutation); + cursors[source_num].reset(source.chunk->getColumns(), header, source.chunk->getNumRows(), input.permutation); source.chunk->all_columns = cursors[source_num].all_columns; source.chunk->sort_columns = cursors[source_num].sort_columns; diff --git a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp index 3a9cf7ee141..28c6cb473e5 100644 --- a/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp +++ b/src/Processors/Merges/Algorithms/MergingSortedAlgorithm.cpp @@ -59,7 +59,7 @@ void MergingSortedAlgorithm::initialize(Inputs inputs) if (!chunk) continue; - cursors[source_num] = SortCursorImpl(header, chunk.getColumns(), description, source_num); + cursors[source_num] = SortCursorImpl(header, chunk.getColumns(), chunk.getNumRows(), description, source_num); } if (sorting_queue_strategy == SortingQueueStrategy::Default) @@ -84,7 +84,7 @@ void MergingSortedAlgorithm::consume(Input & input, size_t source_num) { removeConstAndSparse(input); current_inputs[source_num].swap(input); - cursors[source_num].reset(current_inputs[source_num].chunk.getColumns(), header); + cursors[source_num].reset(current_inputs[source_num].chunk.getColumns(), header, current_inputs[source_num].chunk.getNumRows()); if (sorting_queue_strategy == SortingQueueStrategy::Default) { diff --git a/src/Processors/Transforms/MergeJoinTransform.cpp b/src/Processors/Transforms/MergeJoinTransform.cpp index 1675e5d0386..77a437d4b97 100644 --- a/src/Processors/Transforms/MergeJoinTransform.cpp +++ b/src/Processors/Transforms/MergeJoinTransform.cpp @@ -394,7 +394,7 @@ void FullMergeJoinCursor::setChunk(Chunk && chunk) convertToFullIfSparse(chunk); current_chunk = std::move(chunk); - cursor = SortCursorImpl(sample_block, current_chunk.getColumns(), desc); + cursor = SortCursorImpl(sample_block, current_chunk.getColumns(), current_chunk.getNumRows(), desc); } bool FullMergeJoinCursor::fullyCompleted() const diff --git a/src/Processors/Transforms/SortingTransform.cpp b/src/Processors/Transforms/SortingTransform.cpp index 6e65093e9e2..6a11354e2bf 100644 --- a/src/Processors/Transforms/SortingTransform.cpp +++ b/src/Processors/Transforms/SortingTransform.cpp @@ -42,7 +42,7 @@ MergeSorter::MergeSorter(const Block & header, Chunks chunks_, SortDescription & /// Convert to full column, because some cursors expect non-contant columns convertToFullIfConst(chunk); - cursors.emplace_back(header, chunk.getColumns(), description, chunk_index); + cursors.emplace_back(header, chunk.getColumns(), chunk.getNumRows(), description, chunk_index); has_collation |= cursors.back().has_collation; nonempty_chunks.emplace_back(std::move(chunk)); diff --git a/tests/queries/0_stateless/03261_sort_cursor_crash.reference b/tests/queries/0_stateless/03261_sort_cursor_crash.reference new file mode 100644 index 00000000000..7299f2f5a5f --- /dev/null +++ b/tests/queries/0_stateless/03261_sort_cursor_crash.reference @@ -0,0 +1,4 @@ +42 +43 +44 +45 diff --git a/tests/queries/0_stateless/03261_sort_cursor_crash.sql b/tests/queries/0_stateless/03261_sort_cursor_crash.sql new file mode 100644 index 00000000000..b659f3d4a92 --- /dev/null +++ b/tests/queries/0_stateless/03261_sort_cursor_crash.sql @@ -0,0 +1,24 @@ +-- https://github.com/ClickHouse/ClickHouse/issues/70779 +-- Crash in SortCursorImpl with the old analyzer, which produces a block with 0 columns and 1 row +DROP TABLE IF EXISTS t0; +DROP TABLE IF EXISTS t1; + +CREATE TABLE t0 (c0 Int) ENGINE = AggregatingMergeTree() ORDER BY tuple(); +INSERT INTO TABLE t0 (c0) VALUES (1); +SELECT 42 FROM t0 FINAL PREWHERE t0.c0 = 1; +DROP TABLE t0; + +CREATE TABLE t0 (c0 Int) ENGINE = SummingMergeTree() ORDER BY tuple(); +INSERT INTO TABLE t0 (c0) VALUES (1); +SELECT 43 FROM t0 FINAL PREWHERE t0.c0 = 1; +DROP TABLE t0; + +CREATE TABLE t0 (c0 Int) ENGINE = ReplacingMergeTree() ORDER BY tuple(); +INSERT INTO TABLE t0 (c0) VALUES (1); +SELECT 44 FROM t0 FINAL PREWHERE t0.c0 = 1; +DROP TABLE t0; + +CREATE TABLE t1 (a0 UInt8, c0 Int32, c1 UInt8) ENGINE = AggregatingMergeTree() ORDER BY tuple(); +INSERT INTO TABLE t1 (a0, c0, c1) VALUES (1, 1, 1); +SELECT 45 FROM t1 FINAL PREWHERE t1.c0 = t1.c1; +DROP TABLE t1; From d7da086a2e474b1938568dbd47f6515344ef397f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Tue, 5 Nov 2024 20:50:05 +0100 Subject: [PATCH 449/680] Fix tidy --- src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp b/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp index 943febf4b0e..2add11d0f6a 100644 --- a/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp +++ b/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp @@ -67,7 +67,7 @@ bool ExecuteScalarSubqueriesMatcher::needChildVisit(ASTPtr & node, const ASTPtr return false; } - if (auto tables = node->as()) + if (auto * tables = node->as()) { /// Contrary to what's said in the code block above, ARRAY JOIN needs to resolve the subquery if possible /// and assign an alias for 02367_optimize_trivial_count_with_array_join to pass. Otherwise it will fail in From 9931b61d6fc0989facbc430d353e611d70d44b5c Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Tue, 5 Nov 2024 20:56:04 +0100 Subject: [PATCH 450/680] fix test --- ...03255_parallel_replicas_join_algo_and_analyzer_4.reference | 4 ++-- .../03255_parallel_replicas_join_algo_and_analyzer_4.sh | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference index 52c4e872f84..d846b26b72b 100644 --- a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference +++ b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.reference @@ -84,7 +84,7 @@ SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` GROUP 500020000 500030000 500040000 -SELECT sum(`__table1`.`item_id`) AS `sum(item_id)` FROM (SELECT `__table2`.`item_id` AS `item_id`, `__table2`.`price_sold` AS `price_sold` FROM `default`.`t` AS `__table2`) AS `__table1` GLOBAL ALL LEFT JOIN `_data_4551627371769371400_3093038500622465792` AS `__table3` ON `__table1`.`item_id` = `__table3`.`item_id` GROUP BY `__table1`.`price_sold` ORDER BY `__table1`.`price_sold` ASC +SELECT sum(`__table1`.`item_id`) AS `sum(item_id)` FROM (SELECT `__table2`.`item_id` AS `item_id`, `__table2`.`price_sold` AS `price_sold` FROM `default`.`t` AS `__table2`) AS `__table1` GLOBAL ALL LEFT JOIN `_data_x_y_` AS `__table3` ON `__table1`.`item_id` = `__table3`.`item_id` GROUP BY `__table1`.`price_sold` ORDER BY `__table1`.`price_sold` ASC 4999950000 4999950000 SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t` AS `__table1` GROUP BY `__table1`.`item_id` @@ -113,4 +113,4 @@ SELECT `__table1`.`item_id` AS `item_id` FROM `default`.`t1` AS `__table1` GROUP 500020000 500030000 500040000 -SELECT sum(`__table1`.`item_id`) AS `sum(item_id)` FROM (SELECT `__table2`.`item_id` AS `item_id`, `__table2`.`price_sold` AS `price_sold` FROM `default`.`t` AS `__table2`) AS `__table1` GLOBAL ALL LEFT JOIN `_data_4551627371769371400_3093038500622465792` AS `__table3` ON `__table1`.`item_id` = `__table3`.`item_id` GROUP BY `__table1`.`price_sold` ORDER BY `__table1`.`price_sold` ASC +SELECT sum(`__table1`.`item_id`) AS `sum(item_id)` FROM (SELECT `__table2`.`item_id` AS `item_id`, `__table2`.`price_sold` AS `price_sold` FROM `default`.`t` AS `__table2`) AS `__table1` GLOBAL ALL LEFT JOIN `_data_x_y_` AS `__table3` ON `__table1`.`item_id` = `__table3`.`item_id` GROUP BY `__table1`.`price_sold` ORDER BY `__table1`.`price_sold` ASC diff --git a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh index 18a2fbd317b..19866f26949 100755 --- a/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh +++ b/tests/queries/0_stateless/03255_parallel_replicas_join_algo_and_analyzer_4.sh @@ -88,7 +88,7 @@ for parallel_replicas_prefer_local_join in 1 0; do --SELECT '----- enable_parallel_replicas=$enable_parallel_replicas prefer_local_plan=$prefer_local_plan parallel_replicas_prefer_local_join=$parallel_replicas_prefer_local_join -----'; ${query}; - SELECT replaceRegexpAll(explain, '.*Query: (.*) Replicas:.*', '\\1') + SELECT replaceRegexpAll(replaceRegexpAll(explain, '.*Query: (.*) Replicas:.*', '\\1'), '(.*)_data_[\d]+_[\d]+(.*)', '\1_data_x_y_\2') FROM ( EXPLAIN actions=1 ${query} From 24c5ef9a052b464671cfb78e887b11237281f53b Mon Sep 17 00:00:00 2001 From: alesapin Date: Tue, 5 Nov 2024 23:08:15 +0100 Subject: [PATCH 451/680] Expose base setting for merge selector --- src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp | 2 ++ src/Storages/MergeTree/MergeTreeSettings.cpp | 1 + 2 files changed, 3 insertions(+) diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp index 62ad9d4a52a..6b9638b11d2 100644 --- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp +++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp @@ -71,6 +71,7 @@ namespace MergeTreeSetting extern const MergeTreeSettingsUInt64 parts_to_throw_insert; extern const MergeTreeSettingsMergeSelectorAlgorithm merge_selector_algorithm; extern const MergeTreeSettingsBool merge_selector_enable_heuristic_to_remove_small_parts_at_right; + extern const MergeTreeSettingsFloat merge_selector_base; } namespace ErrorCodes @@ -542,6 +543,7 @@ SelectPartsDecision MergeTreeDataMergerMutator::selectPartsToMergeFromRanges( simple_merge_settings.window_size = (*data_settings)[MergeTreeSetting::merge_selector_window_size]; simple_merge_settings.max_parts_to_merge_at_once = (*data_settings)[MergeTreeSetting::max_parts_to_merge_at_once]; simple_merge_settings.enable_heuristic_to_remove_small_parts_at_right = (*data_settings)[MergeTreeSetting::merge_selector_enable_heuristic_to_remove_small_parts_at_right]; + simple_merge_settings.base = (*data_settings)[MergeTreeSetting::merge_selector_base]; if (!(*data_settings)[MergeTreeSetting::min_age_to_force_merge_on_partition_only]) simple_merge_settings.min_age_to_force_merge = (*data_settings)[MergeTreeSetting::min_age_to_force_merge_seconds]; diff --git a/src/Storages/MergeTree/MergeTreeSettings.cpp b/src/Storages/MergeTree/MergeTreeSettings.cpp index 883191d59ab..33910d1048d 100644 --- a/src/Storages/MergeTree/MergeTreeSettings.cpp +++ b/src/Storages/MergeTree/MergeTreeSettings.cpp @@ -101,6 +101,7 @@ namespace ErrorCodes DECLARE(Milliseconds, background_task_preferred_step_execution_time_ms, 50, "Target time to execution of one step of merge or mutation. Can be exceeded if one step takes longer time", 0) \ DECLARE(MergeSelectorAlgorithm, merge_selector_algorithm, MergeSelectorAlgorithm::SIMPLE, "The algorithm to select parts for merges assignment", EXPERIMENTAL) \ DECLARE(Bool, merge_selector_enable_heuristic_to_remove_small_parts_at_right, true, "Enable heuristic for selecting parts for merge which removes parts from right side of range, if their size is less than specified ratio (0.01) of sum_size. Works for Simple and StochasticSimple merge selectors", 0) \ + DECLARE(Float, merge_selector_base, 5.0, "Affects write amplification of assigned merges (expert level setting, don't change if you don't understand what it is doing). Works for Simple and StochasticSimple merge selectors", 0) \ \ /** Inserts settings. */ \ DECLARE(UInt64, parts_to_delay_insert, 1000, "If table contains at least that many active parts in single partition, artificially slow down insert into table. Disabled if set to 0", 0) \ From 45bdc4d4deaf6a48ec08f52a9bc8a765730a9b88 Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Wed, 6 Nov 2024 01:12:07 +0000 Subject: [PATCH 452/680] Update tests --- .../02932_refreshable_materialized_views_1.reference | 8 ++++---- .../02932_refreshable_materialized_views_2.reference | 4 ++-- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/tests/queries/0_stateless/02932_refreshable_materialized_views_1.reference b/tests/queries/0_stateless/02932_refreshable_materialized_views_1.reference index 3ec0d3b9ee2..b50ea042e86 100644 --- a/tests/queries/0_stateless/02932_refreshable_materialized_views_1.reference +++ b/tests/queries/0_stateless/02932_refreshable_materialized_views_1.reference @@ -1,14 +1,14 @@ <1: created view> a -CREATE MATERIALIZED VIEW default.a\nREFRESH EVERY 2 SECOND\n(\n `x` UInt64\n)\nENGINE = Memory\nAS SELECT number AS x\nFROM numbers(2)\nUNION ALL\nSELECT rand64() AS x +CREATE MATERIALIZED VIEW default.a\nREFRESH EVERY 2 SECOND\n(\n `x` UInt64\n)\nENGINE = Memory\nDEFINER = default SQL SECURITY DEFINER\nAS SELECT number AS x\nFROM numbers(2)\nUNION ALL\nSELECT rand64() AS x <2: refreshed> 3 1 1 <3: time difference at least> 1000 <4.1: fake clock> Scheduled 2050-01-01 00:00:01 2050-01-01 00:00:02 1 3 3 3 0 <4.5: altered> Scheduled 2050-01-01 00:00:01 2052-01-01 00:00:00 -CREATE MATERIALIZED VIEW default.a\nREFRESH EVERY 2 YEAR\n(\n `x` UInt64\n)\nENGINE = Memory\nAS SELECT x * 2 AS x\nFROM default.src +CREATE MATERIALIZED VIEW default.a\nREFRESH EVERY 2 YEAR\n(\n `x` UInt64\n)\nENGINE = Memory\nDEFINER = default SQL SECURITY DEFINER\nAS SELECT x * 2 AS x\nFROM default.src <5: no refresh> 3 <6: refreshed> 2 <7: refreshed> Scheduled 2052-02-03 04:05:06 2054-01-01 00:00:00 -CREATE MATERIALIZED VIEW default.b\nREFRESH EVERY 2 YEAR DEPENDS ON default.a\n(\n `y` Int32\n)\nENGINE = MergeTree\nORDER BY y\nSETTINGS index_granularity = 8192\nAS SELECT x * 10 AS y\nFROM default.a +CREATE MATERIALIZED VIEW default.b\nREFRESH EVERY 2 YEAR DEPENDS ON default.a\n(\n `y` Int32\n)\nENGINE = MergeTree\nORDER BY y\nSETTINGS index_granularity = 8192\nDEFINER = default SQL SECURITY DEFINER\nAS SELECT x * 10 AS y\nFROM default.a <7.5: created dependent> 2052-11-11 11:11:11 <8: refreshed> 20 <9: refreshed> a Scheduled 2054-01-01 00:00:00 @@ -26,4 +26,4 @@ CREATE MATERIALIZED VIEW default.b\nREFRESH EVERY 2 YEAR DEPENDS ON default.a\n( <17: chain-refreshed> a Scheduled 2062-01-01 00:00:00 <17: chain-refreshed> b Scheduled 2062-01-01 00:00:00 <18: removed dependency> b Scheduled 2062-03-03 03:03:03 2062-03-03 03:03:03 2064-01-01 00:00:00 -CREATE MATERIALIZED VIEW default.b\nREFRESH EVERY 2 YEAR\n(\n `y` Int32\n)\nENGINE = MergeTree\nORDER BY y\nSETTINGS index_granularity = 8192\nAS SELECT x * 10 AS y\nFROM default.a +CREATE MATERIALIZED VIEW default.b\nREFRESH EVERY 2 YEAR\n(\n `y` Int32\n)\nENGINE = MergeTree\nORDER BY y\nSETTINGS index_granularity = 8192\nDEFINER = default SQL SECURITY DEFINER\nAS SELECT x * 10 AS y\nFROM default.a diff --git a/tests/queries/0_stateless/02932_refreshable_materialized_views_2.reference b/tests/queries/0_stateless/02932_refreshable_materialized_views_2.reference index 3eeab4f574e..8dcc3d55603 100644 --- a/tests/queries/0_stateless/02932_refreshable_materialized_views_2.reference +++ b/tests/queries/0_stateless/02932_refreshable_materialized_views_2.reference @@ -7,9 +7,9 @@ <25: rename during refresh> f Running <27: cancelled> f Scheduled cancelled <28: drop during refresh> 0 0 -CREATE MATERIALIZED VIEW default.g\nREFRESH EVERY 1 WEEK OFFSET 3 DAY 4 HOUR RANDOMIZE FOR 4 DAY 1 HOUR\n(\n `x` Int64\n)\nENGINE = Memory\nAS SELECT 42 AS x +CREATE MATERIALIZED VIEW default.g\nREFRESH EVERY 1 WEEK OFFSET 3 DAY 4 HOUR RANDOMIZE FOR 4 DAY 1 HOUR\n(\n `x` Int64\n)\nENGINE = Memory\nDEFINER = default SQL SECURITY DEFINER\nAS SELECT 42 AS x <29: randomize> 1 1 -CREATE MATERIALIZED VIEW default.h\nREFRESH EVERY 1 SECOND TO default.dest\n(\n `x` Int64\n)\nAS SELECT x * 10 AS x\nFROM default.src +CREATE MATERIALIZED VIEW default.h\nREFRESH EVERY 1 SECOND TO default.dest\n(\n `x` Int64\n)\nDEFINER = default SQL SECURITY DEFINER\nAS SELECT x * 10 AS x\nFROM default.src <30: to existing table> 10 <31: to existing table> 10 <31: to existing table> 20 From c23dfa343155a7162b1bcf1f98080f5a08b92f7f Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Wed, 6 Nov 2024 12:30:37 +0800 Subject: [PATCH 453/680] fix uninitialized orc data --- .../Impl/NativeORCBlockInputFormat.cpp | 25 +++++++++++++------ 1 file changed, 17 insertions(+), 8 deletions(-) diff --git a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp index 81df330ffb5..5c7637d3e51 100644 --- a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp @@ -1534,15 +1534,24 @@ static ColumnWithTypeAndName readColumnWithDateData( for (size_t i = 0; i < orc_int_column->numElements; ++i) { - Int32 days_num = static_cast(orc_int_column->data[i]); - if (check_date_range && (days_num > DATE_LUT_MAX_EXTEND_DAY_NUM || days_num < -DAYNUM_OFFSET_EPOCH)) - throw Exception( - ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, - "Input value {} of a column \"{}\" exceeds the range of type Date32", - days_num, - column_name); + if (!orc_int_column->hasNulls || orc_int_column->notNull[i]) + { + Int32 days_num = static_cast(orc_int_column->data[i]); + if (check_date_range && (days_num > DATE_LUT_MAX_EXTEND_DAY_NUM || days_num < -DAYNUM_OFFSET_EPOCH)) + throw Exception( + ErrorCodes::VALUE_IS_OUT_OF_RANGE_OF_DATA_TYPE, + "Input value {} of a column \"{}\" exceeds the range of type Date32", + days_num, + column_name); + + column_data.push_back(days_num); + } + else + { + /// ORC library doesn't gurantee that orc_int_column->data[i] is initialized to zero when orc_int_column->notNull[i] is false since https://github.com/ClickHouse/ClickHouse/pull/69473 + column_data.push_back(0); + } - column_data.push_back(days_num); } return {std::move(internal_column), internal_type, column_name}; From 6a8df5ea89724d7686f6c520bc436b7cb80294bd Mon Sep 17 00:00:00 2001 From: nauu Date: Wed, 6 Nov 2024 14:57:14 +0800 Subject: [PATCH 454/680] support the endpoint of oss accelerator --- src/IO/S3/URI.cpp | 12 ++++++++++-- src/IO/tests/gtest_s3_uri.cpp | 16 ++++++++++++++++ 2 files changed, 26 insertions(+), 2 deletions(-) diff --git a/src/IO/S3/URI.cpp b/src/IO/S3/URI.cpp index 7c6a21941eb..ad746ff3326 100644 --- a/src/IO/S3/URI.cpp +++ b/src/IO/S3/URI.cpp @@ -37,7 +37,7 @@ URI::URI(const std::string & uri_, bool allow_archive_path_syntax) /// Case when bucket name represented in domain name of S3 URL. /// E.g. (https://bucket-name.s3.region.amazonaws.com/key) /// https://docs.aws.amazon.com/AmazonS3/latest/dev/VirtualHosting.html#virtual-hosted-style-access - static const RE2 virtual_hosted_style_pattern(R"((.+)\.(s3express[\-a-z0-9]+|s3|cos|obs|oss|eos)([.\-][a-z0-9\-.:]+))"); + static const RE2 virtual_hosted_style_pattern(R"((.+)\.(s3express[\-a-z0-9]+|s3|cos|obs|oss-data-acc|oss|eos)([.\-][a-z0-9\-.:]+))"); /// Case when AWS Private Link Interface is being used /// E.g. (bucket.vpce-07a1cd78f1bd55c5f-j3a3vg6w.s3.us-east-1.vpce.amazonaws.com/bucket-name/key) @@ -115,7 +115,15 @@ URI::URI(const std::string & uri_, bool allow_archive_path_syntax) && re2::RE2::FullMatch(uri.getAuthority(), virtual_hosted_style_pattern, &bucket, &name, &endpoint_authority_from_uri)) { is_virtual_hosted_style = true; - endpoint = uri.getScheme() + "://" + name + endpoint_authority_from_uri; + if (name == "oss-data-acc") + { + bucket = bucket.substr(0, bucket.find(".")); + endpoint = uri.getScheme() + "://" + uri.getHost().substr(bucket.length() + 1); + } + else + { + endpoint = uri.getScheme() + "://" + name + endpoint_authority_from_uri; + } validateBucket(bucket, uri); if (!uri.getPath().empty()) diff --git a/src/IO/tests/gtest_s3_uri.cpp b/src/IO/tests/gtest_s3_uri.cpp index 8696fab0616..6167313b634 100644 --- a/src/IO/tests/gtest_s3_uri.cpp +++ b/src/IO/tests/gtest_s3_uri.cpp @@ -212,6 +212,22 @@ TEST(S3UriTest, validPatterns) ASSERT_EQ("", uri.version_id); ASSERT_EQ(true, uri.is_virtual_hosted_style); } + { + S3::URI uri("https://bucket-test1.oss-cn-beijing-internal.aliyuncs.com/ab-test"); + ASSERT_EQ("https://oss-cn-beijing-internal.aliyuncs.com", uri.endpoint); + ASSERT_EQ("bucket-test1", uri.bucket); + ASSERT_EQ("ab-test", uri.key); + ASSERT_EQ("", uri.version_id); + ASSERT_EQ(true, uri.is_virtual_hosted_style); + } + { + S3::URI uri("https://bucket-test.cn-beijing-internal.oss-data-acc.aliyuncs.com/ab-test"); + ASSERT_EQ("https://cn-beijing-internal.oss-data-acc.aliyuncs.com", uri.endpoint); + ASSERT_EQ("bucket-test", uri.bucket); + ASSERT_EQ("ab-test", uri.key); + ASSERT_EQ("", uri.version_id); + ASSERT_EQ(true, uri.is_virtual_hosted_style); + } } TEST(S3UriTest, versionIdChecks) From 127f324822e7b45259eb6ec9b9f5168933350aa1 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Wed, 6 Nov 2024 15:03:41 +0800 Subject: [PATCH 455/680] add uts --- .../Formats/Impl/NativeORCBlockInputFormat.cpp | 3 +-- .../03259_orc_date_out_of_range.reference | 12 ++++++++++++ .../0_stateless/03259_orc_date_out_of_range.sql | 15 +++++++++++++++ 3 files changed, 28 insertions(+), 2 deletions(-) create mode 100644 tests/queries/0_stateless/03259_orc_date_out_of_range.reference create mode 100644 tests/queries/0_stateless/03259_orc_date_out_of_range.sql diff --git a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp index 5c7637d3e51..26aa3555c2b 100644 --- a/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp +++ b/src/Processors/Formats/Impl/NativeORCBlockInputFormat.cpp @@ -1548,10 +1548,9 @@ static ColumnWithTypeAndName readColumnWithDateData( } else { - /// ORC library doesn't gurantee that orc_int_column->data[i] is initialized to zero when orc_int_column->notNull[i] is false since https://github.com/ClickHouse/ClickHouse/pull/69473 + /// ORC library doesn't guarantee that orc_int_column->data[i] is initialized to zero when orc_int_column->notNull[i] is false since https://github.com/ClickHouse/ClickHouse/pull/69473 column_data.push_back(0); } - } return {std::move(internal_column), internal_type, column_name}; diff --git a/tests/queries/0_stateless/03259_orc_date_out_of_range.reference b/tests/queries/0_stateless/03259_orc_date_out_of_range.reference new file mode 100644 index 00000000000..ddac785369f --- /dev/null +++ b/tests/queries/0_stateless/03259_orc_date_out_of_range.reference @@ -0,0 +1,12 @@ +number Nullable(Int64) +date_field Nullable(Date32) +\N +1970-01-02 +\N +1970-01-04 +\N +1970-01-06 +\N +1970-01-08 +\N +1970-01-10 diff --git a/tests/queries/0_stateless/03259_orc_date_out_of_range.sql b/tests/queries/0_stateless/03259_orc_date_out_of_range.sql new file mode 100644 index 00000000000..470c4ff3817 --- /dev/null +++ b/tests/queries/0_stateless/03259_orc_date_out_of_range.sql @@ -0,0 +1,15 @@ + +-- Tags: no-parallel + +SET session_timezone = 'UTC'; +SET engine_file_truncate_on_insert = 1; + +insert into function file('03259.orc') +select + number, + if (number % 2 = 0, null, toDate32(number)) as date_field + from numbers(10); + +desc file('03259.orc'); + +select date_field from file('03259.orc') order by number; From ef0be4a01cb4fd9c4723ecf31b96aab7ee6a30ac Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Wed, 6 Nov 2024 15:06:00 +0800 Subject: [PATCH 456/680] fix typo --- tests/queries/0_stateless/03259_orc_date_out_of_range.sql | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/queries/0_stateless/03259_orc_date_out_of_range.sql b/tests/queries/0_stateless/03259_orc_date_out_of_range.sql index 470c4ff3817..409e8ce079d 100644 --- a/tests/queries/0_stateless/03259_orc_date_out_of_range.sql +++ b/tests/queries/0_stateless/03259_orc_date_out_of_range.sql @@ -1,4 +1,3 @@ - -- Tags: no-parallel SET session_timezone = 'UTC'; From 0c1aa03cb172ca666b7054863626d563e1de21e7 Mon Sep 17 00:00:00 2001 From: justindeguzman Date: Wed, 6 Nov 2024 00:05:55 -0800 Subject: [PATCH 457/680] [Docs] Update note about Prometheus integration and ClickHouse Cloud --- docs/en/interfaces/prometheus.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/interfaces/prometheus.md b/docs/en/interfaces/prometheus.md index 8e7023cc51f..11f503b54d7 100644 --- a/docs/en/interfaces/prometheus.md +++ b/docs/en/interfaces/prometheus.md @@ -9,7 +9,7 @@ sidebar_label: Prometheus protocols ## Exposing metrics {#expose} :::note -ClickHouse Cloud does not currently support connecting to Prometheus. To be notified when this feature is supported, please contact support@clickhouse.com. +If you are using ClickHouse Cloud, you can expose metrics to Prometheus using the [Prometheus Integration](/en/integrations/prometheus). ::: ClickHouse can expose its own metrics for scraping from Prometheus: From 590029a33bfd844eede8b4ad570464d0cf86c938 Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Wed, 6 Nov 2024 16:38:09 +0800 Subject: [PATCH 458/680] fix orc date32 overflow --- tests/queries/0_stateless/03259_orc_date_out_of_range.sql | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/queries/0_stateless/03259_orc_date_out_of_range.sql b/tests/queries/0_stateless/03259_orc_date_out_of_range.sql index 409e8ce079d..7103b93b147 100644 --- a/tests/queries/0_stateless/03259_orc_date_out_of_range.sql +++ b/tests/queries/0_stateless/03259_orc_date_out_of_range.sql @@ -3,12 +3,12 @@ SET session_timezone = 'UTC'; SET engine_file_truncate_on_insert = 1; -insert into function file('03259.orc') +insert into function file('03259.orc', 'ORC') select number, if (number % 2 = 0, null, toDate32(number)) as date_field - from numbers(10); +from numbers(10); -desc file('03259.orc'); +desc file('03259.orc', 'ORC'); -select date_field from file('03259.orc') order by number; +select date_field from file('03259.orc', 'ORC') order by number; From 4f8099d7aa6d1dff2ad79fc020810fe36a3cfd3b Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Wed, 6 Nov 2024 08:51:44 +0000 Subject: [PATCH 459/680] Simplify the code --- .../MergeTreeIndexVectorSimilarity.cpp | 81 +++++++++---------- .../0_stateless/02354_vector_search_bugs.sql | 2 +- 2 files changed, 40 insertions(+), 43 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp index 498d0131d5a..e55010ac9ec 100644 --- a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp @@ -345,60 +345,57 @@ void MergeTreeIndexAggregatorVectorSimilarity::update(const Block & block, size_ throw Exception(ErrorCodes::INCORRECT_DATA, "Index granularity is too big: more than {} rows per index granule.", std::numeric_limits::max()); if (index_sample_block.columns() > 1) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected block with single column"); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected that index is build over a single column"); - for (size_t i = 0; i < index_sample_block.columns(); ++i) - { - const auto & index_column_with_type_and_name = index_sample_block.getByPosition(i); + const auto & index_column_with_type_and_name = index_sample_block.getByPosition(0); - const auto & index_column_name = index_column_with_type_and_name.name; - const auto & index_column = block.getByName(index_column_name).column; - ColumnPtr column_cut = index_column->cut(*pos, rows_read); + const auto & index_column_name = index_column_with_type_and_name.name; + const auto & index_column = block.getByName(index_column_name).column; + ColumnPtr column_cut = index_column->cut(*pos, rows_read); - const auto * column_array = typeid_cast(column_cut.get()); - if (!column_array) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected Array(Float*) column"); + const auto * column_array = typeid_cast(column_cut.get()); + if (!column_array) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected Array(Float*) column"); - if (column_array->empty()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Array is unexpectedly empty"); + if (column_array->empty()) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Array is unexpectedly empty"); - /// The vector similarity algorithm naturally assumes that the indexed vectors have dimension >= 1. This condition is violated if empty arrays - /// are INSERTed into an vector-similarity-indexed column or if no value was specified at all in which case the arrays take on their default - /// values which is also empty. - if (column_array->isDefaultAt(0)) - throw Exception(ErrorCodes::INCORRECT_DATA, "The arrays in column '{}' must not be empty. Did you try to INSERT default values?", index_column_name); + /// The vector similarity algorithm naturally assumes that the indexed vectors have dimension >= 1. This condition is violated if empty arrays + /// are INSERTed into an vector-similarity-indexed column or if no value was specified at all in which case the arrays take on their default + /// values which is also empty. + if (column_array->isDefaultAt(0)) + throw Exception(ErrorCodes::INCORRECT_DATA, "The arrays in column '{}' must not be empty. Did you try to INSERT default values?", index_column_name); - const size_t rows = column_array->size(); + const size_t rows = column_array->size(); - const auto & column_array_offsets = column_array->getOffsets(); - const size_t dimensions = column_array_offsets[0]; + const auto & column_array_offsets = column_array->getOffsets(); + const size_t dimensions = column_array_offsets[0]; - if (!index) - index = std::make_shared(dimensions, metric_kind, scalar_kind, usearch_hnsw_params); + if (!index) + index = std::make_shared(dimensions, metric_kind, scalar_kind, usearch_hnsw_params); - /// Also check that previously inserted blocks have the same size as this block. - /// Note that this guarantees consistency of dimension only within parts. We are unable to detect inconsistent dimensions across - /// parts - for this, a little help from the user is needed, e.g. CONSTRAINT cnstr CHECK length(array) = 42. - if (index->dimensions() != dimensions) - throw Exception(ErrorCodes::INCORRECT_DATA, "All arrays in column with vector similarity index must have equal length"); + /// Also check that previously inserted blocks have the same size as this block. + /// Note that this guarantees consistency of dimension only within parts. We are unable to detect inconsistent dimensions across + /// parts - for this, a little help from the user is needed, e.g. CONSTRAINT cnstr CHECK length(array) = 42. + if (index->dimensions() != dimensions) + throw Exception(ErrorCodes::INCORRECT_DATA, "All arrays in column with vector similarity index must have equal length"); - /// We use Usearch's index_dense_t as index type which supports only 4 bio entries according to https://github.com/unum-cloud/usearch/tree/main/cpp - if (index->size() + rows > std::numeric_limits::max()) - throw Exception(ErrorCodes::INCORRECT_DATA, "Size of vector similarity index would exceed 4 billion entries"); + /// We use Usearch's index_dense_t as index type which supports only 4 bio entries according to https://github.com/unum-cloud/usearch/tree/main/cpp + if (index->size() + rows > std::numeric_limits::max()) + throw Exception(ErrorCodes::INCORRECT_DATA, "Size of vector similarity index would exceed 4 billion entries"); - DataTypePtr data_type = index_column_with_type_and_name.type; - const auto * data_type_array = typeid_cast(data_type.get()); - if (!data_type_array) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected data type Array(Float*)"); - const TypeIndex nested_type_index = data_type_array->getNestedType()->getTypeId(); + DataTypePtr data_type = index_column_with_type_and_name.type; + const auto * data_type_array = typeid_cast(data_type.get()); + if (!data_type_array) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected data type Array(Float*)"); + const TypeIndex nested_type_index = data_type_array->getNestedType()->getTypeId(); - if (WhichDataType(nested_type_index).isFloat32()) - updateImpl(column_array, column_array_offsets, index, dimensions, rows); - else if (WhichDataType(nested_type_index).isFloat64()) - updateImpl(column_array, column_array_offsets, index, dimensions, rows); - else - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected data type Array(Float*)"); - } + if (WhichDataType(nested_type_index).isFloat32()) + updateImpl(column_array, column_array_offsets, index, dimensions, rows); + else if (WhichDataType(nested_type_index).isFloat64()) + updateImpl(column_array, column_array_offsets, index, dimensions, rows); + else + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected data type Array(Float*)"); *pos += rows_read; diff --git a/tests/queries/0_stateless/02354_vector_search_bugs.sql b/tests/queries/0_stateless/02354_vector_search_bugs.sql index 6bcb0f78e75..276d4eb5b59 100644 --- a/tests/queries/0_stateless/02354_vector_search_bugs.sql +++ b/tests/queries/0_stateless/02354_vector_search_bugs.sql @@ -124,7 +124,7 @@ CREATE TABLE tab( val String, vec Array(Float32), INDEX ann_idx vec TYPE vector_similarity('hnsw', 'cosineDistance'), - INDEX set_idx val TYPE set(100) GRANULARITY 100 + INDEX set_idx val TYPE set(100) ) ENGINE = MergeTree() ORDER BY tuple(); From 6761fccbf30cba1b18331bab993710e89c047aba Mon Sep 17 00:00:00 2001 From: taiyang-li <654010905@qq.com> Date: Wed, 6 Nov 2024 17:10:00 +0800 Subject: [PATCH 460/680] fix orc date32 overflow --- tests/queries/0_stateless/03259_orc_date_out_of_range.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03259_orc_date_out_of_range.sql b/tests/queries/0_stateless/03259_orc_date_out_of_range.sql index 7103b93b147..e73d2faa5dd 100644 --- a/tests/queries/0_stateless/03259_orc_date_out_of_range.sql +++ b/tests/queries/0_stateless/03259_orc_date_out_of_range.sql @@ -1,4 +1,4 @@ --- Tags: no-parallel +-- Tags: no-fasttest, no-parallel SET session_timezone = 'UTC'; SET engine_file_truncate_on_insert = 1; From 918ad5c4d54c27b6c14e1221ae56a40dd937e2cc Mon Sep 17 00:00:00 2001 From: Ilya Golshtein Date: Wed, 6 Nov 2024 09:42:35 +0000 Subject: [PATCH 461/680] fix_test_drop_complex_columns: tests passed --- .../test.py | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/tests/integration/test_replicated_s3_zero_copy_drop_partition/test.py b/tests/integration/test_replicated_s3_zero_copy_drop_partition/test.py index 6d2bb0a3b70..9937c0ed4ea 100644 --- a/tests/integration/test_replicated_s3_zero_copy_drop_partition/test.py +++ b/tests/integration/test_replicated_s3_zero_copy_drop_partition/test.py @@ -68,9 +68,19 @@ CREATE TABLE test_s3(c1 Int8, c2 Date) ENGINE = ReplicatedMergeTree('/test/table def test_drop_complex_columns(started_cluster): + node1 = cluster.instances["node1"] + node1.query( + """ +CREATE TABLE warming_up( +id Int8 +) ENGINE = MergeTree +order by (id) SETTINGS storage_policy = 's3';""" + ) + + # Now we are sure that s3 storage is up and running start_objects = get_objects_in_data_path() print("Objects before", start_objects) - node1 = cluster.instances["node1"] + node1.query( """ CREATE TABLE test_s3_complex_types( @@ -104,3 +114,4 @@ vertical_merge_algorithm_min_columns_to_activate=1;""" end_objects = get_objects_in_data_path() print("Objects after drop", end_objects) assert start_objects == end_objects + node1.query("DROP TABLE warming_up SYNC") From b38dc1d8ca791c6fc686ae9d8efedeb77e354de2 Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Wed, 6 Nov 2024 11:05:43 +0100 Subject: [PATCH 462/680] Update FileCache.cpp --- src/Interpreters/Cache/FileCache.cpp | 2 -- 1 file changed, 2 deletions(-) diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index ae3c9c58fc5..f7b7ffc5aea 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -1438,8 +1438,6 @@ void FileCache::loadMetadataForKeys(const fs::path & keys_dir) "cached file `{}` does not fit in cache anymore (size: {})", size_limit, offset_it->path().string(), size); - chassert(false); /// TODO: remove before merge. - fs::remove(offset_it->path()); } } From 068b4fe8cfa184c4aaecda057b78d7b8acfdbb06 Mon Sep 17 00:00:00 2001 From: maxvostrikov Date: Wed, 6 Nov 2024 12:16:59 +0100 Subject: [PATCH 463/680] squash! Missing tests in several tests in 24.10 Added corner cases for tests for: to_utc_timestamp and from_utc_timestamp (more timezones, spetial timezones, epoch corners does not look right, raising a bug over that) arrayUnion (empty and big arrays) quantilesExactWeightedInterpolated (more data types) --- tests/queries/0_stateless/02812_from_to_utc_timestamp.reference | 2 ++ tests/queries/0_stateless/02812_from_to_utc_timestamp.sh | 2 ++ 2 files changed, 4 insertions(+) diff --git a/tests/queries/0_stateless/02812_from_to_utc_timestamp.reference b/tests/queries/0_stateless/02812_from_to_utc_timestamp.reference index bdce849e069..fb92bdda821 100644 --- a/tests/queries/0_stateless/02812_from_to_utc_timestamp.reference +++ b/tests/queries/0_stateless/02812_from_to_utc_timestamp.reference @@ -6,5 +6,7 @@ 2024-10-24 16:22:33 2024-10-24 06:22:33 leap year: 2024-02-29 16:22:33 2024-02-29 06:22:33 non-leap year: 2023-03-01 16:22:33 2023-03-01 06:22:33 +leap year: 2024-02-29 04:22:33 2024-02-29 19:22:33 +non-leap year: 2023-03-01 04:22:33 2023-02-28 19:22:33 timezone with half-hour offset: 2024-02-29 00:52:33 2024-02-29 21:52:33 jump over a year: 2024-01-01 04:01:01 2023-12-31 20:01:01 diff --git a/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh b/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh index 441fc254256..20ae224332c 100755 --- a/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh +++ b/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh @@ -18,6 +18,8 @@ $CLICKHOUSE_CLIENT -q "select to_utc_timestamp(toDateTime('2024-10-24 11:22:33') $CLICKHOUSE_CLIENT -q "select to_utc_timestamp(toDateTime('2024-10-24 11:22:33'), 'EST'), from_utc_timestamp(toDateTime('2024-10-24 11:22:33'), 'EST')" $CLICKHOUSE_CLIENT -q "select 'leap year:', to_utc_timestamp(toDateTime('2024-02-29 11:22:33'), 'EST'), from_utc_timestamp(toDateTime('2024-02-29 11:22:33'), 'EST')" $CLICKHOUSE_CLIENT -q "select 'non-leap year:', to_utc_timestamp(toDateTime('2023-02-29 11:22:33'), 'EST'), from_utc_timestamp(toDateTime('2023-02-29 11:22:33'), 'EST')" +$CLICKHOUSE_CLIENT -q "select 'leap year:', to_utc_timestamp(toDateTime('2024-02-28 23:22:33'), 'EST'), from_utc_timestamp(toDateTime('2024-03-01 00:22:33'), 'EST')" +$CLICKHOUSE_CLIENT -q "select 'non-leap year:', to_utc_timestamp(toDateTime('2023-02-28 23:22:33'), 'EST'), from_utc_timestamp(toDateTime('2023-03-01 00:22:33'), 'EST')" $CLICKHOUSE_CLIENT -q "select 'timezone with half-hour offset:', to_utc_timestamp(toDateTime('2024-02-29 11:22:33'), 'Australia/Adelaide'), from_utc_timestamp(toDateTime('2024-02-29 11:22:33'), 'Australia/Adelaide')" $CLICKHOUSE_CLIENT -q "select 'jump over a year:', to_utc_timestamp(toDateTime('2023-12-31 23:01:01'), 'EST'), from_utc_timestamp(toDateTime('2024-01-01 01:01:01'), 'EST')" From f0bb69f12667108659b5ed9803f4b290c7faafee Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Wed, 6 Nov 2024 11:46:49 +0000 Subject: [PATCH 464/680] Simplify more --- src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp index e55010ac9ec..f95b840e223 100644 --- a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp @@ -347,9 +347,8 @@ void MergeTreeIndexAggregatorVectorSimilarity::update(const Block & block, size_ if (index_sample_block.columns() > 1) throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected that index is build over a single column"); - const auto & index_column_with_type_and_name = index_sample_block.getByPosition(0); + const auto & index_column_name = index_sample_block.getByPosition(0).name; - const auto & index_column_name = index_column_with_type_and_name.name; const auto & index_column = block.getByName(index_column_name).column; ColumnPtr column_cut = index_column->cut(*pos, rows_read); @@ -384,8 +383,7 @@ void MergeTreeIndexAggregatorVectorSimilarity::update(const Block & block, size_ if (index->size() + rows > std::numeric_limits::max()) throw Exception(ErrorCodes::INCORRECT_DATA, "Size of vector similarity index would exceed 4 billion entries"); - DataTypePtr data_type = index_column_with_type_and_name.type; - const auto * data_type_array = typeid_cast(data_type.get()); + const auto * data_type_array = typeid_cast(block.getByName(index_column_name).type.get()); if (!data_type_array) throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected data type Array(Float*)"); const TypeIndex nested_type_index = data_type_array->getNestedType()->getTypeId(); From 7c6472a09034715bbeb8374667203076c3458e82 Mon Sep 17 00:00:00 2001 From: Joe Lynch Date: Wed, 6 Nov 2024 13:34:39 +0100 Subject: [PATCH 465/680] Fix documentation for system.grants.is_partial_revoke --- docs/en/operations/system-tables/grants.md | 4 ++-- src/Storages/System/StorageSystemGrants.cpp | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/docs/en/operations/system-tables/grants.md b/docs/en/operations/system-tables/grants.md index 262a53a87a5..debc3146008 100644 --- a/docs/en/operations/system-tables/grants.md +++ b/docs/en/operations/system-tables/grants.md @@ -19,7 +19,7 @@ Columns: - `column` ([Nullable](../../sql-reference/data-types/nullable.md)([String](../../sql-reference/data-types/string.md))) — Name of a column to which access is granted. - `is_partial_revoke` ([UInt8](../../sql-reference/data-types/int-uint.md#uint-ranges)) — Logical value. It shows whether some privileges have been revoked. Possible values: -- `0` — The row describes a partial revoke. -- `1` — The row describes a grant. +- `0` — The row describes a grant. +- `1` — The row describes a partial revoke. - `grant_option` ([UInt8](../../sql-reference/data-types/int-uint.md#uint-ranges)) — Permission is granted `WITH GRANT OPTION`, see [GRANT](../../sql-reference/statements/grant.md#granting-privilege-syntax). diff --git a/src/Storages/System/StorageSystemGrants.cpp b/src/Storages/System/StorageSystemGrants.cpp index 5de1f8cef55..aa010e44388 100644 --- a/src/Storages/System/StorageSystemGrants.cpp +++ b/src/Storages/System/StorageSystemGrants.cpp @@ -30,8 +30,8 @@ ColumnsDescription StorageSystemGrants::getColumnsDescription() {"column", std::make_shared(std::make_shared()), "Name of a column to which access is granted."}, {"is_partial_revoke", std::make_shared(), "Logical value. It shows whether some privileges have been revoked. Possible values: " - "0 — The row describes a partial revoke, " - "1 — The row describes a grant." + "0 — The row describes a grant, " + "1 — The row describes a partial revoke." }, {"grant_option", std::make_shared(), "Permission is granted WITH GRANT OPTION."}, }; From 9ee22533a067fc235aea65ff7b89c801b112b918 Mon Sep 17 00:00:00 2001 From: Pablo Marcos Date: Wed, 6 Nov 2024 13:46:30 +0100 Subject: [PATCH 466/680] Move bitShift function changelog entries to backward incompatible Move bitShift function changelog entries to backward incompatible --- CHANGELOG.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index 90285582b4e..dacee73440f 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -488,6 +488,7 @@ * Remove `is_deterministic` field from the `system.functions` table. [#66630](https://github.com/ClickHouse/ClickHouse/pull/66630) ([Alexey Milovidov](https://github.com/alexey-milovidov)). * Function `tuple` will now try to construct named tuples in query (controlled by `enable_named_columns_in_function_tuple`). Introduce function `tupleNames` to extract names from tuples. [#54881](https://github.com/ClickHouse/ClickHouse/pull/54881) ([Amos Bird](https://github.com/amosbird)). * Change how deduplication for Materialized Views works. Fixed a lot of cases like: - on destination table: data is split for 2 or more blocks and that blocks is considered as duplicate when that block is inserted in parallel. - on MV destination table: the equal blocks are deduplicated, that happens when MV often produces equal data as a result for different input data due to performing aggregation. - on MV destination table: the equal blocks which comes from different MV are deduplicated. [#61601](https://github.com/ClickHouse/ClickHouse/pull/61601) ([Sema Checherinda](https://github.com/CheSema)). +* Functions `bitShiftLeft` and `bitShitfRight` return an error for out of bounds shift positions [#65838](https://github.com/ClickHouse/ClickHouse/pull/65838) ([Pablo Marcos](https://github.com/pamarcos)). #### New Feature * Add `ASOF JOIN` support for `full_sorting_join` algorithm. [#55051](https://github.com/ClickHouse/ClickHouse/pull/55051) ([vdimir](https://github.com/vdimir)). @@ -599,7 +600,6 @@ * Functions `bitTest`, `bitTestAll`, and `bitTestAny` now return an error if the specified bit index is out-of-bounds [#65818](https://github.com/ClickHouse/ClickHouse/pull/65818) ([Pablo Marcos](https://github.com/pamarcos)). * Setting `join_any_take_last_row` is supported in any query with hash join. [#65820](https://github.com/ClickHouse/ClickHouse/pull/65820) ([vdimir](https://github.com/vdimir)). * 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)). -* Functions `bitShiftLeft` and `bitShitfRight` return an error for out of bounds shift positions [#65838](https://github.com/ClickHouse/ClickHouse/pull/65838) ([Pablo Marcos](https://github.com/pamarcos)). * Fix growing memory usage in S3Queue. [#65839](https://github.com/ClickHouse/ClickHouse/pull/65839) ([Kseniia Sumarokova](https://github.com/kssenii)). * Fix tie handling in `arrayAUC` to match sklearn. [#65840](https://github.com/ClickHouse/ClickHouse/pull/65840) ([gabrielmcg44](https://github.com/gabrielmcg44)). * Fix possible issues with MySQL server protocol TLS connections. [#65917](https://github.com/ClickHouse/ClickHouse/pull/65917) ([Azat Khuzhin](https://github.com/azat)). From 533009b914761e317025b256b31474f44a9b4734 Mon Sep 17 00:00:00 2001 From: Denny Crane Date: Wed, 6 Nov 2024 08:57:32 -0400 Subject: [PATCH 467/680] Update AlterCommands.cpp --- src/Storages/AlterCommands.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Storages/AlterCommands.cpp b/src/Storages/AlterCommands.cpp index ab4403b3a94..c14775057a5 100644 --- a/src/Storages/AlterCommands.cpp +++ b/src/Storages/AlterCommands.cpp @@ -1496,7 +1496,7 @@ void AlterCommands::validate(const StoragePtr & table, ContextPtr context) const if (command.to_remove == AlterCommand::RemoveProperty::CODEC && column_from_table.codec == nullptr) throw Exception( ErrorCodes::BAD_ARGUMENTS, - "Column {} doesn't have TTL, cannot remove it", + "Column {} doesn't have CODEC, cannot remove it", backQuote(column_name)); if (command.to_remove == AlterCommand::RemoveProperty::COMMENT && column_from_table.comment.empty()) throw Exception( From e5b6a3c1fe9773953e01f7de161bc0c36a75b454 Mon Sep 17 00:00:00 2001 From: Pavel Kruglov <48961922+Avogar@users.noreply.github.com> Date: Wed, 6 Nov 2024 14:33:25 +0100 Subject: [PATCH 468/680] Update 03261_tuple_map_object_to_json_cast.sql --- .../queries/0_stateless/03261_tuple_map_object_to_json_cast.sql | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql b/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql index 91d3f504f92..2e5cecaf502 100644 --- a/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql +++ b/tests/queries/0_stateless/03261_tuple_map_object_to_json_cast.sql @@ -5,6 +5,7 @@ set allow_experimental_object_type = 1; set allow_experimental_variant_type = 1; set use_variant_as_common_type = 1; set enable_named_columns_in_function_tuple = 1; +set enable_analyzer = 1; select 'Map to JSON'; select map('a', number::UInt32, 'b', toDate(number), 'c', range(number), 'd', [map('e', number::UInt32)])::JSON as json, JSONAllPathsWithTypes(json) from numbers(5); From 338af374d88c134b39d75dd1f56f5630cd41fcc2 Mon Sep 17 00:00:00 2001 From: Sema Checherinda Date: Wed, 6 Nov 2024 09:52:25 +0100 Subject: [PATCH 469/680] remove the method remove in favor of the method removeIfExists --- .../AzureBlobStorage/AzureObjectStorage.cpp | 20 ++++++++--------- .../AzureBlobStorage/AzureObjectStorage.h | 4 ++-- .../Cached/CachedObjectStorage.cpp | 22 +++++++++---------- .../Cached/CachedObjectStorage.h | 4 ++-- .../DiskObjectStorageTransaction.cpp | 6 ++--- .../ObjectStorages/HDFS/HDFSObjectStorage.h | 4 ++-- src/Disks/ObjectStorages/IObjectStorage.h | 4 ++-- .../ObjectStorages/Local/LocalObjectStorage.h | 4 ++-- .../MetadataStorageFromPlainObjectStorage.cpp | 4 ++-- ...torageFromPlainObjectStorageOperations.cpp | 4 ++-- .../ObjectStorages/S3/S3ObjectStorage.cpp | 16 +++++++------- src/Disks/ObjectStorages/S3/S3ObjectStorage.h | 4 ++-- .../ObjectStorages/Web/WebObjectStorage.cpp | 16 +++++++------- .../ObjectStorages/Web/WebObjectStorage.h | 4 ++-- .../ObjectStorageQueueSource.cpp | 2 +- 15 files changed, 58 insertions(+), 60 deletions(-) diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp index 673c82806bd..959afa65672 100644 --- a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp +++ b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp @@ -278,17 +278,17 @@ void AzureObjectStorage::removeObjectImpl(const StoredObject & object, const Sha } /// Remove file. Throws exception if file doesn't exists or it's a directory. -void AzureObjectStorage::removeObject(const StoredObject & object) -{ - removeObjectImpl(object, client.get(), false); -} +// void AzureObjectStorage::removeObject(const StoredObject & object) +// { +// removeObjectImpl(object, client.get(), false); +// } -void AzureObjectStorage::removeObjects(const StoredObjects & objects) -{ - auto client_ptr = client.get(); - for (const auto & object : objects) - removeObjectImpl(object, client_ptr, false); -} +// void AzureObjectStorage::removeObjects(const StoredObjects & objects) +// { +// auto client_ptr = client.get(); +// for (const auto & object : objects) +// removeObjectImpl(object, client_ptr, false); +// } void AzureObjectStorage::removeObjectIfExists(const StoredObject & object) { diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h index 58225eccd90..433fe7a852e 100644 --- a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h +++ b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h @@ -60,9 +60,9 @@ public: const WriteSettings & write_settings = {}) override; /// Remove file. Throws exception if file doesn't exists or it's a directory. - void removeObject(const StoredObject & object) override; + //void removeObject(const StoredObject & object) override; - void removeObjects(const StoredObjects & objects) override; + //void removeObjects(const StoredObjects & objects) override; void removeObjectIfExists(const StoredObject & object) override; diff --git a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp index 163ff3a9c68..f2750e6814f 100644 --- a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp +++ b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp @@ -148,19 +148,19 @@ void CachedObjectStorage::removeCacheIfExists(const std::string & path_key_for_c cache->removeKeyIfExists(getCacheKey(path_key_for_cache), FileCache::getCommonUser().user_id); } -void CachedObjectStorage::removeObject(const StoredObject & object) -{ - removeCacheIfExists(object.remote_path); - object_storage->removeObject(object); -} +// void CachedObjectStorage::removeObject(const StoredObject & object) +// { +// removeCacheIfExists(object.remote_path); +// object_storage->removeObject(object); +// } -void CachedObjectStorage::removeObjects(const StoredObjects & objects) -{ - for (const auto & object : objects) - removeCacheIfExists(object.remote_path); +// void CachedObjectStorage::removeObjects(const StoredObjects & objects) +// { +// for (const auto & object : objects) +// removeCacheIfExists(object.remote_path); - object_storage->removeObjects(objects); -} +// object_storage->removeObjects(objects); +// } void CachedObjectStorage::removeObjectIfExists(const StoredObject & object) { diff --git a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h index b77baf21e40..7e10057e04c 100644 --- a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h +++ b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h @@ -45,9 +45,9 @@ public: size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE, const WriteSettings & write_settings = {}) override; - void removeObject(const StoredObject & object) override; + // void removeObject(const StoredObject & object) override; - void removeObjects(const StoredObjects & objects) override; + // void removeObjects(const StoredObjects & objects) override; void removeObjectIfExists(const StoredObject & object) override; diff --git a/src/Disks/ObjectStorages/DiskObjectStorageTransaction.cpp b/src/Disks/ObjectStorages/DiskObjectStorageTransaction.cpp index 64323fb6f3c..19de2bb78af 100644 --- a/src/Disks/ObjectStorages/DiskObjectStorageTransaction.cpp +++ b/src/Disks/ObjectStorages/DiskObjectStorageTransaction.cpp @@ -480,8 +480,7 @@ struct WriteFileObjectStorageOperation final : public IDiskObjectStorageOperatio void undo() override { - if (object_storage.exists(object)) - object_storage.removeObject(object); + object_storage.removeObjectIfExists(object); } void finalize() override @@ -543,8 +542,7 @@ struct CopyFileObjectStorageOperation final : public IDiskObjectStorageOperation void undo() override { - for (const auto & object : created_objects) - destination_object_storage.removeObject(object); + destination_object_storage.removeObjectsIfExist(created_objects); } void finalize() override diff --git a/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.h b/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.h index b53161beb76..317399b4753 100644 --- a/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.h +++ b/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.h @@ -78,9 +78,9 @@ public: const WriteSettings & write_settings = {}) override; /// Remove file. Throws exception if file doesn't exists or it's a directory. - void removeObject(const StoredObject & object) override; + void removeObject(const StoredObject & object); - void removeObjects(const StoredObjects & objects) override; + void removeObjects(const StoredObjects & objects); void removeObjectIfExists(const StoredObject & object) override; diff --git a/src/Disks/ObjectStorages/IObjectStorage.h b/src/Disks/ObjectStorages/IObjectStorage.h index 8dde96b8b16..adb36762539 100644 --- a/src/Disks/ObjectStorages/IObjectStorage.h +++ b/src/Disks/ObjectStorages/IObjectStorage.h @@ -161,11 +161,11 @@ public: virtual bool isRemote() const = 0; /// Remove object. Throws exception if object doesn't exists. - virtual void removeObject(const StoredObject & object) = 0; + // virtual void removeObject(const StoredObject & object) = 0; /// Remove multiple objects. Some object storages can do batch remove in a more /// optimal way. - virtual void removeObjects(const StoredObjects & objects) = 0; + // virtual void removeObjects(const StoredObjects & objects) = 0; /// Remove object on path if exists virtual void removeObjectIfExists(const StoredObject & object) = 0; diff --git a/src/Disks/ObjectStorages/Local/LocalObjectStorage.h b/src/Disks/ObjectStorages/Local/LocalObjectStorage.h index f1a0391a984..ffc151bda04 100644 --- a/src/Disks/ObjectStorages/Local/LocalObjectStorage.h +++ b/src/Disks/ObjectStorages/Local/LocalObjectStorage.h @@ -42,9 +42,9 @@ public: size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE, const WriteSettings & write_settings = {}) override; - void removeObject(const StoredObject & object) override; + void removeObject(const StoredObject & object); - void removeObjects(const StoredObjects & objects) override; + void removeObjects(const StoredObjects & objects); void removeObjectIfExists(const StoredObject & object) override; diff --git a/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.cpp b/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.cpp index d56c5d9143c..27aa9304de7 100644 --- a/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.cpp +++ b/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorage.cpp @@ -203,7 +203,7 @@ void MetadataStorageFromPlainObjectStorageTransaction::unlinkFile(const std::str { auto object_key = metadata_storage.object_storage->generateObjectKeyForPath(path, std::nullopt /* key_prefix */); auto object = StoredObject(object_key.serialize()); - metadata_storage.object_storage->removeObject(object); + metadata_storage.object_storage->removeObjectIfExists(object); } void MetadataStorageFromPlainObjectStorageTransaction::removeDirectory(const std::string & path) @@ -211,7 +211,7 @@ void MetadataStorageFromPlainObjectStorageTransaction::removeDirectory(const std if (metadata_storage.object_storage->isWriteOnce()) { for (auto it = metadata_storage.iterateDirectory(path); it->isValid(); it->next()) - metadata_storage.object_storage->removeObject(StoredObject(it->path())); + metadata_storage.object_storage->removeObjectIfExists(StoredObject(it->path())); } else { diff --git a/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorageOperations.cpp b/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorageOperations.cpp index ea57d691908..62015631aa5 100644 --- a/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorageOperations.cpp +++ b/src/Disks/ObjectStorages/MetadataStorageFromPlainObjectStorageOperations.cpp @@ -107,7 +107,7 @@ void MetadataStorageFromPlainObjectStorageCreateDirectoryOperation::undo(std::un auto metric = object_storage->getMetadataStorageMetrics().directory_map_size; CurrentMetrics::sub(metric, 1); - object_storage->removeObject(StoredObject(metadata_object_key.serialize(), path / PREFIX_PATH_FILE_NAME)); + object_storage->removeObjectIfExists(StoredObject(metadata_object_key.serialize(), path / PREFIX_PATH_FILE_NAME)); } else if (write_created) object_storage->removeObjectIfExists(StoredObject(metadata_object_key.serialize(), path / PREFIX_PATH_FILE_NAME)); @@ -247,7 +247,7 @@ void MetadataStorageFromPlainObjectStorageRemoveDirectoryOperation::execute(std: auto metadata_object_key = createMetadataObjectKey(key_prefix, metadata_key_prefix); auto metadata_object = StoredObject(/*remote_path*/ metadata_object_key.serialize(), /*local_path*/ path / PREFIX_PATH_FILE_NAME); - object_storage->removeObject(metadata_object); + object_storage->removeObjectIfExists(metadata_object); { std::lock_guard lock(path_map.mutex); diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp index 47ef97401f2..7ed118c6b07 100644 --- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp +++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp @@ -326,20 +326,20 @@ void S3ObjectStorage::removeObjectsImpl(const StoredObjects & objects, bool if_e ProfileEvents::DiskS3DeleteObjects); } -void S3ObjectStorage::removeObject(const StoredObject & object) -{ - removeObjectImpl(object, false); -} +// void S3ObjectStorage::removeObject(const StoredObject & object) +// { +// removeObjectImpl(object, false); +// } void S3ObjectStorage::removeObjectIfExists(const StoredObject & object) { removeObjectImpl(object, true); } -void S3ObjectStorage::removeObjects(const StoredObjects & objects) -{ - removeObjectsImpl(objects, false); -} +// void S3ObjectStorage::removeObjects(const StoredObjects & objects) +// { +// removeObjectsImpl(objects, false); +// } void S3ObjectStorage::removeObjectsIfExist(const StoredObjects & objects) { diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.h b/src/Disks/ObjectStorages/S3/S3ObjectStorage.h index d6e84cf57ef..a2aeaf8a43c 100644 --- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.h +++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.h @@ -102,11 +102,11 @@ public: ObjectStorageIteratorPtr iterate(const std::string & path_prefix, size_t max_keys) const override; /// Uses `DeleteObjectRequest`. - void removeObject(const StoredObject & object) override; + //void removeObject(const StoredObject & object) override; /// Uses `DeleteObjectsRequest` if it is allowed by `s3_capabilities`, otherwise `DeleteObjectRequest`. /// `DeleteObjectsRequest` is not supported on GCS, see https://issuetracker.google.com/issues/162653700 . - void removeObjects(const StoredObjects & objects) override; + //void removeObjects(const StoredObjects & objects) override; /// Uses `DeleteObjectRequest`. void removeObjectIfExists(const StoredObject & object) override; diff --git a/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp b/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp index 871d3b506f6..1503d5819eb 100644 --- a/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp +++ b/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp @@ -254,15 +254,15 @@ std::unique_ptr WebObjectStorage::writeObject( /// NOLI throwNotAllowed(); } -void WebObjectStorage::removeObject(const StoredObject &) -{ - throwNotAllowed(); -} +// void WebObjectStorage::removeObject(const StoredObject &) +// { +// throwNotAllowed(); +// } -void WebObjectStorage::removeObjects(const StoredObjects &) -{ - throwNotAllowed(); -} +// void WebObjectStorage::removeObjects(const StoredObjects &) +// { +// throwNotAllowed(); +// } void WebObjectStorage::removeObjectIfExists(const StoredObject &) { diff --git a/src/Disks/ObjectStorages/Web/WebObjectStorage.h b/src/Disks/ObjectStorages/Web/WebObjectStorage.h index 573221b7e21..ae52cc20f9b 100644 --- a/src/Disks/ObjectStorages/Web/WebObjectStorage.h +++ b/src/Disks/ObjectStorages/Web/WebObjectStorage.h @@ -47,9 +47,9 @@ public: size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE, const WriteSettings & write_settings = {}) override; - void removeObject(const StoredObject & object) override; + // void removeObject(const StoredObject & object) override; - void removeObjects(const StoredObjects & objects) override; + // void removeObjects(const StoredObjects & objects) override; void removeObjectIfExists(const StoredObject & object) override; diff --git a/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.cpp b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.cpp index ba1a97bc2fb..e702f07208a 100644 --- a/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.cpp +++ b/src/Storages/ObjectStorageQueue/ObjectStorageQueueSource.cpp @@ -659,7 +659,7 @@ void ObjectStorageQueueSource::applyActionAfterProcessing(const String & path) { if (files_metadata->getTableMetadata().after_processing == ObjectStorageQueueAction::DELETE) { - object_storage->removeObject(StoredObject(path)); + object_storage->removeObjectIfExists(StoredObject(path)); } } From d270885bfa52548dbf342b5ddacf8803a354d2a8 Mon Sep 17 00:00:00 2001 From: Amos Bird Date: Wed, 6 Nov 2024 21:37:47 +0800 Subject: [PATCH 470/680] Allow specifying cmdline flags in integration test --- tests/integration/helpers/cluster.py | 26 +++++++++++++++++--------- 1 file changed, 17 insertions(+), 9 deletions(-) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index 6751f205fb8..e2237363131 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -1653,6 +1653,7 @@ class ClickHouseCluster: copy_common_configs=True, config_root_name="clickhouse", extra_configs=[], + extra_args="", randomize_settings=True, ) -> "ClickHouseInstance": """Add an instance to the cluster. @@ -1740,6 +1741,7 @@ class ClickHouseCluster: with_postgres_cluster=with_postgres_cluster, with_postgresql_java_client=with_postgresql_java_client, clickhouse_start_command=clickhouse_start_command, + clickhouse_start_extra_args=extra_args, main_config_name=main_config_name, users_config_name=users_config_name, copy_common_configs=copy_common_configs, @@ -3368,6 +3370,7 @@ class ClickHouseInstance: with_postgres_cluster, with_postgresql_java_client, clickhouse_start_command=CLICKHOUSE_START_COMMAND, + clickhouse_start_extra_args="", main_config_name="config.xml", users_config_name="users.xml", copy_common_configs=True, @@ -3463,11 +3466,18 @@ class ClickHouseInstance: self.users_config_name = users_config_name self.copy_common_configs = copy_common_configs - self.clickhouse_start_command = clickhouse_start_command.replace( + clickhouse_start_command_with_conf = clickhouse_start_command.replace( "{main_config_file}", self.main_config_name ) - self.clickhouse_stay_alive_command = "bash -c \"trap 'pkill tail' INT TERM; {} --daemon; coproc tail -f /dev/null; wait $$!\"".format( - clickhouse_start_command + + self.clickhouse_start_command = "{} -- {}".format( + clickhouse_start_command_with_conf, clickhouse_start_extra_args + ) + self.clickhouse_start_command_in_daemon = "{} --daemon -- {}".format( + clickhouse_start_command_with_conf, clickhouse_start_extra_args + ) + self.clickhouse_stay_alive_command = "bash -c \"trap 'pkill tail' INT TERM; {}; coproc tail -f /dev/null; wait $$!\"".format( + self.clickhouse_start_command_in_daemon ) self.path = p.join(self.cluster.instances_dir, name) @@ -3910,7 +3920,7 @@ class ClickHouseInstance: if pid is None: logging.debug("No clickhouse process running. Start new one.") self.exec_in_container( - ["bash", "-c", "{} --daemon".format(self.clickhouse_start_command)], + ["bash", "-c", self.clickhouse_start_command_in_daemon], user=str(os.getuid()), ) if expected_to_fail: @@ -4230,7 +4240,7 @@ class ClickHouseInstance: user="root", ) self.exec_in_container( - ["bash", "-c", "{} --daemon".format(self.clickhouse_start_command)], + ["bash", "-c", self.clickhouse_start_command_in_daemon], user=str(os.getuid()), ) @@ -4311,7 +4321,7 @@ class ClickHouseInstance: ] ) self.exec_in_container( - ["bash", "-c", "{} --daemon".format(self.clickhouse_start_command)], + ["bash", "-c", self.clickhouse_start_command_in_daemon], user=str(os.getuid()), ) @@ -4704,9 +4714,7 @@ class ClickHouseInstance: entrypoint_cmd = self.clickhouse_start_command if self.stay_alive: - entrypoint_cmd = self.clickhouse_stay_alive_command.replace( - "{main_config_file}", self.main_config_name - ) + entrypoint_cmd = self.clickhouse_stay_alive_command else: entrypoint_cmd = ( "[" From 71a0e7f07f41c0388b98849717240e845c53dc67 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Wed, 6 Nov 2024 13:34:05 +0000 Subject: [PATCH 471/680] Split tests --- ...> 02354_vector_search_bug_52282.reference} | 0 .../02354_vector_search_bug_52282.sql | 13 ++ ...> 02354_vector_search_bug_69085.reference} | 9 -- .../02354_vector_search_bug_69085.sql | 52 +++++++ .../02354_vector_search_bug_71381.reference | 0 .../02354_vector_search_bug_71381.sql | 20 +++ ...h_bug_adaptive_index_granularity.reference | 0 ..._search_bug_adaptive_index_granularity.sql | 20 +++ ...search_bug_different_array_sizes.reference | 0 ...ector_search_bug_different_array_sizes.sql | 24 ++++ ...ctor_search_bug_multiple_indexes.reference | 0 ...354_vector_search_bug_multiple_indexes.sql | 14 ++ ...vector_search_bug_multiple_marks.reference | 2 + ...02354_vector_search_bug_multiple_marks.sql | 25 ++++ .../0_stateless/02354_vector_search_bugs.sql | 134 ------------------ .../02354_vector_search_multiple_indexes.sql | 1 + 16 files changed, 171 insertions(+), 143 deletions(-) rename tests/queries/0_stateless/{02354_vector_search_multiple_indexes.reference => 02354_vector_search_bug_52282.reference} (100%) create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_52282.sql rename tests/queries/0_stateless/{02354_vector_search_bugs.reference => 02354_vector_search_bug_69085.reference} (68%) create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_69085.sql create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_71381.reference create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_71381.sql create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_adaptive_index_granularity.reference create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_adaptive_index_granularity.sql create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_different_array_sizes.reference create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_different_array_sizes.sql create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_multiple_indexes.reference create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_multiple_indexes.sql create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.reference create mode 100644 tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.sql delete mode 100644 tests/queries/0_stateless/02354_vector_search_bugs.sql diff --git a/tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference b/tests/queries/0_stateless/02354_vector_search_bug_52282.reference similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference rename to tests/queries/0_stateless/02354_vector_search_bug_52282.reference diff --git a/tests/queries/0_stateless/02354_vector_search_bug_52282.sql b/tests/queries/0_stateless/02354_vector_search_bug_52282.sql new file mode 100644 index 00000000000..b8066ce278a --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_bug_52282.sql @@ -0,0 +1,13 @@ +-- Tags: no-fasttest, no-ordinary-database + +SET allow_experimental_vector_similarity_index = 1; + +-- Issue #52258: Vector similarity indexes must reject empty Arrays or Arrays with default values + +DROP TABLE IF EXISTS tab; + +CREATE TABLE tab (id UInt64, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree() ORDER BY id; +INSERT INTO tab VALUES (1, []); -- { serverError INCORRECT_DATA } +INSERT INTO tab (id) VALUES (1); -- { serverError INCORRECT_DATA } + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_bugs.reference b/tests/queries/0_stateless/02354_vector_search_bug_69085.reference similarity index 68% rename from tests/queries/0_stateless/02354_vector_search_bugs.reference rename to tests/queries/0_stateless/02354_vector_search_bug_69085.reference index dec921cf586..3b4e2d9ef17 100644 --- a/tests/queries/0_stateless/02354_vector_search_bugs.reference +++ b/tests/queries/0_stateless/02354_vector_search_bug_69085.reference @@ -1,10 +1,3 @@ -Rejects INSERTs of Arrays with different sizes -Issue #52258: Empty Arrays or Arrays with default values are rejected -It is possible to create parts with different Array vector sizes but there will be an error at query time -Correctness of index with > 1 mark -1 [1,0] 0 -9000 [9000,0] 0 -Issue #69085: Reference vector computed by a subquery Expression (Projection) Limit (preliminary LIMIT (without OFFSET)) Sorting (Sorting for ORDER BY) @@ -40,5 +33,3 @@ Expression (Projection) Condition: true Parts: 1/1 Granules: 4/4 -index_granularity_bytes = 0 is disallowed -Issue #71381: Vector similarity index and other skipping indexes used on the same table diff --git a/tests/queries/0_stateless/02354_vector_search_bug_69085.sql b/tests/queries/0_stateless/02354_vector_search_bug_69085.sql new file mode 100644 index 00000000000..4dbcdf66e36 --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_bug_69085.sql @@ -0,0 +1,52 @@ +-- Tags: no-fasttest, no-ordinary-database + +SET allow_experimental_vector_similarity_index = 1; +SET enable_analyzer = 0; + +-- Issue #69085: Reference vector for vector search is computed by a subquery + +DROP TABLE IF EXISTS tab; + +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'cosineDistance', 'f16', 0, 0) GRANULARITY 2) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 3; +INSERT INTO tab VALUES (0, [4.6, 2.3]), (1, [2.0, 3.2]), (2, [4.2, 3.4]), (3, [5.3, 2.9]), (4, [2.4, 5.2]), (5, [5.3, 2.3]), (6, [1.0, 9.3]), (7, [5.5, 4.7]), (8, [6.4, 3.5]), (9, [5.3, 2.5]), (10, [6.4, 3.4]), (11, [6.4, 3.2]); + +-- works +EXPLAIN indexes = 1 +WITH [0., 2.] AS reference_vec +SELECT + id, + vec, + cosineDistance(vec, reference_vec) AS distance +FROM tab +ORDER BY distance +LIMIT 1; + +-- does not work +EXPLAIN indexes = 1 +WITH ( + SELECT vec + FROM tab + LIMIT 1 +) AS reference_vec +SELECT + id, + vec, + cosineDistance(vec, reference_vec) AS distance +FROM tab +ORDER BY distance +LIMIT 1; + +-- does not work as well +EXPLAIN indexes = 1 +WITH ( + SELECT [0., 2.] +) AS reference_vec +SELECT + id, + vec, + cosineDistance(vec, reference_vec) AS distance +FROM tab +ORDER BY distance +LIMIT 1; + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_bug_71381.reference b/tests/queries/0_stateless/02354_vector_search_bug_71381.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/02354_vector_search_bug_71381.sql b/tests/queries/0_stateless/02354_vector_search_bug_71381.sql new file mode 100644 index 00000000000..9e3246700b8 --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_bug_71381.sql @@ -0,0 +1,20 @@ +-- Tags: no-fasttest, no-ordinary-database + +SET allow_experimental_vector_similarity_index = 1; + +-- Issue #71381: Usage of vector similarity index and further skipping indexes on the same table + +DROP TABLE IF EXISTS tab; + +CREATE TABLE tab( + val String, + vec Array(Float32), + INDEX ann_idx vec TYPE vector_similarity('hnsw', 'cosineDistance'), + INDEX set_idx val TYPE set(100) +) +ENGINE = MergeTree() +ORDER BY tuple(); + +INSERT INTO tab VALUES ('hello world', [0.0]); + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_bug_adaptive_index_granularity.reference b/tests/queries/0_stateless/02354_vector_search_bug_adaptive_index_granularity.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/02354_vector_search_bug_adaptive_index_granularity.sql b/tests/queries/0_stateless/02354_vector_search_bug_adaptive_index_granularity.sql new file mode 100644 index 00000000000..208b5b7a874 --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_bug_adaptive_index_granularity.sql @@ -0,0 +1,20 @@ +-- Tags: no-fasttest, no-ordinary-database + +-- Tests that vector similarity indexes cannot be created with index_granularity_bytes = 0 + +SET allow_experimental_vector_similarity_index = 1; + +DROP TABLE IF EXISTS tab; + +-- If adaptive index granularity is disabled, certain vector search queries with PREWHERE run into LOGICAL_ERRORs. +-- SET allow_experimental_vector_similarity_index = 1; +-- CREATE TABLE tab (`id` Int32, `vec` Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance') GRANULARITY 100000000) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes = 0; +-- INSERT INTO tab SELECT number, [toFloat32(number), 0.] FROM numbers(10000); +-- WITH [1., 0.] AS reference_vec SELECT id, L2Distance(vec, reference_vec) FROM tab PREWHERE toLowCardinality(10) ORDER BY L2Distance(vec, reference_vec) ASC LIMIT 100; +-- As a workaround, force enabled adaptive index granularity for now (it is the default anyways). +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes = 0; -- { serverError INVALID_SETTING_VALUE } + +CREATE TABLE tab(id Int32, vec Array(Float32)) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes = 0; +ALTER TABLE tab ADD INDEX vec_idx1(vec) TYPE vector_similarity('hnsw', 'cosineDistance'); -- { serverError INVALID_SETTING_VALUE } + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_bug_different_array_sizes.reference b/tests/queries/0_stateless/02354_vector_search_bug_different_array_sizes.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/02354_vector_search_bug_different_array_sizes.sql b/tests/queries/0_stateless/02354_vector_search_bug_different_array_sizes.sql new file mode 100644 index 00000000000..41b9d7869e4 --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_bug_different_array_sizes.sql @@ -0,0 +1,24 @@ +-- Tags: no-fasttest, no-ordinary-database + +SET allow_experimental_vector_similarity_index = 1; +SET enable_analyzer = 1; -- 0 vs. 1 produce slightly different error codes, make it future-proof + +DROP TABLE IF EXISTS tab; + +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; + +-- Vector similarity indexes reject INSERTs of Arrays with different sizes +INSERT INTO tab values (0, [2.2, 2.3]) (1, [3.1, 3.2, 3.3]); -- { serverError INCORRECT_DATA } + +-- It is possible to create parts with different Array vector sizes but there will be an error at query time +SYSTEM STOP MERGES tab; +INSERT INTO tab values (0, [2.2, 2.3]) (1, [3.1, 3.2]); +INSERT INTO tab values (2, [2.2, 2.3, 2.4]) (3, [3.1, 3.2, 3.3]); + +WITH [0.0, 2.0] AS reference_vec +SELECT id, vec, L2Distance(vec, reference_vec) +FROM tab +ORDER BY L2Distance(vec, reference_vec) +LIMIT 3; -- { serverError SIZES_OF_ARRAYS_DONT_MATCH } + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_bug_multiple_indexes.reference b/tests/queries/0_stateless/02354_vector_search_bug_multiple_indexes.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/02354_vector_search_bug_multiple_indexes.sql b/tests/queries/0_stateless/02354_vector_search_bug_multiple_indexes.sql new file mode 100644 index 00000000000..f1cfc041233 --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_bug_multiple_indexes.sql @@ -0,0 +1,14 @@ +-- Tags: no-fasttest, no-ordinary-database + +-- Tests that multiple vector similarity indexes can be created on the same column (even if that makes no sense) + +SET allow_experimental_vector_similarity_index = 1; + +DROP TABLE IF EXISTS tab; +CREATE TABLE tab (id Int32, vec Array(Float32), PRIMARY KEY id, INDEX vec_idx(vec) TYPE vector_similarity('hnsw', 'L2Distance')); + +ALTER TABLE tab ADD INDEX idx(vec) TYPE minmax; +ALTER TABLE tab ADD INDEX vec_idx1(vec) TYPE vector_similarity('hnsw', 'cosineDistance'); +ALTER TABLE tab ADD INDEX vec_idx2(vec) TYPE vector_similarity('hnsw', 'L2Distance'); -- silly but creating the same index also works for non-vector indexes ... + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.reference b/tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.reference new file mode 100644 index 00000000000..117bf2cead8 --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.reference @@ -0,0 +1,2 @@ +1 [1,0] 0 +9000 [9000,0] 0 diff --git a/tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.sql b/tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.sql new file mode 100644 index 00000000000..fb99dd2361c --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.sql @@ -0,0 +1,25 @@ +-- Tags: no-fasttest, no-ordinary-database + +-- Tests correctness of vector similarity index with > 1 mark + +SET allow_experimental_vector_similarity_index = 1; +SET enable_analyzer = 0; + +DROP TABLE IF EXISTS tab; + +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 8192; +INSERT INTO tab SELECT number, [toFloat32(number), 0.0] from numbers(10000); + +WITH [1.0, 0.0] AS reference_vec +SELECT id, vec, L2Distance(vec, reference_vec) +FROM tab +ORDER BY L2Distance(vec, reference_vec) +LIMIT 1; + +WITH [9000.0, 0.0] AS reference_vec +SELECT id, vec, L2Distance(vec, reference_vec) +FROM tab +ORDER BY L2Distance(vec, reference_vec) +LIMIT 1; + +DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_bugs.sql b/tests/queries/0_stateless/02354_vector_search_bugs.sql deleted file mode 100644 index 276d4eb5b59..00000000000 --- a/tests/queries/0_stateless/02354_vector_search_bugs.sql +++ /dev/null @@ -1,134 +0,0 @@ --- Tags: no-fasttest, no-ordinary-database - --- Tests various bugs and special cases for vector indexes. - -SET allow_experimental_vector_similarity_index = 1; -SET enable_analyzer = 1; -- 0 vs. 1 produce slightly different error codes, make it future-proof - -DROP TABLE IF EXISTS tab; - -SELECT 'Rejects INSERTs of Arrays with different sizes'; - -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; -INSERT INTO tab values (0, [2.2, 2.3]) (1, [3.1, 3.2, 3.3]); -- { serverError INCORRECT_DATA } -DROP TABLE tab; - -SELECT 'Issue #52258: Empty Arrays or Arrays with default values are rejected'; - -CREATE TABLE tab (id UInt64, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree() ORDER BY id; -INSERT INTO tab VALUES (1, []); -- { serverError INCORRECT_DATA } -INSERT INTO tab (id) VALUES (1); -- { serverError INCORRECT_DATA } -DROP TABLE tab; - -SELECT 'It is possible to create parts with different Array vector sizes but there will be an error at query time'; - -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id; -SYSTEM STOP MERGES tab; -INSERT INTO tab values (0, [2.2, 2.3]) (1, [3.1, 3.2]); -INSERT INTO tab values (2, [2.2, 2.3, 2.4]) (3, [3.1, 3.2, 3.3]); - -WITH [0.0, 2.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab -ORDER BY L2Distance(vec, reference_vec) -LIMIT 3; -- { serverError SIZES_OF_ARRAYS_DONT_MATCH } - -DROP TABLE tab; - -SELECT 'Correctness of index with > 1 mark'; - -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 8192; -INSERT INTO tab SELECT number, [toFloat32(number), 0.0] from numbers(10000); - -WITH [1.0, 0.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab -ORDER BY L2Distance(vec, reference_vec) -LIMIT 1; - -WITH [9000.0, 0.0] AS reference_vec -SELECT id, vec, L2Distance(vec, reference_vec) -FROM tab -ORDER BY L2Distance(vec, reference_vec) -LIMIT 1; - -DROP TABLE tab; - -SELECT 'Issue #69085: Reference vector computed by a subquery'; - -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'cosineDistance', 'f16', 0, 0) GRANULARITY 2) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity = 3; -INSERT INTO tab VALUES (0, [4.6, 2.3]), (1, [2.0, 3.2]), (2, [4.2, 3.4]), (3, [5.3, 2.9]), (4, [2.4, 5.2]), (5, [5.3, 2.3]), (6, [1.0, 9.3]), (7, [5.5, 4.7]), (8, [6.4, 3.5]), (9, [5.3, 2.5]), (10, [6.4, 3.4]), (11, [6.4, 3.2]); - --- works -EXPLAIN indexes = 1 -WITH [0., 2.] AS reference_vec -SELECT - id, - vec, - cosineDistance(vec, reference_vec) AS distance -FROM tab -ORDER BY distance -LIMIT 1 -SETTINGS enable_analyzer = 0; - --- does not work -EXPLAIN indexes = 1 -WITH ( - SELECT vec - FROM tab - LIMIT 1 -) AS reference_vec -SELECT - id, - vec, - cosineDistance(vec, reference_vec) AS distance -FROM tab -ORDER BY distance -LIMIT 1 -SETTINGS enable_analyzer = 0; - --- does not work as well -EXPLAIN indexes = 1 -WITH ( - SELECT [0., 2.] -) AS reference_vec -SELECT - id, - vec, - cosineDistance(vec, reference_vec) AS distance -FROM tab -ORDER BY distance -LIMIT 1 -SETTINGS enable_analyzer = 0; - -DROP TABLE tab; - -SELECT 'index_granularity_bytes = 0 is disallowed'; - --- If adaptive index granularity is disabled, certain vector search queries with PREWHERE run into LOGICAL_ERRORs. --- SET allow_experimental_vector_similarity_index = 1; --- CREATE TABLE tab (`id` Int32, `vec` Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance') GRANULARITY 100000000) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes = 0; --- INSERT INTO tab SELECT number, [toFloat32(number), 0.] FROM numbers(10000); --- WITH [1., 0.] AS reference_vec SELECT id, L2Distance(vec, reference_vec) FROM tab PREWHERE toLowCardinality(10) ORDER BY L2Distance(vec, reference_vec) ASC LIMIT 100; --- As a workaround, force enabled adaptive index granularity for now (it is the default anyways). -CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance')) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes = 0; -- { serverError INVALID_SETTING_VALUE } - -CREATE TABLE tab(id Int32, vec Array(Float32)) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes = 0; -ALTER TABLE tab ADD INDEX vec_idx1(vec) TYPE vector_similarity('hnsw', 'cosineDistance'); -- { serverError INVALID_SETTING_VALUE } - -DROP TABLE tab; - -SELECT 'Issue #71381: Vector similarity index and other skipping indexes used on the same table'; - -CREATE TABLE tab( - val String, - vec Array(Float32), - INDEX ann_idx vec TYPE vector_similarity('hnsw', 'cosineDistance'), - INDEX set_idx val TYPE set(100) -) -ENGINE = MergeTree() -ORDER BY tuple(); - -INSERT INTO tab VALUES ('hello world', [0.0]); - -DROP TABLE tab; diff --git a/tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql b/tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql index f1cfc041233..aedba286a9f 100644 --- a/tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql +++ b/tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql @@ -5,6 +5,7 @@ SET allow_experimental_vector_similarity_index = 1; DROP TABLE IF EXISTS tab; + CREATE TABLE tab (id Int32, vec Array(Float32), PRIMARY KEY id, INDEX vec_idx(vec) TYPE vector_similarity('hnsw', 'L2Distance')); ALTER TABLE tab ADD INDEX idx(vec) TYPE minmax; From 4e3bde24605e1401749703bfe2eb28d7298f6630 Mon Sep 17 00:00:00 2001 From: alesapin Date: Wed, 6 Nov 2024 14:52:59 +0100 Subject: [PATCH 472/680] Add ProfileEvents for merge selector timings --- src/Common/ProfileEvents.cpp | 6 ++++ .../MergeTree/MergeTreeDataMergerMutator.cpp | 30 +++++++++++++++++-- 2 files changed, 34 insertions(+), 2 deletions(-) diff --git a/src/Common/ProfileEvents.cpp b/src/Common/ProfileEvents.cpp index 0774d36462d..7b9f670d340 100644 --- a/src/Common/ProfileEvents.cpp +++ b/src/Common/ProfileEvents.cpp @@ -746,6 +746,12 @@ The server successfully detected this situation and will download merged part fr M(ReadTaskRequestsSentElapsedMicroseconds, "Time spent in callbacks requested from the remote server back to the initiator server to choose the read task (for s3Cluster table function and similar). Measured on the remote server side.", ValueType::Microseconds) \ M(MergeTreeReadTaskRequestsSentElapsedMicroseconds, "Time spent in callbacks requested from the remote server back to the initiator server to choose the read task (for MergeTree tables). Measured on the remote server side.", ValueType::Microseconds) \ M(MergeTreeAllRangesAnnouncementsSentElapsedMicroseconds, "Time spent in sending the announcement from the remote server to the initiator server about the set of data parts (for MergeTree tables). Measured on the remote server side.", ValueType::Microseconds) \ + M(MergerMutatorsGetPartsForMergeElapsedMicroseconds, "Time spent to take data parts snapshot to build ranges from them.", ValueType::Microseconds) \ + M(MergerMutatorPrepareRangesForMergeElapsedMicroseconds, "Time spent to prepare parts ranges which can be merged according to merge predicate.", ValueType::Microseconds) \ + M(MergerMutatorSelectPartsForMergeElapsedMicroseconds, "Time spent to select parts from ranges which can be merged.", ValueType::Microseconds) \ + M(MergerMutatorRangesForMergeCount, "Amount of candidate ranges for merge", ValueType::Number) \ + M(MergerMutatorPartsInRangesForMergeCount, "Amount of candidate parts for merge", ValueType::Number) \ + M(MergerMutatorSelectRangePartsCount, "Amount of parts in selected range for merge", ValueType::Number) \ \ M(ConnectionPoolIsFullMicroseconds, "Total time spent waiting for a slot in connection pool.", ValueType::Microseconds) \ M(AsyncLoaderWaitMicroseconds, "Total time a query was waiting for async loader jobs.", ValueType::Microseconds) \ diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp index 6b9638b11d2..3d935f8b70d 100644 --- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp +++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp @@ -48,6 +48,17 @@ namespace CurrentMetrics { extern const Metric BackgroundMergesAndMutationsPoolTask; } +namespace ProfileEvents +{ + + extern const Event MergeTreeAllRangesAnnouncementsSentElapsedMicroseconds; + extern const Event MergerMutatorsGetPartsForMergeElapsedMicroseconds; + extern const Event MergerMutatorPrepareRangesForMergeElapsedMicroseconds; + extern const Event MergerMutatorSelectPartsForMergeElapsedMicroseconds; + extern const Event MergerMutatorRangesForMergeCount; + extern const Event MergerMutatorPartsInRangesForMergeCount; + extern const Event MergerMutatorSelectRangePartsCount; +} namespace DB { @@ -215,6 +226,7 @@ MergeTreeDataMergerMutator::PartitionIdsHint MergeTreeDataMergerMutator::getPart { PartitionIdsHint res; MergeTreeData::DataPartsVector data_parts = getDataPartsToSelectMergeFrom(txn); + if (data_parts.empty()) return res; @@ -272,6 +284,8 @@ MergeTreeDataMergerMutator::PartitionIdsHint MergeTreeDataMergerMutator::getPart MergeTreeData::DataPartsVector MergeTreeDataMergerMutator::getDataPartsToSelectMergeFrom( const MergeTreeTransactionPtr & txn, const PartitionIdsHint * partitions_hint) const { + + Stopwatch get_data_parts_for_merge_timer; auto res = getDataPartsToSelectMergeFrom(txn); if (!partitions_hint) return res; @@ -280,6 +294,8 @@ MergeTreeData::DataPartsVector MergeTreeDataMergerMutator::getDataPartsToSelectM { return !partitions_hint->contains(part->info.partition_id); }); + + ProfileEvents::increment(ProfileEvents::MergerMutatorsGetPartsForMergeElapsedMicroseconds, get_data_parts_for_merge_timer.elapsedMicroseconds()); return res; } @@ -357,6 +373,7 @@ MergeTreeDataMergerMutator::MergeSelectingInfo MergeTreeDataMergerMutator::getPo const MergeTreeTransactionPtr & txn, PreformattedMessage & out_disable_reason) const { + Stopwatch ranges_for_merge_timer; MergeSelectingInfo res; res.current_time = std::time(nullptr); @@ -457,6 +474,10 @@ MergeTreeDataMergerMutator::MergeSelectingInfo MergeTreeDataMergerMutator::getPo prev_part = ∂ } + ProfileEvents::increment(ProfileEvents::MergerMutatorPartsInRangesForMergeCount, res.parts_selected_precondition); + ProfileEvents::increment(ProfileEvents::MergerMutatorRangesForMergeCount, res.parts_ranges.size()); + ProfileEvents::increment(ProfileEvents::MergerMutatorPrepareRangesForMergeElapsedMicroseconds, ranges_for_merge_timer.elapsedMicroseconds()); + return res; } @@ -471,6 +492,7 @@ SelectPartsDecision MergeTreeDataMergerMutator::selectPartsToMergeFromRanges( PreformattedMessage & out_disable_reason, bool dry_run) { + Stopwatch select_parts_from_ranges_timer; const auto data_settings = data.getSettings(); IMergeSelector::PartsRange parts_to_merge; @@ -570,7 +592,8 @@ SelectPartsDecision MergeTreeDataMergerMutator::selectPartsToMergeFromRanges( if (parts_to_merge.empty()) { - out_disable_reason = PreformattedMessage::create("Did not find any parts to merge (with usual merge selectors)"); + ProfileEvents::increment(ProfileEvents::MergerMutatorSelectPartsForMergeElapsedMicroseconds, select_parts_from_ranges_timer.elapsedMicroseconds()); + out_disable_reason = PreformattedMessage::create("Did not find any parts to merge (with usual merge selectors) in {}", select_parts_from_ranges_timer.elapsedMicroseconds() / 1000); return SelectPartsDecision::CANNOT_SELECT; } } @@ -583,8 +606,11 @@ SelectPartsDecision MergeTreeDataMergerMutator::selectPartsToMergeFromRanges( parts.push_back(part); } - LOG_DEBUG(log, "Selected {} parts from {} to {}", parts.size(), parts.front()->name, parts.back()->name); + LOG_DEBUG(log, "Selected {} parts from {} to {} in {}ms", parts.size(), parts.front()->name, parts.back()->name, select_parts_from_ranges_timer.elapsedMicroseconds() / 1000); + ProfileEvents::increment(ProfileEvents::MergerMutatorSelectRangePartsCount, parts.size()); + future_part->assign(std::move(parts)); + ProfileEvents::increment(ProfileEvents::MergerMutatorSelectPartsForMergeElapsedMicroseconds, select_parts_from_ranges_timer.elapsedMicroseconds()); return SelectPartsDecision::SELECTED; } From afb92f04e62b446fb5c8b0417c658f206ce2a55d Mon Sep 17 00:00:00 2001 From: Alexander Gololobov Date: Wed, 6 Nov 2024 14:56:30 +0100 Subject: [PATCH 473/680] Added ms --- src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp index 3d935f8b70d..4d0fb7f9eeb 100644 --- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp +++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp @@ -593,7 +593,7 @@ SelectPartsDecision MergeTreeDataMergerMutator::selectPartsToMergeFromRanges( if (parts_to_merge.empty()) { ProfileEvents::increment(ProfileEvents::MergerMutatorSelectPartsForMergeElapsedMicroseconds, select_parts_from_ranges_timer.elapsedMicroseconds()); - out_disable_reason = PreformattedMessage::create("Did not find any parts to merge (with usual merge selectors) in {}", select_parts_from_ranges_timer.elapsedMicroseconds() / 1000); + out_disable_reason = PreformattedMessage::create("Did not find any parts to merge (with usual merge selectors) in {}ms", select_parts_from_ranges_timer.elapsedMicroseconds() / 1000); return SelectPartsDecision::CANNOT_SELECT; } } From 7795d43055a3bcf4c5f0710152d4c71cc183d000 Mon Sep 17 00:00:00 2001 From: Dmitry Novik Date: Mon, 4 Nov 2024 17:03:16 +0100 Subject: [PATCH 474/680] Analyzer: Check what happens after if-condition removal --- src/Analyzer/Resolve/QueryAnalyzer.cpp | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index cb3087af707..55bbf4907bb 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -5448,16 +5448,13 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier */ scope.use_identifier_lookup_to_result_cache = false; - if (query_node_typed.getJoinTree()) - { - TableExpressionsAliasVisitor table_expressions_visitor(scope); - table_expressions_visitor.visit(query_node_typed.getJoinTree()); + TableExpressionsAliasVisitor table_expressions_visitor(scope); + table_expressions_visitor.visit(query_node_typed.getJoinTree()); - initializeQueryJoinTreeNode(query_node_typed.getJoinTree(), scope); - scope.aliases.alias_name_to_table_expression_node.clear(); + initializeQueryJoinTreeNode(query_node_typed.getJoinTree(), scope); + scope.aliases.alias_name_to_table_expression_node.clear(); - resolveQueryJoinTreeNode(query_node_typed.getJoinTree(), scope, visitor); - } + resolveQueryJoinTreeNode(query_node_typed.getJoinTree(), scope, visitor); if (!scope.group_by_use_nulls) scope.use_identifier_lookup_to_result_cache = true; From f4c0254254b7cfe1f603dc57350a226c9d5dd993 Mon Sep 17 00:00:00 2001 From: Ilya Golshtein Date: Wed, 6 Nov 2024 14:52:55 +0000 Subject: [PATCH 475/680] fix_test_drop_complex_columns: flaky check for test_drop_after_fetch --- .../test_replicated_s3_zero_copy_drop_partition/test.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/integration/test_replicated_s3_zero_copy_drop_partition/test.py b/tests/integration/test_replicated_s3_zero_copy_drop_partition/test.py index 9937c0ed4ea..7623a24c0ef 100644 --- a/tests/integration/test_replicated_s3_zero_copy_drop_partition/test.py +++ b/tests/integration/test_replicated_s3_zero_copy_drop_partition/test.py @@ -65,6 +65,8 @@ CREATE TABLE test_s3(c1 Int8, c2 Date) ENGINE = ReplicatedMergeTree('/test/table objects_after = get_objects_in_data_path() assert objects_before == objects_after + node1.query("DROP TABLE test_local SYNC") + node1.query("DROP TABLE test_s3 SYNC") def test_drop_complex_columns(started_cluster): From 33bd082149ca207b55915cd78c8c19cdc6aacdc9 Mon Sep 17 00:00:00 2001 From: alesapin Date: Wed, 6 Nov 2024 16:00:25 +0100 Subject: [PATCH 476/680] Followup --- src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp | 1 - 1 file changed, 1 deletion(-) diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp index 3d935f8b70d..40c4db3a69d 100644 --- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp +++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp @@ -51,7 +51,6 @@ namespace CurrentMetrics namespace ProfileEvents { - extern const Event MergeTreeAllRangesAnnouncementsSentElapsedMicroseconds; extern const Event MergerMutatorsGetPartsForMergeElapsedMicroseconds; extern const Event MergerMutatorPrepareRangesForMergeElapsedMicroseconds; extern const Event MergerMutatorSelectPartsForMergeElapsedMicroseconds; From 15337692e68961c247dd809f3b13e89a8acc74b7 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Wed, 6 Nov 2024 15:10:10 +0000 Subject: [PATCH 477/680] Minor: Remove "experimental" mention of analyzer --- src/Core/Settings.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 081e07ca2ce..7e8d0aabce0 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -4239,7 +4239,7 @@ Rewrite aggregate functions with if expression as argument when logically equiva For example, `avg(if(cond, col, null))` can be rewritten to `avgOrNullIf(cond, col)`. It may improve performance. :::note -Supported only with experimental analyzer (`enable_analyzer = 1`). +Supported only with the analyzer (`enable_analyzer = 1`). ::: )", 0) \ DECLARE(Bool, optimize_rewrite_array_exists_to_has, false, R"( From 020b69647a65dd740cddfbf62730f37de14a4eb8 Mon Sep 17 00:00:00 2001 From: avogar Date: Wed, 6 Nov 2024 15:15:29 +0000 Subject: [PATCH 478/680] Fix counting column size in wide part for Dynamid and JSON types --- .../MergeTree/MergeTreeDataPartWide.cpp | 2 +- .../MergeTree/MergeTreeReaderWide.cpp | 2 +- ...umn_sizes_with_dynamic_structure.reference | 1 + ...62_column_sizes_with_dynamic_structure.sql | 22 +++++++++++++++++++ 4 files changed, 25 insertions(+), 2 deletions(-) create mode 100644 tests/queries/0_stateless/03262_column_sizes_with_dynamic_structure.reference create mode 100644 tests/queries/0_stateless/03262_column_sizes_with_dynamic_structure.sql diff --git a/src/Storages/MergeTree/MergeTreeDataPartWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWide.cpp index d6f213463f2..d8470ba8405 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWide.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWide.cpp @@ -108,7 +108,7 @@ ColumnSize MergeTreeDataPartWide::getColumnSizeImpl( auto mrk_checksum = checksums.files.find(*stream_name + getMarksFileExtension()); if (mrk_checksum != checksums.files.end()) size.marks += mrk_checksum->second.file_size; - }); + }, column.type, getColumnSample(column)); return size; } diff --git a/src/Storages/MergeTree/MergeTreeReaderWide.cpp b/src/Storages/MergeTree/MergeTreeReaderWide.cpp index 77231d8d392..885bd1ded8c 100644 --- a/src/Storages/MergeTree/MergeTreeReaderWide.cpp +++ b/src/Storages/MergeTree/MergeTreeReaderWide.cpp @@ -172,7 +172,7 @@ size_t MergeTreeReaderWide::readRows( throw; } - if (column->empty()) + if (column->empty() && max_rows_to_read > 0) res_columns[pos] = nullptr; } diff --git a/tests/queries/0_stateless/03262_column_sizes_with_dynamic_structure.reference b/tests/queries/0_stateless/03262_column_sizes_with_dynamic_structure.reference new file mode 100644 index 00000000000..5cab16ed96d --- /dev/null +++ b/tests/queries/0_stateless/03262_column_sizes_with_dynamic_structure.reference @@ -0,0 +1 @@ +test 10.00 million 352.87 MiB 39.43 MiB 39.45 MiB diff --git a/tests/queries/0_stateless/03262_column_sizes_with_dynamic_structure.sql b/tests/queries/0_stateless/03262_column_sizes_with_dynamic_structure.sql new file mode 100644 index 00000000000..21e6515fc99 --- /dev/null +++ b/tests/queries/0_stateless/03262_column_sizes_with_dynamic_structure.sql @@ -0,0 +1,22 @@ +-- Tags: no-random-settings + +set allow_experimental_dynamic_type = 1; +set allow_experimental_json_type = 1; + +drop table if exists test; +create table test (d Dynamic, json JSON) engine=MergeTree order by tuple() settings min_rows_for_wide_part=0, min_bytes_for_wide_part=1; +insert into test select number, '{"a" : 42, "b" : "Hello, World"}' from numbers(10000000); + +SELECT + `table`, + formatReadableQuantity(sum(rows)) AS rows, + formatReadableSize(sum(data_uncompressed_bytes)) AS data_size_uncompressed, + formatReadableSize(sum(data_compressed_bytes)) AS data_size_compressed, + formatReadableSize(sum(bytes_on_disk)) AS total_size_on_disk +FROM system.parts +WHERE active AND (database = currentDatabase()) AND (`table` = 'test') +GROUP BY `table` +ORDER BY `table` ASC; + +drop table test; + From 12ab488453796a46f1f37d91cf60c6a6007e0134 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Wed, 6 Nov 2024 16:20:57 +0100 Subject: [PATCH 479/680] Revert "Selection of hash join inner table" --- src/Core/Joins.h | 11 - src/Core/Settings.cpp | 3 - src/Core/Settings.h | 1 - src/Core/SettingsEnums.cpp | 4 - src/Core/SettingsEnums.h | 2 +- src/Interpreters/ConcurrentHashJoin.h | 11 - src/Interpreters/FullSortingMergeJoin.h | 2 +- src/Interpreters/HashJoin/HashJoin.cpp | 16 +- src/Interpreters/HashJoin/HashJoin.h | 5 +- .../HashJoin/HashJoinMethodsImpl.h | 18 +- src/Interpreters/InterpreterSelectQuery.cpp | 4 +- src/Interpreters/TableJoin.cpp | 56 +---- src/Interpreters/TableJoin.h | 19 +- src/Interpreters/TreeRewriter.cpp | 5 +- src/Parsers/CreateQueryUUIDs.cpp | 2 +- src/Planner/CollectColumnIdentifiers.cpp | 1 - src/Planner/PlannerJoinTree.cpp | 152 +++++-------- src/Processors/QueryPlan/JoinStep.cpp | 103 +-------- src/Processors/QueryPlan/JoinStep.h | 17 +- .../QueryPlan/Optimizations/Optimizations.h | 1 - .../QueryPlan/Optimizations/optimizeJoin.cpp | 102 --------- .../QueryPlan/Optimizations/optimizeTree.cpp | 3 - .../QueryPlan/ReadFromMemoryStorageStep.h | 2 - .../Transforms/ColumnPermuteTransform.cpp | 49 ----- .../Transforms/ColumnPermuteTransform.h | 30 --- .../Transforms/JoiningTransform.cpp | 1 - tests/clickhouse-test | 4 - tests/integration/helpers/cluster.py | 13 +- tests/integration/helpers/random_settings.py | 2 - .../test_peak_memory_usage/test.py | 2 +- .../0_stateless/00826_cross_to_inner_join.sql | 13 +- .../00847_multiple_join_same_column.sql | 14 +- .../01015_empty_in_inner_right_join.sql.j2 | 2 - .../01107_join_right_table_totals.reference | 7 - .../01107_join_right_table_totals.sql | 10 +- .../01763_filter_push_down_bugs.reference | 2 +- .../01881_join_on_conditions_hash.sql.j2 | 10 +- .../0_stateless/02000_join_on_const.reference | 18 +- .../0_stateless/02000_join_on_const.sql | 16 +- .../02001_join_on_const_bs_long.sql.j2 | 4 +- ...oin_with_nullable_lowcardinality_crash.sql | 5 +- .../0_stateless/02282_array_distance.sql | 12 +- .../02381_join_dup_columns_in_plan.reference | 1 + .../0_stateless/02461_join_lc_issue_42380.sql | 3 +- ...emove_redundant_sorting_analyzer.reference | 4 +- ...move_redundant_distinct_analyzer.reference | 18 +- .../02514_analyzer_drop_join_on.reference | 55 +++-- .../02514_analyzer_drop_join_on.sql | 1 - ...oin_with_totals_and_subquery_bug.reference | 2 +- .../02835_join_step_explain.reference | 32 +-- .../0_stateless/02835_join_step_explain.sql | 2 - .../02962_join_using_bug_57894.reference | 1 - .../02962_join_using_bug_57894.sql | 2 - ...filter_push_down_equivalent_sets.reference | 206 ++++++++---------- ..._join_filter_push_down_equivalent_sets.sql | 40 +--- .../03038_recursive_cte_postgres_4.reference | 4 +- .../03038_recursive_cte_postgres_4.sql | 4 +- .../0_stateless/03094_one_thousand_joins.sql | 1 - ...convert_outer_join_to_inner_join.reference | 36 +-- ...03130_convert_outer_join_to_inner_join.sql | 13 +- ...ter_push_down_equivalent_columns.reference | 3 +- .../03236_squashing_high_memory.sql | 1 - 62 files changed, 314 insertions(+), 869 deletions(-) delete mode 100644 src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp delete mode 100644 src/Processors/Transforms/ColumnPermuteTransform.cpp delete mode 100644 src/Processors/Transforms/ColumnPermuteTransform.h diff --git a/src/Core/Joins.h b/src/Core/Joins.h index dd6d86fc902..0964bf86e6b 100644 --- a/src/Core/Joins.h +++ b/src/Core/Joins.h @@ -119,15 +119,4 @@ enum class JoinTableSide : uint8_t const char * toString(JoinTableSide join_table_side); -/// Setting to choose which table to use as the inner table in hash join -enum class JoinInnerTableSelectionMode : uint8_t -{ - /// Use left table - Left, - /// Use right table - Right, - /// Use the table with the smallest number of rows - Auto, -}; - } diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 081e07ca2ce..ada6b674c87 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -1912,9 +1912,6 @@ See also: For single JOIN in case of identifier ambiguity prefer left table )", IMPORTANT) \ \ - DECLARE(JoinInnerTableSelectionMode, query_plan_join_inner_table_selection, JoinInnerTableSelectionMode::Auto, R"( -Select the side of the join to be the inner table in the query plan. Supported only for `ALL` join strictness with `JOIN ON` clause. Possible values: 'auto', 'left', 'right'. -)", 0) \ DECLARE(UInt64, preferred_block_size_bytes, 1000000, R"( This setting adjusts the data block size for query processing and represents additional fine-tuning to the more rough 'max_block_size' setting. If the columns are large and with 'max_block_size' rows the block size is likely to be larger than the specified amount of bytes, its size will be lowered for better CPU cache locality. )", 0) \ diff --git a/src/Core/Settings.h b/src/Core/Settings.h index 1cc58deb94a..ac3b1fe651e 100644 --- a/src/Core/Settings.h +++ b/src/Core/Settings.h @@ -66,7 +66,6 @@ class WriteBuffer; M(CLASS_NAME, IntervalOutputFormat) \ M(CLASS_NAME, JoinAlgorithm) \ M(CLASS_NAME, JoinStrictness) \ - M(CLASS_NAME, JoinInnerTableSelectionMode) \ M(CLASS_NAME, LightweightMutationProjectionMode) \ M(CLASS_NAME, LoadBalancing) \ M(CLASS_NAME, LocalFSReadMethod) \ diff --git a/src/Core/SettingsEnums.cpp b/src/Core/SettingsEnums.cpp index 89e9cb295c3..cef63039277 100644 --- a/src/Core/SettingsEnums.cpp +++ b/src/Core/SettingsEnums.cpp @@ -55,10 +55,6 @@ IMPLEMENT_SETTING_MULTI_ENUM(JoinAlgorithm, ErrorCodes::UNKNOWN_JOIN, {"full_sorting_merge", JoinAlgorithm::FULL_SORTING_MERGE}, {"grace_hash", JoinAlgorithm::GRACE_HASH}}) -IMPLEMENT_SETTING_ENUM(JoinInnerTableSelectionMode, ErrorCodes::BAD_ARGUMENTS, - {{"left", JoinInnerTableSelectionMode::Left}, - {"right", JoinInnerTableSelectionMode::Right}, - {"auto", JoinInnerTableSelectionMode::Auto}}) IMPLEMENT_SETTING_ENUM(TotalsMode, ErrorCodes::UNKNOWN_TOTALS_MODE, {{"before_having", TotalsMode::BEFORE_HAVING}, diff --git a/src/Core/SettingsEnums.h b/src/Core/SettingsEnums.h index 35bdb8a7f65..607011b505b 100644 --- a/src/Core/SettingsEnums.h +++ b/src/Core/SettingsEnums.h @@ -128,8 +128,8 @@ constexpr auto getEnumValues(); DECLARE_SETTING_ENUM(LoadBalancing) DECLARE_SETTING_ENUM(JoinStrictness) + DECLARE_SETTING_MULTI_ENUM(JoinAlgorithm) -DECLARE_SETTING_ENUM(JoinInnerTableSelectionMode) /// Which rows should be included in TOTALS. diff --git a/src/Interpreters/ConcurrentHashJoin.h b/src/Interpreters/ConcurrentHashJoin.h index b377727a134..a911edaccc3 100644 --- a/src/Interpreters/ConcurrentHashJoin.h +++ b/src/Interpreters/ConcurrentHashJoin.h @@ -60,17 +60,6 @@ public: IBlocksStreamPtr getNonJoinedBlocks(const Block & left_sample_block, const Block & result_sample_block, UInt64 max_block_size) const override; - - bool isCloneSupported() const override - { - return !getTotals() && getTotalRowCount() == 0; - } - - std::shared_ptr clone(const std::shared_ptr & table_join_, const Block &, const Block & right_sample_block_) const override - { - return std::make_shared(context, table_join_, slots, right_sample_block_, stats_collecting_params); - } - private: struct InternalHashJoin { diff --git a/src/Interpreters/FullSortingMergeJoin.h b/src/Interpreters/FullSortingMergeJoin.h index faa9114c618..3f1e0d59287 100644 --- a/src/Interpreters/FullSortingMergeJoin.h +++ b/src/Interpreters/FullSortingMergeJoin.h @@ -36,7 +36,7 @@ public: bool isCloneSupported() const override { - return !getTotals(); + return true; } std::shared_ptr clone(const std::shared_ptr & table_join_, diff --git a/src/Interpreters/HashJoin/HashJoin.cpp b/src/Interpreters/HashJoin/HashJoin.cpp index dad8a487745..3e7f3deea8b 100644 --- a/src/Interpreters/HashJoin/HashJoin.cpp +++ b/src/Interpreters/HashJoin/HashJoin.cpp @@ -383,16 +383,6 @@ size_t HashJoin::getTotalByteCount() const return res; } -bool HashJoin::isUsedByAnotherAlgorithm() const -{ - return table_join->isEnabledAlgorithm(JoinAlgorithm::AUTO) || table_join->isEnabledAlgorithm(JoinAlgorithm::GRACE_HASH); -} - -bool HashJoin::canRemoveColumnsFromLeftBlock() const -{ - return table_join->enableEnalyzer() && !table_join->hasUsing() && !isUsedByAnotherAlgorithm(); -} - void HashJoin::initRightBlockStructure(Block & saved_block_sample) { if (isCrossOrComma(kind)) @@ -404,7 +394,8 @@ void HashJoin::initRightBlockStructure(Block & saved_block_sample) bool multiple_disjuncts = !table_join->oneDisjunct(); /// We could remove key columns for LEFT | INNER HashJoin but we should keep them for JoinSwitcher (if any). - bool save_key_columns = isUsedByAnotherAlgorithm() || + bool save_key_columns = table_join->isEnabledAlgorithm(JoinAlgorithm::AUTO) || + table_join->isEnabledAlgorithm(JoinAlgorithm::GRACE_HASH) || isRightOrFull(kind) || multiple_disjuncts || table_join->getMixedJoinExpression(); @@ -1237,10 +1228,7 @@ IBlocksStreamPtr HashJoin::getNonJoinedBlocks(const Block & left_sample_block, { if (!JoinCommon::hasNonJoinedBlocks(*table_join)) return {}; - size_t left_columns_count = left_sample_block.columns(); - if (canRemoveColumnsFromLeftBlock()) - left_columns_count = table_join->getOutputColumns(JoinTableSide::Left).size(); bool flag_per_row = needUsedFlagsForPerRightTableRow(table_join); if (!flag_per_row) diff --git a/src/Interpreters/HashJoin/HashJoin.h b/src/Interpreters/HashJoin/HashJoin.h index 8a27961354a..4c1ebbcdc66 100644 --- a/src/Interpreters/HashJoin/HashJoin.h +++ b/src/Interpreters/HashJoin/HashJoin.h @@ -127,7 +127,7 @@ public: bool isCloneSupported() const override { - return !getTotals() && getTotalRowCount() == 0; + return true; } std::shared_ptr clone(const std::shared_ptr & table_join_, @@ -464,9 +464,6 @@ private: bool empty() const; - bool isUsedByAnotherAlgorithm() const; - bool canRemoveColumnsFromLeftBlock() const; - void validateAdditionalFilterExpression(std::shared_ptr additional_filter_expression); bool needUsedFlagsForPerRightTableRow(std::shared_ptr table_join_) const; diff --git a/src/Interpreters/HashJoin/HashJoinMethodsImpl.h b/src/Interpreters/HashJoin/HashJoinMethodsImpl.h index 7e8a2658b9c..45a766e2df6 100644 --- a/src/Interpreters/HashJoin/HashJoinMethodsImpl.h +++ b/src/Interpreters/HashJoin/HashJoinMethodsImpl.h @@ -56,6 +56,7 @@ Block HashJoinMethods::joinBlockImpl( const auto & key_names = !is_join_get ? onexprs[i].key_names_left : onexprs[i].key_names_right; join_on_keys.emplace_back(block, key_names, onexprs[i].condColumnNames().first, join.key_sizes[i]); } + size_t existing_columns = block.columns(); /** If you use FULL or RIGHT JOIN, then the columns from the "left" table must be materialized. * Because if they are constants, then in the "not joined" rows, they may have different values @@ -98,22 +99,6 @@ Block HashJoinMethods::joinBlockImpl( added_columns.buildJoinGetOutput(); else added_columns.buildOutput(); - - const auto & table_join = join.table_join; - std::set block_columns_to_erase; - if (join.canRemoveColumnsFromLeftBlock()) - { - std::unordered_set left_output_columns; - for (const auto & out_column : table_join->getOutputColumns(JoinTableSide::Left)) - left_output_columns.insert(out_column.name); - for (size_t i = 0; i < block.columns(); ++i) - { - if (!left_output_columns.contains(block.getByPosition(i).name)) - block_columns_to_erase.insert(i); - } - } - size_t existing_columns = block.columns(); - for (size_t i = 0; i < added_columns.size(); ++i) block.insert(added_columns.moveColumn(i)); @@ -175,7 +160,6 @@ Block HashJoinMethods::joinBlockImpl( block.safeGetByPosition(pos).column = block.safeGetByPosition(pos).column->replicate(*offsets_to_replicate); } } - block.erase(block_columns_to_erase); return remaining_block; } diff --git a/src/Interpreters/InterpreterSelectQuery.cpp b/src/Interpreters/InterpreterSelectQuery.cpp index 8ddf51fa25e..3918c1c37ea 100644 --- a/src/Interpreters/InterpreterSelectQuery.cpp +++ b/src/Interpreters/InterpreterSelectQuery.cpp @@ -1888,9 +1888,7 @@ void InterpreterSelectQuery::executeImpl(QueryPlan & query_plan, std::optional

setStepDescription(fmt::format("JOIN {}", expressions.join->pipelineType())); std::vector plans; diff --git a/src/Interpreters/TableJoin.cpp b/src/Interpreters/TableJoin.cpp index 555aaff2e06..2532dddba3c 100644 --- a/src/Interpreters/TableJoin.cpp +++ b/src/Interpreters/TableJoin.cpp @@ -41,7 +41,6 @@ namespace DB namespace Setting { extern const SettingsBool allow_experimental_join_right_table_sorting; - extern const SettingsBool allow_experimental_analyzer; extern const SettingsUInt64 cross_join_min_bytes_to_compress; extern const SettingsUInt64 cross_join_min_rows_to_compress; extern const SettingsUInt64 default_max_bytes_in_join; @@ -144,7 +143,6 @@ TableJoin::TableJoin(const Settings & settings, VolumePtr tmp_volume_, Temporary , max_memory_usage(settings[Setting::max_memory_usage]) , tmp_volume(tmp_volume_) , tmp_data(tmp_data_) - , enable_analyzer(settings[Setting::allow_experimental_analyzer]) { } @@ -163,8 +161,6 @@ void TableJoin::resetCollected() clauses.clear(); columns_from_joined_table.clear(); columns_added_by_join.clear(); - columns_from_left_table.clear(); - result_columns_from_left_table.clear(); original_names.clear(); renames.clear(); left_type_map.clear(); @@ -207,20 +203,6 @@ size_t TableJoin::rightKeyInclusion(const String & name) const return count; } -void TableJoin::setInputColumns(NamesAndTypesList left_output_columns, NamesAndTypesList right_output_columns) -{ - columns_from_left_table = std::move(left_output_columns); - columns_from_joined_table = std::move(right_output_columns); -} - - -const NamesAndTypesList & TableJoin::getOutputColumns(JoinTableSide side) -{ - if (side == JoinTableSide::Left) - return result_columns_from_left_table; - return columns_added_by_join; -} - void TableJoin::deduplicateAndQualifyColumnNames(const NameSet & left_table_columns, const String & right_table_prefix) { NameSet joined_columns; @@ -369,18 +351,9 @@ bool TableJoin::rightBecomeNullable(const DataTypePtr & column_type) const return forceNullableRight() && JoinCommon::canBecomeNullable(column_type); } -void TableJoin::setUsedColumn(const NameAndTypePair & joined_column, JoinTableSide side) -{ - if (side == JoinTableSide::Left) - result_columns_from_left_table.push_back(joined_column); - else - columns_added_by_join.push_back(joined_column); - -} - void TableJoin::addJoinedColumn(const NameAndTypePair & joined_column) { - setUsedColumn(joined_column, JoinTableSide::Right); + columns_added_by_join.emplace_back(joined_column); } NamesAndTypesList TableJoin::correctedColumnsAddedByJoin() const @@ -1022,32 +995,5 @@ size_t TableJoin::getMaxMemoryUsage() const return max_memory_usage; } -void TableJoin::swapSides() -{ - assertEnableEnalyzer(); - - std::swap(key_asts_left, key_asts_right); - std::swap(left_type_map, right_type_map); - for (auto & clause : clauses) - { - std::swap(clause.key_names_left, clause.key_names_right); - std::swap(clause.on_filter_condition_left, clause.on_filter_condition_right); - std::swap(clause.analyzer_left_filter_condition_column_name, clause.analyzer_right_filter_condition_column_name); - } - - std::swap(columns_from_left_table, columns_from_joined_table); - std::swap(result_columns_from_left_table, columns_added_by_join); - - if (table_join.kind == JoinKind::Left) - table_join.kind = JoinKind::Right; - else if (table_join.kind == JoinKind::Right) - table_join.kind = JoinKind::Left; -} - -void TableJoin::assertEnableEnalyzer() const -{ - if (!enable_analyzer) - throw DB::Exception(ErrorCodes::NOT_IMPLEMENTED, "TableJoin: analyzer is disabled"); -} } diff --git a/src/Interpreters/TableJoin.h b/src/Interpreters/TableJoin.h index e0e1926fb12..e1bae55a4ed 100644 --- a/src/Interpreters/TableJoin.h +++ b/src/Interpreters/TableJoin.h @@ -167,9 +167,6 @@ private: ASOFJoinInequality asof_inequality = ASOFJoinInequality::GreaterOrEquals; - NamesAndTypesList columns_from_left_table; - NamesAndTypesList result_columns_from_left_table; - /// All columns which can be read from joined table. Duplicating names are qualified. NamesAndTypesList columns_from_joined_table; /// Columns will be added to block by JOIN. @@ -205,8 +202,6 @@ private: bool is_join_with_constant = false; - bool enable_analyzer = false; - Names requiredJoinedNames() const; /// Create converting actions and change key column names if required @@ -271,8 +266,6 @@ public: VolumePtr getGlobalTemporaryVolume() { return tmp_volume; } TemporaryDataOnDiskScopePtr getTempDataOnDisk() { return tmp_data; } - bool enableEnalyzer() const { return enable_analyzer; } - void assertEnableEnalyzer() const; ActionsDAG createJoinedBlockActions(ContextPtr context) const; @@ -289,7 +282,6 @@ public: } bool allowParallelHashJoin() const; - void swapSides(); bool joinUseNulls() const { return join_use_nulls; } @@ -380,9 +372,6 @@ public: bool leftBecomeNullable(const DataTypePtr & column_type) const; bool rightBecomeNullable(const DataTypePtr & column_type) const; void addJoinedColumn(const NameAndTypePair & joined_column); - - void setUsedColumn(const NameAndTypePair & joined_column, JoinTableSide side); - void setColumnsAddedByJoin(const NamesAndTypesList & columns_added_by_join_value) { columns_added_by_join = columns_added_by_join_value; @@ -408,17 +397,11 @@ public: ASTPtr leftKeysList() const; ASTPtr rightKeysList() const; /// For ON syntax only - void setColumnsFromJoinedTable(NamesAndTypesList columns_from_joined_table_value, const NameSet & left_table_columns, const String & right_table_prefix, const NamesAndTypesList & columns_from_left_table_) + void setColumnsFromJoinedTable(NamesAndTypesList columns_from_joined_table_value, const NameSet & left_table_columns, const String & right_table_prefix) { columns_from_joined_table = std::move(columns_from_joined_table_value); deduplicateAndQualifyColumnNames(left_table_columns, right_table_prefix); - result_columns_from_left_table = columns_from_left_table_; - columns_from_left_table = columns_from_left_table_; } - - void setInputColumns(NamesAndTypesList left_output_columns, NamesAndTypesList right_output_columns); - const NamesAndTypesList & getOutputColumns(JoinTableSide side); - const NamesAndTypesList & columnsFromJoinedTable() const { return columns_from_joined_table; } const NamesAndTypesList & columnsAddedByJoin() const { return columns_added_by_join; } diff --git a/src/Interpreters/TreeRewriter.cpp b/src/Interpreters/TreeRewriter.cpp index 28e11166762..ea08fd92339 100644 --- a/src/Interpreters/TreeRewriter.cpp +++ b/src/Interpreters/TreeRewriter.cpp @@ -1353,15 +1353,12 @@ TreeRewriterResultPtr TreeRewriter::analyzeSelect( if (tables_with_columns.size() > 1) { - auto columns_from_left_table = tables_with_columns[0].columns; const auto & right_table = tables_with_columns[1]; auto columns_from_joined_table = right_table.columns; /// query can use materialized or aliased columns from right joined table, /// we want to request it for right table columns_from_joined_table.insert(columns_from_joined_table.end(), right_table.hidden_columns.begin(), right_table.hidden_columns.end()); - columns_from_left_table.insert(columns_from_left_table.end(), tables_with_columns[0].hidden_columns.begin(), tables_with_columns[0].hidden_columns.end()); - result.analyzed_join->setColumnsFromJoinedTable( - std::move(columns_from_joined_table), source_columns_set, right_table.table.getQualifiedNamePrefix(), columns_from_left_table); + result.analyzed_join->setColumnsFromJoinedTable(std::move(columns_from_joined_table), source_columns_set, right_table.table.getQualifiedNamePrefix()); } translateQualifiedNames(query, *select_query, source_columns_set, tables_with_columns); diff --git a/src/Parsers/CreateQueryUUIDs.cpp b/src/Parsers/CreateQueryUUIDs.cpp index 70848440a0e..c788cc7a025 100644 --- a/src/Parsers/CreateQueryUUIDs.cpp +++ b/src/Parsers/CreateQueryUUIDs.cpp @@ -31,7 +31,7 @@ CreateQueryUUIDs::CreateQueryUUIDs(const ASTCreateQuery & query, bool generate_r /// If we generate random UUIDs for already existing tables then those UUIDs will not be correct making those inner target table inaccessible. /// Thus it's not safe for example to replace /// "ATTACH MATERIALIZED VIEW mv AS SELECT a FROM b" with - /// "ATTACH MATERIALIZED VIEW mv TO INNER UUID '123e4567-e89b-12d3-a456-426614174000' AS SELECT a FROM b" + /// "ATTACH MATERIALIZED VIEW mv TO INNER UUID "XXXX" AS SELECT a FROM b" /// This replacement is safe only for CREATE queries when inner target tables don't exist yet. if (!query.attach) { diff --git a/src/Planner/CollectColumnIdentifiers.cpp b/src/Planner/CollectColumnIdentifiers.cpp index dd5bdd4d141..95f1c7d53d8 100644 --- a/src/Planner/CollectColumnIdentifiers.cpp +++ b/src/Planner/CollectColumnIdentifiers.cpp @@ -2,7 +2,6 @@ #include #include -#include #include diff --git a/src/Planner/PlannerJoinTree.cpp b/src/Planner/PlannerJoinTree.cpp index a1ce455f266..5c153f6db39 100644 --- a/src/Planner/PlannerJoinTree.cpp +++ b/src/Planner/PlannerJoinTree.cpp @@ -104,7 +104,6 @@ namespace Setting extern const SettingsBool optimize_move_to_prewhere; extern const SettingsBool optimize_move_to_prewhere_if_final; extern const SettingsBool use_concurrency_control; - extern const SettingsJoinInnerTableSelectionMode query_plan_join_inner_table_selection; } namespace ErrorCodes @@ -1242,55 +1241,6 @@ void joinCastPlanColumnsToNullable(QueryPlan & plan_to_add_cast, PlannerContextP plan_to_add_cast.addStep(std::move(cast_join_columns_step)); } -std::optional createStepToDropColumns( - const Block & header, - const ColumnIdentifierSet & outer_scope_columns, - const PlannerContextPtr & planner_context) -{ - ActionsDAG drop_unused_columns_after_join_actions_dag(header.getColumnsWithTypeAndName()); - ActionsDAG::NodeRawConstPtrs drop_unused_columns_after_join_actions_dag_updated_outputs; - std::unordered_set drop_unused_columns_after_join_actions_dag_updated_outputs_names; - std::optional first_skipped_column_node_index; - - auto & drop_unused_columns_after_join_actions_dag_outputs = drop_unused_columns_after_join_actions_dag.getOutputs(); - size_t drop_unused_columns_after_join_actions_dag_outputs_size = drop_unused_columns_after_join_actions_dag_outputs.size(); - - const auto & global_planner_context = planner_context->getGlobalPlannerContext(); - - for (size_t i = 0; i < drop_unused_columns_after_join_actions_dag_outputs_size; ++i) - { - const auto & output = drop_unused_columns_after_join_actions_dag_outputs[i]; - - if (drop_unused_columns_after_join_actions_dag_updated_outputs_names.contains(output->result_name) - || !global_planner_context->hasColumnIdentifier(output->result_name)) - continue; - - if (!outer_scope_columns.contains(output->result_name)) - { - if (!first_skipped_column_node_index) - first_skipped_column_node_index = i; - continue; - } - - drop_unused_columns_after_join_actions_dag_updated_outputs.push_back(output); - drop_unused_columns_after_join_actions_dag_updated_outputs_names.insert(output->result_name); - } - - if (!first_skipped_column_node_index) - return {}; - - /** It is expected that JOIN TREE query plan will contain at least 1 column, even if there are no columns in outer scope. - * - * Example: SELECT count() FROM test_table_1 AS t1, test_table_2 AS t2; - */ - if (drop_unused_columns_after_join_actions_dag_updated_outputs.empty() && first_skipped_column_node_index) - drop_unused_columns_after_join_actions_dag_updated_outputs.push_back(drop_unused_columns_after_join_actions_dag_outputs[*first_skipped_column_node_index]); - - drop_unused_columns_after_join_actions_dag_outputs = std::move(drop_unused_columns_after_join_actions_dag_updated_outputs); - - return drop_unused_columns_after_join_actions_dag; -} - JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_expression, JoinTreeQueryPlan left_join_tree_query_plan, JoinTreeQueryPlan right_join_tree_query_plan, @@ -1563,37 +1513,21 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ } const Block & left_header = left_plan.getCurrentHeader(); + auto left_table_names = left_header.getNames(); + NameSet left_table_names_set(left_table_names.begin(), left_table_names.end()); + + auto columns_from_joined_table = right_plan.getCurrentHeader().getNamesAndTypesList(); + table_join->setColumnsFromJoinedTable(columns_from_joined_table, left_table_names_set, ""); + + for (auto & column_from_joined_table : columns_from_joined_table) + { + /// Add columns from joined table only if they are presented in outer scope, otherwise they can be dropped + if (planner_context->getGlobalPlannerContext()->hasColumnIdentifier(column_from_joined_table.name) && + outer_scope_columns.contains(column_from_joined_table.name)) + table_join->addJoinedColumn(column_from_joined_table); + } + const Block & right_header = right_plan.getCurrentHeader(); - - auto columns_from_left_table = left_header.getNamesAndTypesList(); - auto columns_from_right_table = right_header.getNamesAndTypesList(); - - table_join->setInputColumns(columns_from_left_table, columns_from_right_table); - - for (auto & column_from_joined_table : columns_from_left_table) - { - /// Add columns to output only if they are presented in outer scope, otherwise they can be dropped - if (planner_context->getGlobalPlannerContext()->hasColumnIdentifier(column_from_joined_table.name) && - outer_scope_columns.contains(column_from_joined_table.name)) - table_join->setUsedColumn(column_from_joined_table, JoinTableSide::Left); - } - - for (auto & column_from_joined_table : columns_from_right_table) - { - /// Add columns to output only if they are presented in outer scope, otherwise they can be dropped - if (planner_context->getGlobalPlannerContext()->hasColumnIdentifier(column_from_joined_table.name) && - outer_scope_columns.contains(column_from_joined_table.name)) - table_join->setUsedColumn(column_from_joined_table, JoinTableSide::Right); - } - - if (table_join->getOutputColumns(JoinTableSide::Left).empty() && table_join->getOutputColumns(JoinTableSide::Right).empty()) - { - if (!columns_from_left_table.empty()) - table_join->setUsedColumn(columns_from_left_table.front(), JoinTableSide::Left); - else if (!columns_from_right_table.empty()) - table_join->setUsedColumn(columns_from_right_table.front(), JoinTableSide::Right); - } - auto join_algorithm = chooseJoinAlgorithm(table_join, join_node.getRightTableExpression(), left_header, right_header, planner_context); auto result_plan = QueryPlan(); @@ -1681,26 +1615,13 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ } auto join_pipeline_type = join_algorithm->pipelineType(); - - ColumnIdentifierSet outer_scope_columns_nonempty; - if (outer_scope_columns.empty()) - { - if (left_header.columns() > 1) - outer_scope_columns_nonempty.insert(left_header.getByPosition(0).name); - else if (right_header.columns() > 1) - outer_scope_columns_nonempty.insert(right_header.getByPosition(0).name); - } - auto join_step = std::make_unique( left_plan.getCurrentHeader(), right_plan.getCurrentHeader(), std::move(join_algorithm), settings[Setting::max_block_size], settings[Setting::max_threads], - outer_scope_columns.empty() ? outer_scope_columns_nonempty : outer_scope_columns, - false /*optimize_read_in_order*/, - true /*optimize_skip_unused_shards*/); - join_step->inner_table_selection_mode = settings[Setting::query_plan_join_inner_table_selection]; + false /*optimize_read_in_order*/); join_step->setStepDescription(fmt::format("JOIN {}", join_pipeline_type)); @@ -1711,18 +1632,47 @@ JoinTreeQueryPlan buildQueryPlanForJoinNode(const QueryTreeNodePtr & join_table_ result_plan.unitePlans(std::move(join_step), {std::move(plans)}); } - const auto & header_after_join = result_plan.getCurrentHeader(); - if (header_after_join.columns() > outer_scope_columns.size()) + ActionsDAG drop_unused_columns_after_join_actions_dag(result_plan.getCurrentHeader().getColumnsWithTypeAndName()); + ActionsDAG::NodeRawConstPtrs drop_unused_columns_after_join_actions_dag_updated_outputs; + std::unordered_set drop_unused_columns_after_join_actions_dag_updated_outputs_names; + std::optional first_skipped_column_node_index; + + auto & drop_unused_columns_after_join_actions_dag_outputs = drop_unused_columns_after_join_actions_dag.getOutputs(); + size_t drop_unused_columns_after_join_actions_dag_outputs_size = drop_unused_columns_after_join_actions_dag_outputs.size(); + + for (size_t i = 0; i < drop_unused_columns_after_join_actions_dag_outputs_size; ++i) { - auto drop_unused_columns_after_join_actions_dag = createStepToDropColumns(header_after_join, outer_scope_columns, planner_context); - if (drop_unused_columns_after_join_actions_dag) + const auto & output = drop_unused_columns_after_join_actions_dag_outputs[i]; + + const auto & global_planner_context = planner_context->getGlobalPlannerContext(); + if (drop_unused_columns_after_join_actions_dag_updated_outputs_names.contains(output->result_name) + || !global_planner_context->hasColumnIdentifier(output->result_name)) + continue; + + if (!outer_scope_columns.contains(output->result_name)) { - auto drop_unused_columns_after_join_transform_step = std::make_unique(result_plan.getCurrentHeader(), std::move(*drop_unused_columns_after_join_actions_dag)); - drop_unused_columns_after_join_transform_step->setStepDescription("Drop unused columns after JOIN"); - result_plan.addStep(std::move(drop_unused_columns_after_join_transform_step)); + if (!first_skipped_column_node_index) + first_skipped_column_node_index = i; + continue; } + + drop_unused_columns_after_join_actions_dag_updated_outputs.push_back(output); + drop_unused_columns_after_join_actions_dag_updated_outputs_names.insert(output->result_name); } + /** It is expected that JOIN TREE query plan will contain at least 1 column, even if there are no columns in outer scope. + * + * Example: SELECT count() FROM test_table_1 AS t1, test_table_2 AS t2; + */ + if (drop_unused_columns_after_join_actions_dag_updated_outputs.empty() && first_skipped_column_node_index) + drop_unused_columns_after_join_actions_dag_updated_outputs.push_back(drop_unused_columns_after_join_actions_dag_outputs[*first_skipped_column_node_index]); + + drop_unused_columns_after_join_actions_dag_outputs = std::move(drop_unused_columns_after_join_actions_dag_updated_outputs); + + auto drop_unused_columns_after_join_transform_step = std::make_unique(result_plan.getCurrentHeader(), std::move(drop_unused_columns_after_join_actions_dag)); + drop_unused_columns_after_join_transform_step->setStepDescription("DROP unused columns after JOIN"); + result_plan.addStep(std::move(drop_unused_columns_after_join_transform_step)); + for (const auto & right_join_tree_query_plan_row_policy : right_join_tree_query_plan.used_row_policies) left_join_tree_query_plan.used_row_policies.insert(right_join_tree_query_plan_row_policy); diff --git a/src/Processors/QueryPlan/JoinStep.cpp b/src/Processors/QueryPlan/JoinStep.cpp index 7ade437822e..018b52a5c68 100644 --- a/src/Processors/QueryPlan/JoinStep.cpp +++ b/src/Processors/QueryPlan/JoinStep.cpp @@ -6,7 +6,6 @@ #include #include #include -#include namespace DB { @@ -37,37 +36,6 @@ std::vector> describeJoinActions(const JoinPtr & join) return description; } -std::vector getPermutationForBlock( - const Block & block, - const Block & lhs_block, - const Block & rhs_block, - const NameSet & name_filter) -{ - std::vector permutation; - permutation.reserve(block.columns()); - Block::NameMap name_map = block.getNamesToIndexesMap(); - - bool is_trivial = true; - for (const auto & other_block : {lhs_block, rhs_block}) - { - for (const auto & col : other_block) - { - if (!name_filter.contains(col.name)) - continue; - if (auto it = name_map.find(col.name); it != name_map.end()) - { - is_trivial = is_trivial && it->second == permutation.size(); - permutation.push_back(it->second); - } - } - } - - if (is_trivial && permutation.size() == block.columns()) - return {}; - - return permutation; -} - } JoinStep::JoinStep( @@ -76,15 +44,8 @@ JoinStep::JoinStep( JoinPtr join_, size_t max_block_size_, size_t max_streams_, - NameSet required_output_, - bool keep_left_read_in_order_, - bool use_new_analyzer_) - : join(std::move(join_)) - , max_block_size(max_block_size_) - , max_streams(max_streams_) - , required_output(std::move(required_output_)) - , keep_left_read_in_order(keep_left_read_in_order_) - , use_new_analyzer(use_new_analyzer_) + bool keep_left_read_in_order_) + : join(std::move(join_)), max_block_size(max_block_size_), max_streams(max_streams_), keep_left_read_in_order(keep_left_read_in_order_) { updateInputHeaders({left_header_, right_header_}); } @@ -94,43 +55,23 @@ QueryPipelineBuilderPtr JoinStep::updatePipeline(QueryPipelineBuilders pipelines if (pipelines.size() != 2) throw Exception(ErrorCodes::LOGICAL_ERROR, "JoinStep expect two input steps"); - Block lhs_header = pipelines[0]->getHeader(); - Block rhs_header = pipelines[1]->getHeader(); - - if (swap_streams) - std::swap(pipelines[0], pipelines[1]); - if (join->pipelineType() == JoinPipelineType::YShaped) { auto joined_pipeline = QueryPipelineBuilder::joinPipelinesYShaped( - std::move(pipelines[0]), std::move(pipelines[1]), join, join_algorithm_header, max_block_size, &processors); + std::move(pipelines[0]), std::move(pipelines[1]), join, *output_header, max_block_size, &processors); joined_pipeline->resize(max_streams); return joined_pipeline; } - auto pipeline = QueryPipelineBuilder::joinPipelinesRightLeft( + return QueryPipelineBuilder::joinPipelinesRightLeft( std::move(pipelines[0]), std::move(pipelines[1]), join, - join_algorithm_header, + *output_header, max_block_size, max_streams, keep_left_read_in_order, &processors); - - if (!use_new_analyzer) - return pipeline; - - auto column_permutation = getPermutationForBlock(pipeline->getHeader(), lhs_header, rhs_header, required_output); - if (!column_permutation.empty()) - { - pipeline->addSimpleTransform([&column_permutation](const Block & header) - { - return std::make_shared(header, column_permutation); - }); - } - - return pipeline; } bool JoinStep::allowPushDownToRight() const @@ -149,49 +90,17 @@ void JoinStep::describeActions(FormatSettings & settings) const for (const auto & [name, value] : describeJoinActions(join)) settings.out << prefix << name << ": " << value << '\n'; - if (swap_streams) - settings.out << prefix << "Swapped: true\n"; } void JoinStep::describeActions(JSONBuilder::JSONMap & map) const { for (const auto & [name, value] : describeJoinActions(join)) map.add(name, value); - if (swap_streams) - map.add("Swapped", true); -} - -void JoinStep::setJoin(JoinPtr join_, bool swap_streams_) -{ - join_algorithm_header.clear(); - swap_streams = swap_streams_; - join = std::move(join_); - updateOutputHeader(); } void JoinStep::updateOutputHeader() { - if (join_algorithm_header) - return; - - const auto & header = swap_streams ? input_headers[1] : input_headers[0]; - - Block result_header = JoiningTransform::transformHeader(header, join); - join_algorithm_header = result_header; - - if (!use_new_analyzer) - { - if (swap_streams) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot swap streams without new analyzer"); - output_header = result_header; - return; - } - - auto column_permutation = getPermutationForBlock(result_header, input_headers[0], input_headers[1], required_output); - if (!column_permutation.empty()) - result_header = ColumnPermuteTransform::permute(result_header, column_permutation); - - output_header = result_header; + output_header = JoiningTransform::transformHeader(input_headers.front(), join); } static ITransformingStep::Traits getStorageJoinTraits() diff --git a/src/Processors/QueryPlan/JoinStep.h b/src/Processors/QueryPlan/JoinStep.h index 1eca42c62cf..2793784d633 100644 --- a/src/Processors/QueryPlan/JoinStep.h +++ b/src/Processors/QueryPlan/JoinStep.h @@ -2,7 +2,6 @@ #include #include -#include namespace DB { @@ -20,9 +19,7 @@ public: JoinPtr join_, size_t max_block_size_, size_t max_streams_, - NameSet required_output_, - bool keep_left_read_in_order_, - bool use_new_analyzer_); + bool keep_left_read_in_order_); String getName() const override { return "Join"; } @@ -34,26 +31,16 @@ public: void describeActions(FormatSettings & settings) const override; const JoinPtr & getJoin() const { return join; } - void setJoin(JoinPtr join_, bool swap_streams_ = false); + void setJoin(JoinPtr join_) { join = std::move(join_); } bool allowPushDownToRight() const; - JoinInnerTableSelectionMode inner_table_selection_mode = JoinInnerTableSelectionMode::Right; - private: void updateOutputHeader() override; - /// Header that expected to be returned from IJoin - Block join_algorithm_header; - JoinPtr join; size_t max_block_size; size_t max_streams; - - const NameSet required_output; - std::set columns_to_remove; bool keep_left_read_in_order; - bool use_new_analyzer = false; - bool swap_streams = false; }; /// Special step for the case when Join is already filled. diff --git a/src/Processors/QueryPlan/Optimizations/Optimizations.h b/src/Processors/QueryPlan/Optimizations/Optimizations.h index c1c4d1e1635..751d5182dc3 100644 --- a/src/Processors/QueryPlan/Optimizations/Optimizations.h +++ b/src/Processors/QueryPlan/Optimizations/Optimizations.h @@ -113,7 +113,6 @@ void optimizePrimaryKeyConditionAndLimit(const Stack & stack); void optimizePrewhere(Stack & stack, QueryPlan::Nodes & nodes); void optimizeReadInOrder(QueryPlan::Node & node, QueryPlan::Nodes & nodes); void optimizeAggregationInOrder(QueryPlan::Node & node, QueryPlan::Nodes &); -void optimizeJoin(QueryPlan::Node & node, QueryPlan::Nodes &); void optimizeDistinctInOrder(QueryPlan::Node & node, QueryPlan::Nodes &); /// A separate tree traverse to apply sorting properties after *InOrder optimizations. diff --git a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp b/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp deleted file mode 100644 index c0b31864eac..00000000000 --- a/src/Processors/QueryPlan/Optimizations/optimizeJoin.cpp +++ /dev/null @@ -1,102 +0,0 @@ -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include - -#include -#include -#include - -namespace DB::QueryPlanOptimizations -{ - -static std::optional estimateReadRowsCount(QueryPlan::Node & node) -{ - IQueryPlanStep * step = node.step.get(); - if (const auto * reading = typeid_cast(step)) - { - if (auto analyzed_result = reading->getAnalyzedResult()) - return analyzed_result->selected_rows; - if (auto analyzed_result = reading->selectRangesToRead()) - return analyzed_result->selected_rows; - return {}; - } - - if (const auto * reading = typeid_cast(step)) - return reading->getStorage()->totalRows(Settings{}); - - if (node.children.size() != 1) - return {}; - - if (typeid_cast(step) || typeid_cast(step)) - return estimateReadRowsCount(*node.children.front()); - - return {}; -} - -void optimizeJoin(QueryPlan::Node & node, QueryPlan::Nodes &) -{ - auto * join_step = typeid_cast(node.step.get()); - if (!join_step || node.children.size() != 2) - return; - - const auto & join = join_step->getJoin(); - if (join->pipelineType() != JoinPipelineType::FillRightFirst || !join->isCloneSupported()) - return; - - const auto & table_join = join->getTableJoin(); - - /// Algorithms other than HashJoin may not support OUTER JOINs - if (table_join.kind() != JoinKind::Inner && !typeid_cast(join.get())) - return; - - /// fixme: USING clause handled specially in join algorithm, so swap breaks it - /// fixme: Swapping for SEMI and ANTI joins should be alright, need to try to enable it and test - if (table_join.hasUsing() || table_join.strictness() != JoinStrictness::All) - return; - - bool need_swap = false; - if (join_step->inner_table_selection_mode == JoinInnerTableSelectionMode::Auto) - { - auto lhs_extimation = estimateReadRowsCount(*node.children[0]); - auto rhs_extimation = estimateReadRowsCount(*node.children[1]); - LOG_TRACE(getLogger("optimizeJoin"), "Left table estimation: {}, right table estimation: {}", - lhs_extimation.transform(toString).value_or("unknown"), - rhs_extimation.transform(toString).value_or("unknown")); - - if (lhs_extimation && rhs_extimation && *lhs_extimation < *rhs_extimation) - need_swap = true; - } - else if (join_step->inner_table_selection_mode == JoinInnerTableSelectionMode::Left) - { - need_swap = true; - } - - if (!need_swap) - return; - - const auto & headers = join_step->getInputHeaders(); - if (headers.size() != 2) - return; - - const auto & left_stream_input_header = headers.front(); - const auto & right_stream_input_header = headers.back(); - - auto updated_table_join = std::make_shared(table_join); - updated_table_join->swapSides(); - auto updated_join = join->clone(updated_table_join, right_stream_input_header, left_stream_input_header); - join_step->setJoin(std::move(updated_join), /* swap_streams= */ true); -} - -} diff --git a/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp b/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp index c034ca79181..03418c752d4 100644 --- a/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp +++ b/src/Processors/QueryPlan/Optimizations/optimizeTree.cpp @@ -227,9 +227,6 @@ void addStepsToBuildSets(QueryPlan & plan, QueryPlan::Node & root, QueryPlan::No /// NOTE: frame cannot be safely used after stack was modified. auto & frame = stack.back(); - if (frame.next_child == 0) - optimizeJoin(*frame.node, nodes); - /// Traverse all children first. if (frame.next_child < frame.node->children.size()) { diff --git a/src/Processors/QueryPlan/ReadFromMemoryStorageStep.h b/src/Processors/QueryPlan/ReadFromMemoryStorageStep.h index a9c2d2df2c4..238c1a3aad0 100644 --- a/src/Processors/QueryPlan/ReadFromMemoryStorageStep.h +++ b/src/Processors/QueryPlan/ReadFromMemoryStorageStep.h @@ -35,8 +35,6 @@ public: void initializePipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings &) override; - const StoragePtr & getStorage() const { return storage; } - private: static constexpr auto name = "ReadFromMemoryStorage"; diff --git a/src/Processors/Transforms/ColumnPermuteTransform.cpp b/src/Processors/Transforms/ColumnPermuteTransform.cpp deleted file mode 100644 index f371689814c..00000000000 --- a/src/Processors/Transforms/ColumnPermuteTransform.cpp +++ /dev/null @@ -1,49 +0,0 @@ -#include - -namespace DB -{ - -namespace -{ - -template -void applyPermutation(std::vector & data, const std::vector & permutation) -{ - std::vector res; - res.reserve(permutation.size()); - for (size_t i : permutation) - res.push_back(data[i]); - data = std::move(res); -} - -void permuteChunk(Chunk & chunk, const std::vector & permutation) -{ - size_t num_rows = chunk.getNumRows(); - auto columns = chunk.detachColumns(); - applyPermutation(columns, permutation); - chunk.setColumns(std::move(columns), num_rows); -} - -} - -Block ColumnPermuteTransform::permute(const Block & block, const std::vector & permutation) -{ - auto columns = block.getColumnsWithTypeAndName(); - applyPermutation(columns, permutation); - return Block(columns); -} - -ColumnPermuteTransform::ColumnPermuteTransform(const Block & header_, const std::vector & permutation_) - : ISimpleTransform(header_, permute(header_, permutation_), false) - , permutation(permutation_) -{ -} - - -void ColumnPermuteTransform::transform(Chunk & chunk) -{ - permuteChunk(chunk, permutation); -} - - -} diff --git a/src/Processors/Transforms/ColumnPermuteTransform.h b/src/Processors/Transforms/ColumnPermuteTransform.h deleted file mode 100644 index 25f3a8d0825..00000000000 --- a/src/Processors/Transforms/ColumnPermuteTransform.h +++ /dev/null @@ -1,30 +0,0 @@ -#pragma once - -#include -#include -#include -#include -#include -#include - -namespace DB -{ - -class ColumnPermuteTransform : public ISimpleTransform -{ -public: - ColumnPermuteTransform(const Block & header_, const std::vector & permutation_); - - String getName() const override { return "ColumnPermuteTransform"; } - - void transform(Chunk & chunk) override; - - static Block permute(const Block & block, const std::vector & permutation); - -private: - Names column_names; - std::vector permutation; -}; - - -} diff --git a/src/Processors/Transforms/JoiningTransform.cpp b/src/Processors/Transforms/JoiningTransform.cpp index 187f4bf6728..f2fb6327129 100644 --- a/src/Processors/Transforms/JoiningTransform.cpp +++ b/src/Processors/Transforms/JoiningTransform.cpp @@ -19,7 +19,6 @@ Block JoiningTransform::transformHeader(Block header, const JoinPtr & join) join->initialize(header); ExtraBlockPtr tmp; join->joinBlock(header, tmp); - materializeBlockInplace(header); LOG_TEST(getLogger("JoiningTransform"), "After join block: '{}'", header.dumpStructure()); return header; } diff --git a/tests/clickhouse-test b/tests/clickhouse-test index f4c3b368632..9c035b7cc35 100755 --- a/tests/clickhouse-test +++ b/tests/clickhouse-test @@ -789,7 +789,6 @@ def get_localzone(): return os.getenv("TZ", "/".join(os.readlink("/etc/localtime").split("/")[-2:])) -# Refer to `tests/integration/helpers/random_settings.py` for integration test random settings class SettingsRandomizer: settings = { "max_insert_threads": lambda: ( @@ -920,9 +919,6 @@ class SettingsRandomizer: "max_parsing_threads": lambda: random.choice([0, 1, 10]), "optimize_functions_to_subcolumns": lambda: random.randint(0, 1), "parallel_replicas_local_plan": lambda: random.randint(0, 1), - "query_plan_join_inner_table_selection": lambda: random.choice( - ["left", "auto", "right"] - ), "output_format_native_write_json_as_string": lambda: random.randint(0, 1), "enable_vertical_final": lambda: random.randint(0, 1), } diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index 6751f205fb8..7c531cdd493 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -67,7 +67,6 @@ DEFAULT_ENV_NAME = ".env" DEFAULT_BASE_CONFIG_DIR = os.environ.get( "CLICKHOUSE_TESTS_BASE_CONFIG_DIR", "/etc/clickhouse-server/" ) -DOCKER_BASE_TAG = os.environ.get("DOCKER_BASE_TAG", "latest") SANITIZER_SIGN = "==================" @@ -504,6 +503,7 @@ class ClickHouseCluster: "CLICKHOUSE_TESTS_DOCKERD_HOST" ) self.docker_api_version = os.environ.get("DOCKER_API_VERSION") + self.docker_base_tag = os.environ.get("DOCKER_BASE_TAG", "latest") self.base_cmd = ["docker", "compose"] if custom_dockerd_host: @@ -1079,7 +1079,7 @@ class ClickHouseCluster: env_variables["keeper_binary"] = binary_path env_variables["keeper_cmd_prefix"] = keeper_cmd_prefix - env_variables["image"] = "clickhouse/integration-test:" + DOCKER_BASE_TAG + env_variables["image"] = "clickhouse/integration-test:" + self.docker_base_tag env_variables["user"] = str(os.getuid()) env_variables["keeper_fs"] = "bind" for i in range(1, 4): @@ -1675,7 +1675,7 @@ class ClickHouseCluster: ) if tag is None: - tag = DOCKER_BASE_TAG + tag = self.docker_base_tag if not env_variables: env_variables = {} self.use_keeper = use_keeper @@ -4538,12 +4538,7 @@ class ClickHouseInstance: if len(self.custom_dictionaries_paths): write_embedded_config("0_common_enable_dictionaries.xml", self.config_d_dir) - if ( - self.randomize_settings - and self.image == "clickhouse/integration-test" - and self.tag == DOCKER_BASE_TAG - and self.base_config_dir == DEFAULT_BASE_CONFIG_DIR - ): + if self.randomize_settings and self.base_config_dir == DEFAULT_BASE_CONFIG_DIR: # If custom main config is used, do not apply random settings to it write_random_settings_config(Path(users_d_dir) / "0_random_settings.xml") diff --git a/tests/integration/helpers/random_settings.py b/tests/integration/helpers/random_settings.py index 32cde54d0e7..b2319561fd7 100644 --- a/tests/integration/helpers/random_settings.py +++ b/tests/integration/helpers/random_settings.py @@ -5,8 +5,6 @@ def randomize_settings(): yield "max_joined_block_size_rows", random.randint(8000, 100000) if random.random() < 0.5: yield "max_block_size", random.randint(8000, 100000) - if random.random() < 0.5: - yield "query_plan_join_inner_table_selection", random.choice(["auto", "left"]) def write_random_settings_config(destination): diff --git a/tests/integration/test_peak_memory_usage/test.py b/tests/integration/test_peak_memory_usage/test.py index 69057573173..51268dcf386 100644 --- a/tests/integration/test_peak_memory_usage/test.py +++ b/tests/integration/test_peak_memory_usage/test.py @@ -91,7 +91,7 @@ def test_clickhouse_client_max_peak_memory_usage_distributed(started_cluster): with client(name="client1>", log=client_output, command=command_text) as client1: client1.expect(prompt) client1.send( - "SELECT COUNT(*) FROM distributed_fixed_numbers JOIN fixed_numbers_2 ON distributed_fixed_numbers.number=fixed_numbers_2.number SETTINGS query_plan_join_inner_table_selection = 'right'", + "SELECT COUNT(*) FROM distributed_fixed_numbers JOIN fixed_numbers_2 ON distributed_fixed_numbers.number=fixed_numbers_2.number", ) client1.expect("Peak memory usage", timeout=60) client1.expect(prompt) diff --git a/tests/queries/0_stateless/00826_cross_to_inner_join.sql b/tests/queries/0_stateless/00826_cross_to_inner_join.sql index 5ab7a2d0626..e9f9e13e2d3 100644 --- a/tests/queries/0_stateless/00826_cross_to_inner_join.sql +++ b/tests/queries/0_stateless/00826_cross_to_inner_join.sql @@ -15,9 +15,9 @@ INSERT INTO t2_00826 values (1,1), (1,2); INSERT INTO t2_00826 (a) values (2), (3); SELECT '--- cross ---'; -SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.a = t2_00826.a ORDER BY ALL; +SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.a = t2_00826.a; SELECT '--- cross nullable ---'; -SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.b = t2_00826.b ORDER BY ALL; +SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.b = t2_00826.b; SELECT '--- cross nullable vs not nullable ---'; SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.a = t2_00826.b ORDER BY t1_00826.a; SELECT '--- cross self ---'; @@ -41,15 +41,14 @@ SELECT '--- is null or ---'; SELECT * FROM t1_00826 cross join t2_00826 where t1_00826.b = t2_00826.a AND (t2_00826.b IS NULL OR t2_00826.b > t2_00826.a) ORDER BY t1_00826.a; SELECT '--- do not rewrite alias ---'; -SELECT a as b FROM t1_00826 cross join t2_00826 where t1_00826.b = t2_00826.a AND b > 0 ORDER BY ALL; +SELECT a as b FROM t1_00826 cross join t2_00826 where t1_00826.b = t2_00826.a AND b > 0; SELECT '--- comma ---'; -SELECT * FROM t1_00826, t2_00826 where t1_00826.a = t2_00826.a ORDER BY ALL; +SELECT * FROM t1_00826, t2_00826 where t1_00826.a = t2_00826.a; SELECT '--- comma nullable ---'; -SELECT * FROM t1_00826, t2_00826 where t1_00826.b = t2_00826.b ORDER BY ALL; +SELECT * FROM t1_00826, t2_00826 where t1_00826.b = t2_00826.b; SELECT '--- comma and or ---'; -SELECT * FROM t1_00826, t2_00826 where t1_00826.a = t2_00826.a AND (t2_00826.b IS NULL OR t2_00826.b < 2) -ORDER BY ALL; +SELECT * FROM t1_00826, t2_00826 where t1_00826.a = t2_00826.a AND (t2_00826.b IS NULL OR t2_00826.b < 2); SELECT '--- cross ---'; diff --git a/tests/queries/0_stateless/00847_multiple_join_same_column.sql b/tests/queries/0_stateless/00847_multiple_join_same_column.sql index bbb4eb12466..c7f0c6383c2 100644 --- a/tests/queries/0_stateless/00847_multiple_join_same_column.sql +++ b/tests/queries/0_stateless/00847_multiple_join_same_column.sql @@ -20,42 +20,42 @@ select t.a, s.b, s.a, s.b, y.a, y.b from t left join s on (t.a = s.a and s.b = t.b) left join y on (y.a = s.a and y.b = s.b) order by t.a -format PrettyCompactMonoBlock; +format PrettyCompactNoEscapes; select t.a as t_a from t left join s on s.a = t_a order by t.a -format PrettyCompactMonoBlock; +format PrettyCompactNoEscapes; select t.a, s.a as s_a from t left join s on s.a = t.a left join y on y.b = s.b order by t.a -format PrettyCompactMonoBlock; +format PrettyCompactNoEscapes; select t.a, t.a, t.b as t_b from t left join s on t.a = s.a left join y on y.b = s.b order by t.a -format PrettyCompactMonoBlock; +format PrettyCompactNoEscapes; select s.a, s.a, s.b as s_b, s.b from t left join s on s.a = t.a left join y on s.b = y.b order by t.a -format PrettyCompactMonoBlock; +format PrettyCompactNoEscapes; select y.a, y.a, y.b as y_b, y.b from t left join s on s.a = t.a left join y on y.b = s.b order by t.a -format PrettyCompactMonoBlock; +format PrettyCompactNoEscapes; select t.a, t.a as t_a, s.a, s.a as s_a, y.a, y.a as y_a from t left join s on t.a = s.a left join y on y.b = s.b order by t.a -format PrettyCompactMonoBlock; +format PrettyCompactNoEscapes; drop table t; drop table s; diff --git a/tests/queries/0_stateless/01015_empty_in_inner_right_join.sql.j2 b/tests/queries/0_stateless/01015_empty_in_inner_right_join.sql.j2 index cdbb0542ffb..cdb9d253b9b 100644 --- a/tests/queries/0_stateless/01015_empty_in_inner_right_join.sql.j2 +++ b/tests/queries/0_stateless/01015_empty_in_inner_right_join.sql.j2 @@ -1,7 +1,5 @@ SET joined_subquery_requires_alias = 0; -SET query_plan_join_inner_table_selection = 'auto'; - {% for join_algorithm in ['partial_merge', 'hash'] -%} SET join_algorithm = '{{ join_algorithm }}'; diff --git a/tests/queries/0_stateless/01107_join_right_table_totals.reference b/tests/queries/0_stateless/01107_join_right_table_totals.reference index aa569ff9331..daf503b776d 100644 --- a/tests/queries/0_stateless/01107_join_right_table_totals.reference +++ b/tests/queries/0_stateless/01107_join_right_table_totals.reference @@ -18,35 +18,28 @@ 0 0 0 0 -- 1 1 1 1 0 0 -- 1 1 1 1 0 0 -- 1 1 1 1 0 0 -- 1 1 1 1 0 0 -- 1 1 0 0 -- 1 foo 1 1 300 0 foo 1 0 300 -- 1 100 1970-01-01 1 100 1970-01-01 1 100 1970-01-01 1 200 1970-01-02 1 200 1970-01-02 1 100 1970-01-01 diff --git a/tests/queries/0_stateless/01107_join_right_table_totals.sql b/tests/queries/0_stateless/01107_join_right_table_totals.sql index 7e549282489..ad8954d5d70 100644 --- a/tests/queries/0_stateless/01107_join_right_table_totals.sql +++ b/tests/queries/0_stateless/01107_join_right_table_totals.sql @@ -64,47 +64,39 @@ USING (id); INSERT INTO t VALUES (1, 100, '1970-01-01'), (1, 200, '1970-01-02'); -SELECT '-'; SELECT * FROM (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id) l LEFT JOIN (SELECT item_id FROM t ) r ON l.item_id = r.item_id; -SELECT '-'; SELECT * FROM (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id) l RIGHT JOIN (SELECT item_id FROM t ) r ON l.item_id = r.item_id; -SELECT '-'; SELECT * FROM (SELECT item_id FROM t) l LEFT JOIN (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id ) r ON l.item_id = r.item_id; -SELECT '-'; SELECT * FROM (SELECT item_id FROM t) l RIGHT JOIN (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id ) r ON l.item_id = r.item_id; -SELECT '-'; SELECT * FROM (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id) l LEFT JOIN (SELECT item_id FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id ) r ON l.item_id = r.item_id; -SELECT '-'; SELECT * FROM (SELECT item_id, 'foo' AS key, 1 AS val FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id) l LEFT JOIN (SELECT item_id, sum(price_sold) AS val FROM t GROUP BY item_id WITH TOTALS ORDER BY item_id ) r ON l.item_id = r.item_id; -SELECT '-'; SELECT * FROM (SELECT * FROM t GROUP BY item_id, price_sold, date WITH TOTALS ORDER BY item_id, price_sold, date) l LEFT JOIN (SELECT * FROM t GROUP BY item_id, price_sold, date WITH TOTALS ORDER BY item_id, price_sold, date ) r -ON l.item_id = r.item_id -ORDER BY ALL; +ON l.item_id = r.item_id; DROP TABLE t; diff --git a/tests/queries/0_stateless/01763_filter_push_down_bugs.reference b/tests/queries/0_stateless/01763_filter_push_down_bugs.reference index 229ac6eae09..19018a610b7 100644 --- a/tests/queries/0_stateless/01763_filter_push_down_bugs.reference +++ b/tests/queries/0_stateless/01763_filter_push_down_bugs.reference @@ -26,7 +26,7 @@ Expression ((Projection + Before ORDER BY)) Parts: 1/1 Granules: 1/1 Expression ((Project names + Projection)) - Filter (WHERE) + Filter ((WHERE + DROP unused columns after JOIN)) Join (JOIN FillRightFirst) Expression ReadFromMergeTree (default.t1) diff --git a/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 b/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 index c13722f431a..c2d85cefb18 100644 --- a/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 +++ b/tests/queries/0_stateless/01881_join_on_conditions_hash.sql.j2 @@ -75,7 +75,7 @@ SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t2.key; -- { serverErro SELECT * FROM t1 JOIN t2_nullable as t2 ON t2.key == t2.key2 AND (t1.id == t2.id OR isNull(t2.key2)); -- { serverError 403 } SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 OR t1.id == t2.id; -- { serverError 403 } SELECT * FROM t1 JOIN t2 ON (t2.key == t2.key2 AND (t1.key == t1.key2 AND t1.key != 'XXX' OR t1.id == t2.id)) AND t1.id == t2.id; -- { serverError 403 } -SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 AND t1.key == t1.key2 AND t1.key != 'XXX' AND t1.id == t2.id OR t2.key == t2.key2 AND t1.id == t2.id AND t1.id == t2.id ORDER BY ALL; +SELECT * FROM t1 JOIN t2 ON t2.key == t2.key2 AND t1.key == t1.key2 AND t1.key != 'XXX' AND t1.id == t2.id OR t2.key == t2.key2 AND t1.id == t2.id AND t1.id == t2.id; -- non-equi condition containing columns from different tables doesn't supported yet SELECT * FROM t1 INNER ALL JOIN t2 ON t1.id == t2.id AND t1.id >= t2.id; -- { serverError 403 } SELECT * FROM t1 INNER ANY JOIN t2 ON t1.id == t2.id AND t2.key == t2.key2 AND t1.key == t1.key2 AND t1.id >= length(t2.key); -- { serverError 403 } @@ -89,10 +89,10 @@ SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.idd and (t1.id == t22.id OR t22 SELECT 't22', * FROM t1 JOIN t22 ON (t22.key == t22.key2 OR t1.id == t22.id) and t1.id == t22.idd; -- { serverError 403 } SELECT 't22', * FROM t1 JOIN t22 ON (t1.id == t22.id OR t22.key == t22.key2) and t1.id == t22.idd; -- { serverError 403 } SELECT 't22', * FROM t1 JOIN t22 ON (t1.id == t22.id OR t22.key == t22.key2) and (t1.id == t22.idd AND (t1.key2 = 'a1' OR t1.key2 = 'a2' OR t1.key2 = 'a3' OR t1.key2 = 'a4' OR t1.key2 = 'a5' OR t1.key2 = 'a6' OR t1.key2 = 'a7' OR t1.key2 = 'a8' OR t1.key2 = 'a9' OR t1.key2 = 'a10' OR t1.key2 = 'a11' OR t1.key2 = 'a12' OR t1.key2 = 'a13' OR t1.key2 = 'a14' OR t1.key2 = 'a15' OR t1.key2 = 'a16' OR t1.key2 = 'a17' OR t1.key2 = 'a18' OR t1.key2 = 'a19' OR t1.key2 = '111')); -- { serverError 403 } -SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.idd and t22.key == t22.key2 OR t1.id == t22.idd and t1.id == t22.id ORDER BY ALL; -SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.idd and t1.id == t22.id OR t1.id == t22.idd and t22.key == t22.key2 ORDER BY ALL; -SELECT 't22', * FROM t1 JOIN t22 ON t22.key == t22.key2 and t1.id == t22.idd OR t1.id == t22.id and t1.id == t22.idd ORDER BY ALL; -SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.id and t1.id == t22.idd OR t22.key == t22.key2 and t1.id == t22.idd ORDER BY ALL; +SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.idd and t22.key == t22.key2 OR t1.id == t22.idd and t1.id == t22.id; +SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.idd and t1.id == t22.id OR t1.id == t22.idd and t22.key == t22.key2; +SELECT 't22', * FROM t1 JOIN t22 ON t22.key == t22.key2 and t1.id == t22.idd OR t1.id == t22.id and t1.id == t22.idd; +SELECT 't22', * FROM t1 JOIN t22 ON t1.id == t22.id and t1.id == t22.idd OR t22.key == t22.key2 and t1.id == t22.idd; {% endfor -%} diff --git a/tests/queries/0_stateless/02000_join_on_const.reference b/tests/queries/0_stateless/02000_join_on_const.reference index f8e46a2b976..3bd1633ce32 100644 --- a/tests/queries/0_stateless/02000_join_on_const.reference +++ b/tests/queries/0_stateless/02000_join_on_const.reference @@ -33,23 +33,23 @@ 2 2 2 2 -- { echoOn } -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; 1 0 2 2 -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 1 SETTINGS enable_analyzer = 1; -0 3 +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; 2 2 -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 2, 1 SETTINGS enable_analyzer = 1; +0 3 +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; 1 0 2 2 0 3 -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; 1 0 2 0 -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 2 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; 0 2 0 3 -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 2, 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; 1 0 2 0 0 2 @@ -59,11 +59,11 @@ SELECT * FROM (SELECT 1 as a) as t1 LEFT JOIN ( SELECT ('b', 256) as b ) AS t2 1 ('',0) SELECT * FROM (SELECT 1 as a) as t1 RIGHT JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; 0 ('b',256) -SELECT * FROM (SELECT 1 as a) as t1 FULL JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL ORDER BY 2; +SELECT * FROM (SELECT 1 as a) as t1 FULL JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; 1 ('',0) 0 ('b',256) SELECT * FROM (SELECT 1 as a) as t1 SEMI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; -SELECT * FROM (SELECT 1 as a) as t1 ANTI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL ORDER BY 2; +SELECT * FROM (SELECT 1 as a) as t1 ANTI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; 1 ('',0) 2 4 2 Nullable(UInt64) UInt8 diff --git a/tests/queries/0_stateless/02000_join_on_const.sql b/tests/queries/0_stateless/02000_join_on_const.sql index 33638edafa5..da70973ed87 100644 --- a/tests/queries/0_stateless/02000_join_on_const.sql +++ b/tests/queries/0_stateless/02000_join_on_const.sql @@ -73,20 +73,20 @@ SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 SETTINGS enable_analyzer = 0; -- SELECT * FROM t1 JOIN t2 ON t1.id = t2.id AND 1 SETTINGS enable_analyzer = 1; -- { echoOn } -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 1 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 1 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 ORDER BY 2, 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 1 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 1 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 2 SETTINGS enable_analyzer = 1; -SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 ORDER BY 2, 1 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 LEFT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 RIGHT JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; +SELECT * FROM t1 FULL JOIN t2 ON t1.id = t2.id AND 1 = 2 SETTINGS enable_analyzer = 1; SELECT * FROM (SELECT 1 as a) as t1 INNER JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; SELECT * FROM (SELECT 1 as a) as t1 LEFT JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; SELECT * FROM (SELECT 1 as a) as t1 RIGHT JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; -SELECT * FROM (SELECT 1 as a) as t1 FULL JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL ORDER BY 2; +SELECT * FROM (SELECT 1 as a) as t1 FULL JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; SELECT * FROM (SELECT 1 as a) as t1 SEMI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; -SELECT * FROM (SELECT 1 as a) as t1 ANTI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL ORDER BY 2; +SELECT * FROM (SELECT 1 as a) as t1 ANTI JOIN ( SELECT ('b', 256) as b ) AS t2 ON NULL; -- { echoOff } diff --git a/tests/queries/0_stateless/02001_join_on_const_bs_long.sql.j2 b/tests/queries/0_stateless/02001_join_on_const_bs_long.sql.j2 index 83548e087bd..1726bcb7062 100644 --- a/tests/queries/0_stateless/02001_join_on_const_bs_long.sql.j2 +++ b/tests/queries/0_stateless/02001_join_on_const_bs_long.sql.j2 @@ -1,8 +1,8 @@ DROP TABLE IF EXISTS t1; DROP TABLE IF EXISTS t2; -CREATE TABLE t1 (id Int) ENGINE = TinyLog; -CREATE TABLE t2 (id Int) ENGINE = TinyLog; +CREATE TABLE t1 (id Int) ENGINE = MergeTree ORDER BY id; +CREATE TABLE t2 (id Int) ENGINE = MergeTree ORDER BY id; INSERT INTO t1 VALUES (1), (2); INSERT INTO t2 SELECT number + 5 AS x FROM (SELECT * FROM system.numbers LIMIT 1111); diff --git a/tests/queries/0_stateless/02245_join_with_nullable_lowcardinality_crash.sql b/tests/queries/0_stateless/02245_join_with_nullable_lowcardinality_crash.sql index c3c84ebaded..abc2ee41402 100644 --- a/tests/queries/0_stateless/02245_join_with_nullable_lowcardinality_crash.sql +++ b/tests/queries/0_stateless/02245_join_with_nullable_lowcardinality_crash.sql @@ -12,9 +12,8 @@ CREATE TABLE without_nullable insert into with_nullable values(0,'f'),(0,'usa'); insert into without_nullable values(0,'usa'),(0,'us2a'); -select if(t0.country is null ,t2.country,t0.country) "country" -from without_nullable t0 right outer join with_nullable t2 on t0.country=t2.country -ORDER BY 1 DESC; +select if(t0.country is null ,t2.country,t0.country) "country" +from without_nullable t0 right outer join with_nullable t2 on t0.country=t2.country; drop table with_nullable; drop table without_nullable; diff --git a/tests/queries/0_stateless/02282_array_distance.sql b/tests/queries/0_stateless/02282_array_distance.sql index 85abc8fa381..2cca853fd67 100644 --- a/tests/queries/0_stateless/02282_array_distance.sql +++ b/tests/queries/0_stateless/02282_array_distance.sql @@ -48,8 +48,7 @@ SELECT L2SquaredDistance(v1.v, v2.v), cosineDistance(v1.v, v2.v) FROM vec2 v1, vec2 v2 -WHERE length(v1.v) == length(v2.v) -ORDER BY ALL; +WHERE length(v1.v) == length(v2.v); INSERT INTO vec2f VALUES (1, [100, 200, 0]), (2, [888, 777, 666]), (3, range(1, 35, 1)), (4, range(3, 37, 1)), (5, range(1, 135, 1)), (6, range(3, 137, 1)); SELECT @@ -62,8 +61,7 @@ SELECT L2SquaredDistance(v1.v, v2.v), cosineDistance(v1.v, v2.v) FROM vec2f v1, vec2f v2 -WHERE length(v1.v) == length(v2.v) -ORDER BY ALL; +WHERE length(v1.v) == length(v2.v); INSERT INTO vec2d VALUES (1, [100, 200, 0]), (2, [888, 777, 666]), (3, range(1, 35, 1)), (4, range(3, 37, 1)), (5, range(1, 135, 1)), (6, range(3, 137, 1)); SELECT @@ -76,8 +74,7 @@ SELECT L2SquaredDistance(v1.v, v2.v), cosineDistance(v1.v, v2.v) FROM vec2d v1, vec2d v2 -WHERE length(v1.v) == length(v2.v) -ORDER BY ALL; +WHERE length(v1.v) == length(v2.v); SELECT v1.id, @@ -89,8 +86,7 @@ SELECT L2SquaredDistance(v1.v, v2.v), cosineDistance(v1.v, v2.v) FROM vec2f v1, vec2d v2 -WHERE length(v1.v) == length(v2.v) -ORDER BY ALL; +WHERE length(v1.v) == length(v2.v); SELECT L1Distance([0, 0], [1]); -- { serverError SIZES_OF_ARRAYS_DONT_MATCH } SELECT L2Distance([1, 2], (3,4)); -- { serverError ILLEGAL_TYPE_OF_ARGUMENT } diff --git a/tests/queries/0_stateless/02381_join_dup_columns_in_plan.reference b/tests/queries/0_stateless/02381_join_dup_columns_in_plan.reference index 90aab0a0eb2..365725f8ffe 100644 --- a/tests/queries/0_stateless/02381_join_dup_columns_in_plan.reference +++ b/tests/queries/0_stateless/02381_join_dup_columns_in_plan.reference @@ -148,6 +148,7 @@ Header: key String value String Join Header: __table1.key String + __table3.key String __table3.value String Sorting Header: __table1.key String diff --git a/tests/queries/0_stateless/02461_join_lc_issue_42380.sql b/tests/queries/0_stateless/02461_join_lc_issue_42380.sql index 8b5c6846bd0..f0ecbf64e58 100644 --- a/tests/queries/0_stateless/02461_join_lc_issue_42380.sql +++ b/tests/queries/0_stateless/02461_join_lc_issue_42380.sql @@ -9,5 +9,4 @@ CREATE TABLE t2__fuzz_47 (id LowCardinality(Int16)) ENGINE = MergeTree() ORDER B INSERT INTO t1__fuzz_13 VALUES (1); INSERT INTO t2__fuzz_47 VALUES (1); -SELECT * FROM t1__fuzz_13 FULL OUTER JOIN t2__fuzz_47 ON 1 = 2 -ORDER BY ALL; +SELECT * FROM t1__fuzz_13 FULL OUTER JOIN t2__fuzz_47 ON 1 = 2; diff --git a/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference b/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference index c9bf36f88ea..3c68d14fdf2 100644 --- a/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference +++ b/tests/queries/0_stateless/02496_remove_redundant_sorting_analyzer.reference @@ -117,7 +117,7 @@ ORDER BY t1.number, t2.number -- explain Expression (Project names) Sorting (Sorting for ORDER BY) - Expression ((Before ORDER BY + Projection)) + Expression ((Before ORDER BY + (Projection + DROP unused columns after JOIN))) Join (JOIN FillRightFirst) Expression ((Change column names to column identifiers + (Project names + (Before ORDER BY + (Projection + (Change column names to column identifiers + (Project names + (Before ORDER BY + (Projection + Change column names to column identifiers))))))))) ReadFromSystemNumbers @@ -161,7 +161,7 @@ ORDER BY t1.number, t2.number -- explain Expression (Project names) Sorting (Sorting for ORDER BY) - Expression ((Before ORDER BY + Projection)) + Expression ((Before ORDER BY + (Projection + DROP unused columns after JOIN))) Join (JOIN FillRightFirst) Expression ((Change column names to column identifiers + (Project names + (Before ORDER BY + (Projection + (Change column names to column identifiers + (Project names + (Before ORDER BY + (Projection + Change column names to column identifiers))))))))) ReadFromSystemNumbers diff --git a/tests/queries/0_stateless/02500_remove_redundant_distinct_analyzer.reference b/tests/queries/0_stateless/02500_remove_redundant_distinct_analyzer.reference index baa2be9dfdb..867ae394c1f 100644 --- a/tests/queries/0_stateless/02500_remove_redundant_distinct_analyzer.reference +++ b/tests/queries/0_stateless/02500_remove_redundant_distinct_analyzer.reference @@ -79,7 +79,7 @@ Expression (Project names) Sorting (Sorting for ORDER BY) Expression (Before ORDER BY) Distinct (Preliminary DISTINCT) - Expression (Projection) + Expression ((Projection + DROP unused columns after JOIN)) Join (JOIN FillRightFirst) Expression ((Change column names to column identifiers + Project names)) Distinct (DISTINCT) @@ -244,7 +244,7 @@ Expression ((Project names + (Projection + (Change column names to column identi Sorting (Sorting for ORDER BY) Expression ((Before ORDER BY + Projection)) Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -280,7 +280,7 @@ Expression (Project names) Sorting (Sorting for ORDER BY) Expression ((Before ORDER BY + Projection)) Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -315,7 +315,7 @@ Expression (Project names) Expression ((Before ORDER BY + Projection)) Rollup Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -348,7 +348,7 @@ Expression ((Project names + (Projection + (Change column names to column identi Expression ((Before ORDER BY + Projection)) Rollup Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -386,7 +386,7 @@ Expression (Project names) Expression ((Before ORDER BY + Projection)) Cube Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -419,7 +419,7 @@ Expression ((Project names + (Projection + (Change column names to column identi Expression ((Before ORDER BY + Projection)) Cube Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -457,7 +457,7 @@ Expression (Project names) Expression ((Before ORDER BY + Projection)) TotalsHaving Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers @@ -491,7 +491,7 @@ Expression ((Project names + (Projection + (Change column names to column identi Expression ((Before ORDER BY + Projection)) TotalsHaving Aggregating - Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + Projection)))) + Expression ((Before GROUP BY + (Change column names to column identifiers + (Project names + (Projection + DROP unused columns after JOIN))))) Join (JOIN FillRightFirst) Expression (Change column names to column identifiers) ReadFromSystemNumbers diff --git a/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference b/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference index bbfdf1ad5f4..2c62e278050 100644 --- a/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference +++ b/tests/queries/0_stateless/02514_analyzer_drop_join_on.reference @@ -8,21 +8,24 @@ Header: count() UInt64 Aggregating Header: __table1.a2 String count() UInt64 - Expression (Before GROUP BY) + Expression ((Before GROUP BY + DROP unused columns after JOIN)) Header: __table1.a2 String Join (JOIN FillRightFirst) Header: __table1.a2 String - Expression (JOIN actions) + __table3.c1 UInt64 + Expression ((JOIN actions + DROP unused columns after JOIN)) Header: __table1.a2 String __table3.c1 UInt64 Join (JOIN FillRightFirst) Header: __table1.a2 String + __table2.b1 UInt64 __table3.c1 UInt64 - Expression (JOIN actions) + Expression ((JOIN actions + DROP unused columns after JOIN)) Header: __table1.a2 String __table2.b1 UInt64 Join (JOIN FillRightFirst) - Header: __table1.a2 String + Header: __table1.a1 UInt64 + __table1.a2 String __table2.b1 UInt64 Expression ((JOIN actions + Change column names to column identifiers)) Header: __table1.a1 UInt64 @@ -45,32 +48,39 @@ Header: count() UInt64 EXPLAIN PLAN header = 1 SELECT a.a2, d.d2 FROM a JOIN b USING (k) JOIN c USING (k) JOIN d USING (k) ; -Expression ((Project names + Projection)) +Expression ((Project names + (Projection + DROP unused columns after JOIN))) Header: a2 String d2 String Join (JOIN FillRightFirst) Header: __table1.a2 String + __table1.k UInt64 __table4.d2 String - Join (JOIN FillRightFirst) + Expression (DROP unused columns after JOIN) Header: __table1.a2 String __table1.k UInt64 Join (JOIN FillRightFirst) Header: __table1.a2 String __table1.k UInt64 - Expression (Change column names to column identifiers) + Expression (DROP unused columns after JOIN) Header: __table1.a2 String __table1.k UInt64 - ReadFromMemoryStorage - Header: a2 String - k UInt64 + Join (JOIN FillRightFirst) + Header: __table1.a2 String + __table1.k UInt64 + Expression (Change column names to column identifiers) + Header: __table1.a2 String + __table1.k UInt64 + ReadFromMemoryStorage + Header: a2 String + k UInt64 + Expression (Change column names to column identifiers) + Header: __table2.k UInt64 + ReadFromMemoryStorage + Header: k UInt64 Expression (Change column names to column identifiers) - Header: __table2.k UInt64 + Header: __table3.k UInt64 ReadFromMemoryStorage Header: k UInt64 - Expression (Change column names to column identifiers) - Header: __table3.k UInt64 - ReadFromMemoryStorage - Header: k UInt64 Expression (Change column names to column identifiers) Header: __table4.d2 String __table4.k UInt64 @@ -96,24 +106,27 @@ Header: bx String Header: __table1.a2 String __table2.bx String __table4.c2 String + __table4.c1 UInt64 Expression Header: __table1.a2 String __table2.bx String - __table4.c1 UInt64 __table4.c2 String + __table4.c1 UInt64 Join (JOIN FillRightFirst) Header: __table1.a2 String __table2.bx String - __table4.c1 UInt64 + __table2.b1 UInt64 __table4.c2 String - Expression (JOIN actions) + __table4.c1 UInt64 + Expression ((JOIN actions + DROP unused columns after JOIN)) Header: __table1.a2 String - __table2.b1 UInt64 __table2.bx String + __table2.b1 UInt64 Join (JOIN FillRightFirst) - Header: __table1.a2 String - __table2.b1 UInt64 + Header: __table1.a1 UInt64 + __table1.a2 String __table2.bx String + __table2.b1 UInt64 Expression ((JOIN actions + Change column names to column identifiers)) Header: __table1.a1 UInt64 __table1.a2 String diff --git a/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql b/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql index b10bf38e495..df84e2f50b2 100644 --- a/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql +++ b/tests/queries/0_stateless/02514_analyzer_drop_join_on.sql @@ -16,7 +16,6 @@ CREATE TABLE d (k UInt64, d1 UInt64, d2 String) ENGINE = Memory; INSERT INTO d VALUES (1, 1, 'a'), (2, 2, 'b'), (3, 3, 'c'); SET enable_analyzer = 1; -SET query_plan_join_inner_table_selection = 'right'; -- { echoOn } diff --git a/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.reference b/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.reference index 116c78a15e4..86e7e2a6a49 100644 --- a/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.reference +++ b/tests/queries/0_stateless/02516_join_with_totals_and_subquery_bug.reference @@ -5,7 +5,7 @@ 1 1 -0 +1 \N 100000000000000000000 diff --git a/tests/queries/0_stateless/02835_join_step_explain.reference b/tests/queries/0_stateless/02835_join_step_explain.reference index bdbc019d4f8..06f4a9cfc99 100644 --- a/tests/queries/0_stateless/02835_join_step_explain.reference +++ b/tests/queries/0_stateless/02835_join_step_explain.reference @@ -1,22 +1,22 @@ -Expression ((Project names + Projection)) +Expression ((Project names + (Projection + DROP unused columns after JOIN))) Header: id UInt64 value_1 String rhs.id UInt64 rhs.value_1 String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value_1 String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value_1 String : 3 + INPUT : 2 -> __table2.value_1 String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value_1 :: 1 -> value_1 String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value_1 :: 3 -> rhs.value_1 String : 2 -Positions: 4 0 1 2 + ALIAS __table2.value_1 :: 2 -> rhs.value_1 String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 0 2 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value_1 String - __table2.id UInt64 __table2.value_1 String + __table2.id UInt64 Type: INNER Strictness: ALL Algorithm: HashJoin @@ -50,25 +50,29 @@ Positions: 4 0 1 2 Parts: 1 Granules: 1 -- -Expression ((Project names + Projection)) +Expression ((Project names + (Projection + DROP unused columns after JOIN))) Header: id UInt64 value_1 String rhs.id UInt64 rhs.value_1 String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value_1 String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 + INPUT :: 2 -> __table1.value_2 UInt64 : 2 INPUT : 3 -> __table2.value_1 String : 3 - ALIAS __table1.id :: 0 -> id UInt64 : 4 + INPUT :: 4 -> __table2.value_2 UInt64 : 4 + INPUT : 5 -> __table2.id UInt64 : 5 + ALIAS __table1.id :: 0 -> id UInt64 : 6 ALIAS __table1.value_1 :: 1 -> value_1 String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value_1 :: 3 -> rhs.value_1 String : 2 -Positions: 4 0 1 2 + ALIAS __table2.value_1 :: 3 -> rhs.value_1 String : 1 + ALIAS __table2.id :: 5 -> rhs.id UInt64 : 3 +Positions: 6 0 3 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value_1 String - __table2.id UInt64 + __table1.value_2 UInt64 __table2.value_1 String + __table2.value_2 UInt64 + __table2.id UInt64 Type: INNER Strictness: ASOF Algorithm: HashJoin diff --git a/tests/queries/0_stateless/02835_join_step_explain.sql b/tests/queries/0_stateless/02835_join_step_explain.sql index b803ddbd911..1cdd3684a0b 100644 --- a/tests/queries/0_stateless/02835_join_step_explain.sql +++ b/tests/queries/0_stateless/02835_join_step_explain.sql @@ -19,8 +19,6 @@ CREATE TABLE test_table_2 INSERT INTO test_table_1 VALUES (0, 'Value', 0); INSERT INTO test_table_2 VALUES (0, 'Value', 0); -SET query_plan_join_inner_table_selection = 'right'; - EXPLAIN header = 1, actions = 1 SELECT lhs.id, lhs.value_1, rhs.id, rhs.value_1 FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id; diff --git a/tests/queries/0_stateless/02962_join_using_bug_57894.reference b/tests/queries/0_stateless/02962_join_using_bug_57894.reference index fc6fe462205..454655081df 100644 --- a/tests/queries/0_stateless/02962_join_using_bug_57894.reference +++ b/tests/queries/0_stateless/02962_join_using_bug_57894.reference @@ -31,7 +31,6 @@ 8 9 \N ---- analyzer --- 0 1 2 diff --git a/tests/queries/0_stateless/02962_join_using_bug_57894.sql b/tests/queries/0_stateless/02962_join_using_bug_57894.sql index e29347beb5e..96190241da5 100644 --- a/tests/queries/0_stateless/02962_join_using_bug_57894.sql +++ b/tests/queries/0_stateless/02962_join_using_bug_57894.sql @@ -21,8 +21,6 @@ SETTINGS join_algorithm = 'partial_merge'; SELECT x FROM t FULL JOIN r USING (x) ORDER BY ALL SETTINGS join_algorithm = 'full_sorting_merge'; -SELECT '--- analyzer ---'; - SET enable_analyzer = 1; SELECT x FROM t FULL JOIN r USING (x) ORDER BY ALL diff --git a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference index b7718d926c6..80f4e309505 100644 --- a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference +++ b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference @@ -2,9 +2,7 @@ EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5; Expression ((Project names + (Projection + ))) Header: id UInt64 rhs.id UInt64 @@ -12,18 +10,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 1 0 2 + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 2 0 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: INNER Strictness: ALL Algorithm: HashJoin @@ -71,9 +69,7 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right'; -; +WHERE rhs.id = 5; Expression ((Project names + (Projection + ))) Header: id UInt64 rhs.id UInt64 @@ -81,18 +77,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 1 0 2 + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 2 0 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: INNER Strictness: ALL Algorithm: HashJoin @@ -140,9 +136,7 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 AND rhs.id = 6 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5 AND rhs.id = 6; Expression ((Project names + (Projection + ))) Header: id UInt64 rhs.id UInt64 @@ -150,18 +144,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 1 0 2 + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 2 0 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: INNER Strictness: ALL Algorithm: HashJoin @@ -212,9 +206,7 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5; Expression ((Project names + (Projection + ))) Header: id UInt64 rhs.id UInt64 @@ -222,18 +214,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 1 0 2 + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 2 0 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: LEFT Strictness: ALL Algorithm: HashJoin @@ -281,9 +273,7 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE rhs.id = 5; Expression ((Project names + Projection)) Header: id UInt64 rhs.id UInt64 @@ -291,31 +281,31 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 1 0 2 - Filter (WHERE) + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 2 0 1 + Filter ((WHERE + DROP unused columns after JOIN)) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Filter column: equals(__table2.id, 5_UInt8) (removed) Actions: INPUT :: 0 -> __table1.id UInt64 : 0 INPUT :: 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT :: 3 -> __table2.value String : 3 + INPUT :: 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 4 - FUNCTION equals(__table2.id : 2, 5_UInt8 :: 4) -> equals(__table2.id, 5_UInt8) UInt8 : 5 + FUNCTION equals(__table2.id : 3, 5_UInt8 :: 4) -> equals(__table2.id, 5_UInt8) UInt8 : 5 Positions: 5 0 1 2 3 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: LEFT Strictness: ALL Algorithm: HashJoin @@ -357,9 +347,7 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5; Expression ((Project names + Projection)) Header: id UInt64 rhs.id UInt64 @@ -367,31 +355,31 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 1 0 2 - Filter (WHERE) + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 2 0 1 + Filter ((WHERE + DROP unused columns after JOIN)) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Filter column: equals(__table1.id, 5_UInt8) (removed) Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT :: 1 -> __table1.value String : 1 - INPUT :: 2 -> __table2.id UInt64 : 2 - INPUT :: 3 -> __table2.value String : 3 + INPUT :: 2 -> __table2.value String : 2 + INPUT :: 3 -> __table2.id UInt64 : 3 COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 4 FUNCTION equals(__table1.id : 0, 5_UInt8 :: 4) -> equals(__table1.id, 5_UInt8) UInt8 : 5 Positions: 5 0 1 2 3 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: RIGHT Strictness: ALL Algorithm: HashJoin @@ -433,9 +421,7 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE rhs.id = 5; Expression ((Project names + (Projection + ))) Header: id UInt64 rhs.id UInt64 @@ -443,18 +429,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 1 0 2 + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 2 0 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: RIGHT Strictness: ALL Algorithm: HashJoin @@ -502,9 +488,7 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5; Expression ((Project names + Projection)) Header: id UInt64 rhs.id UInt64 @@ -512,31 +496,31 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 1 0 2 - Filter (WHERE) + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 2 0 1 + Filter ((WHERE + DROP unused columns after JOIN)) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Filter column: equals(__table1.id, 5_UInt8) (removed) Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT :: 1 -> __table1.value String : 1 - INPUT :: 2 -> __table2.id UInt64 : 2 - INPUT :: 3 -> __table2.value String : 3 + INPUT :: 2 -> __table2.value String : 2 + INPUT :: 3 -> __table2.id UInt64 : 3 COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 4 FUNCTION equals(__table1.id : 0, 5_UInt8 :: 4) -> equals(__table1.id, 5_UInt8) UInt8 : 5 Positions: 5 0 1 2 3 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: FULL Strictness: ALL Algorithm: HashJoin @@ -578,9 +562,7 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE rhs.id = 5; Expression ((Project names + Projection)) Header: id UInt64 rhs.id UInt64 @@ -588,31 +570,31 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 1 0 2 - Filter (WHERE) + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 2 0 1 + Filter ((WHERE + DROP unused columns after JOIN)) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Filter column: equals(__table2.id, 5_UInt8) (removed) Actions: INPUT :: 0 -> __table1.id UInt64 : 0 INPUT :: 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT :: 3 -> __table2.value String : 3 + INPUT :: 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 4 - FUNCTION equals(__table2.id : 2, 5_UInt8 :: 4) -> equals(__table2.id, 5_UInt8) UInt8 : 5 + FUNCTION equals(__table2.id : 3, 5_UInt8 :: 4) -> equals(__table2.id, 5_UInt8) UInt8 : 5 Positions: 5 0 1 2 3 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: FULL Strictness: ALL Algorithm: HashJoin @@ -654,9 +636,7 @@ SELECT '--'; -- EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 AND rhs.id = 6 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5 AND rhs.id = 6; Expression ((Project names + Projection)) Header: id UInt64 rhs.id UInt64 @@ -664,34 +644,34 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 1 0 2 - Filter (WHERE) + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 2 0 1 + Filter ((WHERE + DROP unused columns after JOIN)) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Filter column: and(equals(__table1.id, 5_UInt8), equals(__table2.id, 6_UInt8)) (removed) Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT :: 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT :: 3 -> __table2.value String : 3 + INPUT :: 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 4 COLUMN Const(UInt8) -> 6_UInt8 UInt8 : 5 FUNCTION equals(__table1.id : 0, 5_UInt8 :: 4) -> equals(__table1.id, 5_UInt8) UInt8 : 6 - FUNCTION equals(__table2.id : 2, 6_UInt8 :: 5) -> equals(__table2.id, 6_UInt8) UInt8 : 4 + FUNCTION equals(__table2.id : 3, 6_UInt8 :: 5) -> equals(__table2.id, 6_UInt8) UInt8 : 4 FUNCTION and(equals(__table1.id, 5_UInt8) :: 6, equals(__table2.id, 6_UInt8) :: 4) -> and(equals(__table1.id, 5_UInt8), equals(__table2.id, 6_UInt8)) UInt8 : 5 Positions: 5 0 1 2 3 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: FULL Strictness: ALL Algorithm: HashJoin diff --git a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql index d6dcc34c796..e1a13d1ce71 100644 --- a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql +++ b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.sql @@ -22,9 +22,7 @@ INSERT INTO test_table_2 SELECT number, number FROM numbers(10); EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5; SELECT '--'; @@ -35,9 +33,7 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right'; -; +WHERE rhs.id = 5; SELECT '--'; @@ -48,9 +44,7 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 AND rhs.id = 6 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5 AND rhs.id = 6; SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs INNER JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE lhs.id = 5 AND rhs.id = 6; @@ -59,9 +53,7 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5; SELECT '--'; @@ -72,9 +64,7 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE rhs.id = 5; SELECT '--'; @@ -85,9 +75,7 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5; SELECT '--'; @@ -98,9 +86,7 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE rhs.id = 5; SELECT '--'; @@ -111,9 +97,7 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5; SELECT '--'; @@ -124,9 +108,7 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE rhs.id = 5 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE rhs.id = 5; SELECT '--'; @@ -137,9 +119,7 @@ SELECT '--'; EXPLAIN header = 1, actions = 1 SELECT lhs.id, rhs.id, lhs.value, rhs.value FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id -WHERE lhs.id = 5 AND rhs.id = 6 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +WHERE lhs.id = 5 AND rhs.id = 6; SELECT '--'; diff --git a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference index 7df38e855f6..cf070eebc38 100644 --- a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference +++ b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.reference @@ -52,9 +52,7 @@ WITH RECURSIVE search_graph AS ( FROM graph g, search_graph sg WHERE g.f = sg.t AND NOT is_cycle ) -SELECT * FROM search_graph -SETTINGS query_plan_join_inner_table_selection = 'right' -; +SELECT * FROM search_graph; 1 2 arc 1 -> 2 false [(1,2)] 1 3 arc 1 -> 3 false [(1,3)] 2 3 arc 2 -> 3 false [(2,3)] diff --git a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql index d33ca7b078e..7dad74893b9 100644 --- a/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql +++ b/tests/queries/0_stateless/03038_recursive_cte_postgres_4.sql @@ -55,9 +55,7 @@ WITH RECURSIVE search_graph AS ( FROM graph g, search_graph sg WHERE g.f = sg.t AND NOT is_cycle ) -SELECT * FROM search_graph -SETTINGS query_plan_join_inner_table_selection = 'right' -; +SELECT * FROM search_graph; -- ordering by the path column has same effect as SEARCH DEPTH FIRST WITH RECURSIVE search_graph AS ( diff --git a/tests/queries/0_stateless/03094_one_thousand_joins.sql b/tests/queries/0_stateless/03094_one_thousand_joins.sql index 69c4fb42a6b..6ae4e4d4d3c 100644 --- a/tests/queries/0_stateless/03094_one_thousand_joins.sql +++ b/tests/queries/0_stateless/03094_one_thousand_joins.sql @@ -3,7 +3,6 @@ SET join_algorithm = 'default'; -- for 'full_sorting_merge' the query is 10x slower SET enable_analyzer = 1; -- old analyzer returns TOO_DEEP_SUBQUERIES -SET query_plan_join_inner_table_selection = 'auto'; -- 'left' is slower -- Bug 33446, marked as 'long' because it still runs around 10 sec SELECT * FROM (SELECT 1 AS x) t1 JOIN (SELECT 1 AS x) t2 ON t1.x = t2.x JOIN (SELECT 1 AS x) t3 ON t1.x = t3.x JOIN (SELECT 1 AS x) t4 ON t1.x = t4.x JOIN (SELECT 1 AS x) t5 ON t1.x = t5.x JOIN (SELECT 1 AS x) t6 ON t1.x = t6.x JOIN (SELECT 1 AS x) t7 ON t1.x = t7.x JOIN (SELECT 1 AS x) t8 ON t1.x = t8.x JOIN (SELECT 1 AS x) t9 ON t1.x = t9.x JOIN (SELECT 1 AS x) t10 ON t1.x = t10.x JOIN (SELECT 1 AS x) t11 ON t1.x = t11.x JOIN (SELECT 1 AS x) t12 ON t1.x = t12.x JOIN (SELECT 1 AS x) t13 ON t1.x = t13.x JOIN (SELECT 1 AS x) t14 ON t1.x = t14.x JOIN (SELECT 1 AS x) t15 ON t1.x = t15.x JOIN (SELECT 1 AS x) t16 ON t1.x = t16.x JOIN (SELECT 1 AS x) t17 ON t1.x = t17.x JOIN (SELECT 1 AS x) t18 ON t1.x = t18.x JOIN (SELECT 1 AS x) t19 ON t1.x = t19.x JOIN (SELECT 1 AS x) t20 ON t1.x = t20.x JOIN (SELECT 1 AS x) t21 ON t1.x = t21.x JOIN (SELECT 1 AS x) t22 ON t1.x = t22.x JOIN (SELECT 1 AS x) t23 ON t1.x = t23.x JOIN (SELECT 1 AS x) t24 ON t1.x = t24.x JOIN (SELECT 1 AS x) t25 ON t1.x = t25.x JOIN (SELECT 1 AS x) t26 ON t1.x = t26.x JOIN (SELECT 1 AS x) t27 ON t1.x = t27.x JOIN (SELECT 1 AS x) t28 ON t1.x = t28.x JOIN (SELECT 1 AS x) t29 ON t1.x = t29.x JOIN (SELECT 1 AS x) t30 ON t1.x = t30.x JOIN (SELECT 1 AS x) t31 ON t1.x = t31.x JOIN (SELECT 1 AS x) t32 ON t1.x = t32.x JOIN (SELECT 1 AS x) t33 ON t1.x = t33.x JOIN (SELECT 1 AS x) t34 ON t1.x = t34.x JOIN (SELECT 1 AS x) t35 ON t1.x = t35.x JOIN (SELECT 1 AS x) t36 ON t1.x = t36.x JOIN (SELECT 1 AS x) t37 ON t1.x = t37.x JOIN (SELECT 1 AS x) t38 ON t1.x = t38.x JOIN (SELECT 1 AS x) t39 ON t1.x = t39.x JOIN (SELECT 1 AS x) t40 ON t1.x = t40.x JOIN (SELECT 1 AS x) t41 ON t1.x = t41.x JOIN (SELECT 1 AS x) t42 ON t1.x = t42.x JOIN (SELECT 1 AS x) t43 ON t1.x = t43.x JOIN (SELECT 1 AS x) t44 ON t1.x = t44.x JOIN (SELECT 1 AS x) t45 ON t1.x = t45.x JOIN (SELECT 1 AS x) t46 ON t1.x = t46.x JOIN (SELECT 1 AS x) t47 ON t1.x = t47.x JOIN (SELECT 1 AS x) t48 ON t1.x = t48.x JOIN (SELECT 1 AS x) t49 ON t1.x = t49.x JOIN (SELECT 1 AS x) t50 ON t1.x = t50.x JOIN (SELECT 1 AS x) t51 ON t1.x = t51.x JOIN (SELECT 1 AS x) t52 ON t1.x = t52.x JOIN (SELECT 1 AS x) t53 ON t1.x = t53.x JOIN (SELECT 1 AS x) t54 ON t1.x = t54.x JOIN (SELECT 1 AS x) t55 ON t1.x = t55.x JOIN (SELECT 1 AS x) t56 ON t1.x = t56.x JOIN (SELECT 1 AS x) t57 ON t1.x = t57.x JOIN (SELECT 1 AS x) t58 ON t1.x = t58.x JOIN (SELECT 1 AS x) t59 ON t1.x = t59.x JOIN (SELECT 1 AS x) t60 ON t1.x = t60.x JOIN (SELECT 1 AS x) t61 ON t1.x = t61.x JOIN (SELECT 1 AS x) t62 ON t1.x = t62.x JOIN (SELECT 1 AS x) t63 ON t1.x = t63.x JOIN (SELECT 1 AS x) t64 ON t1.x = t64.x JOIN (SELECT 1 AS x) t65 ON t1.x = t65.x JOIN (SELECT 1 AS x) t66 ON t1.x = t66.x JOIN (SELECT 1 AS x) t67 ON t1.x = t67.x JOIN (SELECT 1 AS x) t68 ON t1.x = t68.x JOIN (SELECT 1 AS x) t69 ON t1.x = t69.x JOIN (SELECT 1 AS x) t70 ON t1.x = t70.x JOIN (SELECT 1 AS x) t71 ON t1.x = t71.x JOIN (SELECT 1 AS x) t72 ON t1.x = t72.x JOIN (SELECT 1 AS x) t73 ON t1.x = t73.x JOIN (SELECT 1 AS x) t74 ON t1.x = t74.x JOIN (SELECT 1 AS x) t75 ON t1.x = t75.x JOIN (SELECT 1 AS x) t76 ON t1.x = t76.x JOIN (SELECT 1 AS x) t77 ON t1.x = t77.x JOIN (SELECT 1 AS x) t78 ON t1.x = t78.x JOIN (SELECT 1 AS x) t79 ON t1.x = t79.x JOIN (SELECT 1 AS x) t80 ON t1.x = t80.x JOIN (SELECT 1 AS x) t81 ON t1.x = t81.x JOIN (SELECT 1 AS x) t82 ON t1.x = t82.x JOIN (SELECT 1 AS x) t83 ON t1.x = t83.x JOIN (SELECT 1 AS x) t84 ON t1.x = t84.x JOIN (SELECT 1 AS x) t85 ON t1.x = t85.x JOIN (SELECT 1 AS x) t86 ON t1.x = t86.x JOIN (SELECT 1 AS x) t87 ON t1.x = t87.x JOIN (SELECT 1 AS x) t88 ON t1.x = t88.x JOIN (SELECT 1 AS x) t89 ON t1.x = t89.x JOIN (SELECT 1 AS x) t90 ON t1.x = t90.x JOIN (SELECT 1 AS x) t91 ON t1.x = t91.x JOIN (SELECT 1 AS x) t92 ON t1.x = t92.x JOIN (SELECT 1 AS x) t93 ON t1.x = t93.x JOIN (SELECT 1 AS x) t94 ON t1.x = t94.x JOIN (SELECT 1 AS x) t95 ON t1.x = t95.x JOIN (SELECT 1 AS x) t96 ON t1.x = t96.x JOIN (SELECT 1 AS x) t97 ON t1.x = t97.x JOIN (SELECT 1 AS x) t98 ON t1.x = t98.x JOIN (SELECT 1 AS x) t99 ON t1.x = t99.x JOIN (SELECT 1 AS x) t100 ON t1.x = t100.x JOIN (SELECT 1 AS x) t101 ON t1.x = t101.x JOIN (SELECT 1 AS x) t102 ON t1.x = t102.x JOIN (SELECT 1 AS x) t103 ON t1.x = t103.x JOIN (SELECT 1 AS x) t104 ON t1.x = t104.x JOIN (SELECT 1 AS x) t105 ON t1.x = t105.x JOIN (SELECT 1 AS x) t106 ON t1.x = t106.x JOIN (SELECT 1 AS x) t107 ON t1.x = t107.x JOIN (SELECT 1 AS x) t108 ON t1.x = t108.x JOIN (SELECT 1 AS x) t109 ON t1.x = t109.x JOIN (SELECT 1 AS x) t110 ON t1.x = t110.x JOIN (SELECT 1 AS x) t111 ON t1.x = t111.x JOIN (SELECT 1 AS x) t112 ON t1.x = t112.x JOIN (SELECT 1 AS x) t113 ON t1.x = t113.x JOIN (SELECT 1 AS x) t114 ON t1.x = t114.x JOIN (SELECT 1 AS x) t115 ON t1.x = t115.x JOIN (SELECT 1 AS x) t116 ON t1.x = t116.x JOIN (SELECT 1 AS x) t117 ON t1.x = t117.x JOIN (SELECT 1 AS x) t118 ON t1.x = t118.x JOIN (SELECT 1 AS x) t119 ON t1.x = t119.x JOIN (SELECT 1 AS x) t120 ON t1.x = t120.x JOIN (SELECT 1 AS x) t121 ON t1.x = t121.x JOIN (SELECT 1 AS x) t122 ON t1.x = t122.x JOIN (SELECT 1 AS x) t123 ON t1.x = t123.x JOIN (SELECT 1 AS x) t124 ON t1.x = t124.x JOIN (SELECT 1 AS x) t125 ON t1.x = t125.x JOIN (SELECT 1 AS x) t126 ON t1.x = t126.x JOIN (SELECT 1 AS x) t127 ON t1.x = t127.x JOIN (SELECT 1 AS x) t128 ON t1.x = t128.x JOIN (SELECT 1 AS x) t129 ON t1.x = t129.x JOIN (SELECT 1 AS x) t130 ON t1.x = t130.x JOIN (SELECT 1 AS x) t131 ON t1.x = t131.x JOIN (SELECT 1 AS x) t132 ON t1.x = t132.x JOIN (SELECT 1 AS x) t133 ON t1.x = t133.x JOIN (SELECT 1 AS x) t134 ON t1.x = t134.x JOIN (SELECT 1 AS x) t135 ON t1.x = t135.x JOIN (SELECT 1 AS x) t136 ON t1.x = t136.x JOIN (SELECT 1 AS x) t137 ON t1.x = t137.x JOIN (SELECT 1 AS x) t138 ON t1.x = t138.x JOIN (SELECT 1 AS x) t139 ON t1.x = t139.x JOIN (SELECT 1 AS x) t140 ON t1.x = t140.x JOIN (SELECT 1 AS x) t141 ON t1.x = t141.x JOIN (SELECT 1 AS x) t142 ON t1.x = t142.x JOIN (SELECT 1 AS x) t143 ON t1.x = t143.x JOIN (SELECT 1 AS x) t144 ON t1.x = t144.x JOIN (SELECT 1 AS x) t145 ON t1.x = t145.x JOIN (SELECT 1 AS x) t146 ON t1.x = t146.x JOIN (SELECT 1 AS x) t147 ON t1.x = t147.x JOIN (SELECT 1 AS x) t148 ON t1.x = t148.x JOIN (SELECT 1 AS x) t149 ON t1.x = t149.x JOIN (SELECT 1 AS x) t150 ON t1.x = t150.x JOIN (SELECT 1 AS x) t151 ON t1.x = t151.x JOIN (SELECT 1 AS x) t152 ON t1.x = t152.x JOIN (SELECT 1 AS x) t153 ON t1.x = t153.x JOIN (SELECT 1 AS x) t154 ON t1.x = t154.x JOIN (SELECT 1 AS x) t155 ON t1.x = t155.x JOIN (SELECT 1 AS x) t156 ON t1.x = t156.x JOIN (SELECT 1 AS x) t157 ON t1.x = t157.x JOIN (SELECT 1 AS x) t158 ON t1.x = t158.x JOIN (SELECT 1 AS x) t159 ON t1.x = t159.x JOIN (SELECT 1 AS x) t160 ON t1.x = t160.x JOIN (SELECT 1 AS x) t161 ON t1.x = t161.x JOIN (SELECT 1 AS x) t162 ON t1.x = t162.x JOIN (SELECT 1 AS x) t163 ON t1.x = t163.x JOIN (SELECT 1 AS x) t164 ON t1.x = t164.x JOIN (SELECT 1 AS x) t165 ON t1.x = t165.x JOIN (SELECT 1 AS x) t166 ON t1.x = t166.x JOIN (SELECT 1 AS x) t167 ON t1.x = t167.x JOIN (SELECT 1 AS x) t168 ON t1.x = t168.x JOIN (SELECT 1 AS x) t169 ON t1.x = t169.x JOIN (SELECT 1 AS x) t170 ON t1.x = t170.x JOIN (SELECT 1 AS x) t171 ON t1.x = t171.x JOIN (SELECT 1 AS x) t172 ON t1.x = t172.x JOIN (SELECT 1 AS x) t173 ON t1.x = t173.x JOIN (SELECT 1 AS x) t174 ON t1.x = t174.x JOIN (SELECT 1 AS x) t175 ON t1.x = t175.x JOIN (SELECT 1 AS x) t176 ON t1.x = t176.x JOIN (SELECT 1 AS x) t177 ON t1.x = t177.x JOIN (SELECT 1 AS x) t178 ON t1.x = t178.x JOIN (SELECT 1 AS x) t179 ON t1.x = t179.x JOIN (SELECT 1 AS x) t180 ON t1.x = t180.x JOIN (SELECT 1 AS x) t181 ON t1.x = t181.x JOIN (SELECT 1 AS x) t182 ON t1.x = t182.x JOIN (SELECT 1 AS x) t183 ON t1.x = t183.x JOIN (SELECT 1 AS x) t184 ON t1.x = t184.x JOIN (SELECT 1 AS x) t185 ON t1.x = t185.x JOIN (SELECT 1 AS x) t186 ON t1.x = t186.x JOIN (SELECT 1 AS x) t187 ON t1.x = t187.x JOIN (SELECT 1 AS x) t188 ON t1.x = t188.x JOIN (SELECT 1 AS x) t189 ON t1.x = t189.x JOIN (SELECT 1 AS x) t190 ON t1.x = t190.x JOIN (SELECT 1 AS x) t191 ON t1.x = t191.x JOIN (SELECT 1 AS x) t192 ON t1.x = t192.x JOIN (SELECT 1 AS x) t193 ON t1.x = t193.x JOIN (SELECT 1 AS x) t194 ON t1.x = t194.x JOIN (SELECT 1 AS x) t195 ON t1.x = t195.x JOIN (SELECT 1 AS x) t196 ON t1.x = t196.x JOIN (SELECT 1 AS x) t197 ON t1.x = t197.x JOIN (SELECT 1 AS x) t198 ON t1.x = t198.x JOIN (SELECT 1 AS x) t199 ON t1.x = t199.x JOIN (SELECT 1 AS x) t200 ON t1.x = t200.x JOIN (SELECT 1 AS x) t201 ON t1.x = t201.x JOIN (SELECT 1 AS x) t202 ON t1.x = t202.x JOIN (SELECT 1 AS x) t203 ON t1.x = t203.x JOIN (SELECT 1 AS x) t204 ON t1.x = t204.x JOIN (SELECT 1 AS x) t205 ON t1.x = t205.x JOIN (SELECT 1 AS x) t206 ON t1.x = t206.x JOIN (SELECT 1 AS x) t207 ON t1.x = t207.x JOIN (SELECT 1 AS x) t208 ON t1.x = t208.x JOIN (SELECT 1 AS x) t209 ON t1.x = t209.x JOIN (SELECT 1 AS x) t210 ON t1.x = t210.x JOIN (SELECT 1 AS x) t211 ON t1.x = t211.x JOIN (SELECT 1 AS x) t212 ON t1.x = t212.x JOIN (SELECT 1 AS x) t213 ON t1.x = t213.x JOIN (SELECT 1 AS x) t214 ON t1.x = t214.x JOIN (SELECT 1 AS x) t215 ON t1.x = t215.x JOIN (SELECT 1 AS x) t216 ON t1.x = t216.x JOIN (SELECT 1 AS x) t217 ON t1.x = t217.x JOIN (SELECT 1 AS x) t218 ON t1.x = t218.x JOIN (SELECT 1 AS x) t219 ON t1.x = t219.x JOIN (SELECT 1 AS x) t220 ON t1.x = t220.x JOIN (SELECT 1 AS x) t221 ON t1.x = t221.x JOIN (SELECT 1 AS x) t222 ON t1.x = t222.x JOIN (SELECT 1 AS x) t223 ON t1.x = t223.x JOIN (SELECT 1 AS x) t224 ON t1.x = t224.x JOIN (SELECT 1 AS x) t225 ON t1.x = t225.x JOIN (SELECT 1 AS x) t226 ON t1.x = t226.x JOIN (SELECT 1 AS x) t227 ON t1.x = t227.x JOIN (SELECT 1 AS x) t228 ON t1.x = t228.x JOIN (SELECT 1 AS x) t229 ON t1.x = t229.x JOIN (SELECT 1 AS x) t230 ON t1.x = t230.x JOIN (SELECT 1 AS x) t231 ON t1.x = t231.x JOIN (SELECT 1 AS x) t232 ON t1.x = t232.x JOIN (SELECT 1 AS x) t233 ON t1.x = t233.x JOIN (SELECT 1 AS x) t234 ON t1.x = t234.x JOIN (SELECT 1 AS x) t235 ON t1.x = t235.x JOIN (SELECT 1 AS x) t236 ON t1.x = t236.x JOIN (SELECT 1 AS x) t237 ON t1.x = t237.x JOIN (SELECT 1 AS x) t238 ON t1.x = t238.x JOIN (SELECT 1 AS x) t239 ON t1.x = t239.x JOIN (SELECT 1 AS x) t240 ON t1.x = t240.x JOIN (SELECT 1 AS x) t241 ON t1.x = t241.x JOIN (SELECT 1 AS x) t242 ON t1.x = t242.x JOIN (SELECT 1 AS x) t243 ON t1.x = t243.x JOIN (SELECT 1 AS x) t244 ON t1.x = t244.x JOIN (SELECT 1 AS x) t245 ON t1.x = t245.x JOIN (SELECT 1 AS x) t246 ON t1.x = t246.x JOIN (SELECT 1 AS x) t247 ON t1.x = t247.x JOIN (SELECT 1 AS x) t248 ON t1.x = t248.x JOIN (SELECT 1 AS x) t249 ON t1.x = t249.x JOIN (SELECT 1 AS x) t250 ON t1.x = t250.x JOIN (SELECT 1 AS x) t251 ON t1.x = t251.x JOIN (SELECT 1 AS x) t252 ON t1.x = t252.x JOIN (SELECT 1 AS x) t253 ON t1.x = t253.x JOIN (SELECT 1 AS x) t254 ON t1.x = t254.x JOIN (SELECT 1 AS x) t255 ON t1.x = t255.x JOIN (SELECT 1 AS x) t256 ON t1.x = t256.x JOIN (SELECT 1 AS x) t257 ON t1.x = t257.x JOIN (SELECT 1 AS x) t258 ON t1.x = t258.x JOIN (SELECT 1 AS x) t259 ON t1.x = t259.x JOIN (SELECT 1 AS x) t260 ON t1.x = t260.x JOIN (SELECT 1 AS x) t261 ON t1.x = t261.x JOIN (SELECT 1 AS x) t262 ON t1.x = t262.x JOIN (SELECT 1 AS x) t263 ON t1.x = t263.x JOIN (SELECT 1 AS x) t264 ON t1.x = t264.x JOIN (SELECT 1 AS x) t265 ON t1.x = t265.x JOIN (SELECT 1 AS x) t266 ON t1.x = t266.x JOIN (SELECT 1 AS x) t267 ON t1.x = t267.x JOIN (SELECT 1 AS x) t268 ON t1.x = t268.x JOIN (SELECT 1 AS x) t269 ON t1.x = t269.x JOIN (SELECT 1 AS x) t270 ON t1.x = t270.x JOIN (SELECT 1 AS x) t271 ON t1.x = t271.x JOIN (SELECT 1 AS x) t272 ON t1.x = t272.x JOIN (SELECT 1 AS x) t273 ON t1.x = t273.x JOIN (SELECT 1 AS x) t274 ON t1.x = t274.x JOIN (SELECT 1 AS x) t275 ON t1.x = t275.x JOIN (SELECT 1 AS x) t276 ON t1.x = t276.x JOIN (SELECT 1 AS x) t277 ON t1.x = t277.x JOIN (SELECT 1 AS x) t278 ON t1.x = t278.x JOIN (SELECT 1 AS x) t279 ON t1.x = t279.x JOIN (SELECT 1 AS x) t280 ON t1.x = t280.x JOIN (SELECT 1 AS x) t281 ON t1.x = t281.x JOIN (SELECT 1 AS x) t282 ON t1.x = t282.x JOIN (SELECT 1 AS x) t283 ON t1.x = t283.x JOIN (SELECT 1 AS x) t284 ON t1.x = t284.x JOIN (SELECT 1 AS x) t285 ON t1.x = t285.x JOIN (SELECT 1 AS x) t286 ON t1.x = t286.x JOIN (SELECT 1 AS x) t287 ON t1.x = t287.x JOIN (SELECT 1 AS x) t288 ON t1.x = t288.x JOIN (SELECT 1 AS x) t289 ON t1.x = t289.x JOIN (SELECT 1 AS x) t290 ON t1.x = t290.x JOIN (SELECT 1 AS x) t291 ON t1.x = t291.x JOIN (SELECT 1 AS x) t292 ON t1.x = t292.x JOIN (SELECT 1 AS x) t293 ON t1.x = t293.x JOIN (SELECT 1 AS x) t294 ON t1.x = t294.x JOIN (SELECT 1 AS x) t295 ON t1.x = t295.x JOIN (SELECT 1 AS x) t296 ON t1.x = t296.x JOIN (SELECT 1 AS x) t297 ON t1.x = t297.x JOIN (SELECT 1 AS x) t298 ON t1.x = t298.x JOIN (SELECT 1 AS x) t299 ON t1.x = t299.x JOIN (SELECT 1 AS x) t300 ON t1.x = t300.x JOIN (SELECT 1 AS x) t301 ON t1.x = t301.x JOIN (SELECT 1 AS x) t302 ON t1.x = t302.x JOIN (SELECT 1 AS x) t303 ON t1.x = t303.x JOIN (SELECT 1 AS x) t304 ON t1.x = t304.x JOIN (SELECT 1 AS x) t305 ON t1.x = t305.x JOIN (SELECT 1 AS x) t306 ON t1.x = t306.x JOIN (SELECT 1 AS x) t307 ON t1.x = t307.x JOIN (SELECT 1 AS x) t308 ON t1.x = t308.x JOIN (SELECT 1 AS x) t309 ON t1.x = t309.x JOIN (SELECT 1 AS x) t310 ON t1.x = t310.x JOIN (SELECT 1 AS x) t311 ON t1.x = t311.x JOIN (SELECT 1 AS x) t312 ON t1.x = t312.x JOIN (SELECT 1 AS x) t313 ON t1.x = t313.x JOIN (SELECT 1 AS x) t314 ON t1.x = t314.x JOIN (SELECT 1 AS x) t315 ON t1.x = t315.x JOIN (SELECT 1 AS x) t316 ON t1.x = t316.x JOIN (SELECT 1 AS x) t317 ON t1.x = t317.x JOIN (SELECT 1 AS x) t318 ON t1.x = t318.x JOIN (SELECT 1 AS x) t319 ON t1.x = t319.x JOIN (SELECT 1 AS x) t320 ON t1.x = t320.x JOIN (SELECT 1 AS x) t321 ON t1.x = t321.x JOIN (SELECT 1 AS x) t322 ON t1.x = t322.x JOIN (SELECT 1 AS x) t323 ON t1.x = t323.x JOIN (SELECT 1 AS x) t324 ON t1.x = t324.x JOIN (SELECT 1 AS x) t325 ON t1.x = t325.x JOIN (SELECT 1 AS x) t326 ON t1.x = t326.x JOIN (SELECT 1 AS x) t327 ON t1.x = t327.x JOIN (SELECT 1 AS x) t328 ON t1.x = t328.x JOIN (SELECT 1 AS x) t329 ON t1.x = t329.x JOIN (SELECT 1 AS x) t330 ON t1.x = t330.x JOIN (SELECT 1 AS x) t331 ON t1.x = t331.x JOIN (SELECT 1 AS x) t332 ON t1.x = t332.x JOIN (SELECT 1 AS x) t333 ON t1.x = t333.x JOIN (SELECT 1 AS x) t334 ON t1.x = t334.x JOIN (SELECT 1 AS x) t335 ON t1.x = t335.x JOIN (SELECT 1 AS x) t336 ON t1.x = t336.x JOIN (SELECT 1 AS x) t337 ON t1.x = t337.x JOIN (SELECT 1 AS x) t338 ON t1.x = t338.x JOIN (SELECT 1 AS x) t339 ON t1.x = t339.x JOIN (SELECT 1 AS x) t340 ON t1.x = t340.x JOIN (SELECT 1 AS x) t341 ON t1.x = t341.x JOIN (SELECT 1 AS x) t342 ON t1.x = t342.x JOIN (SELECT 1 AS x) t343 ON t1.x = t343.x JOIN (SELECT 1 AS x) t344 ON t1.x = t344.x JOIN (SELECT 1 AS x) t345 ON t1.x = t345.x JOIN (SELECT 1 AS x) t346 ON t1.x = t346.x JOIN (SELECT 1 AS x) t347 ON t1.x = t347.x JOIN (SELECT 1 AS x) t348 ON t1.x = t348.x JOIN (SELECT 1 AS x) t349 ON t1.x = t349.x JOIN (SELECT 1 AS x) t350 ON t1.x = t350.x JOIN (SELECT 1 AS x) t351 ON t1.x = t351.x JOIN (SELECT 1 AS x) t352 ON t1.x = t352.x JOIN (SELECT 1 AS x) t353 ON t1.x = t353.x JOIN (SELECT 1 AS x) t354 ON t1.x = t354.x JOIN (SELECT 1 AS x) t355 ON t1.x = t355.x JOIN (SELECT 1 AS x) t356 ON t1.x = t356.x JOIN (SELECT 1 AS x) t357 ON t1.x = t357.x JOIN (SELECT 1 AS x) t358 ON t1.x = t358.x JOIN (SELECT 1 AS x) t359 ON t1.x = t359.x JOIN (SELECT 1 AS x) t360 ON t1.x = t360.x JOIN (SELECT 1 AS x) t361 ON t1.x = t361.x JOIN (SELECT 1 AS x) t362 ON t1.x = t362.x JOIN (SELECT 1 AS x) t363 ON t1.x = t363.x JOIN (SELECT 1 AS x) t364 ON t1.x = t364.x JOIN (SELECT 1 AS x) t365 ON t1.x = t365.x JOIN (SELECT 1 AS x) t366 ON t1.x = t366.x JOIN (SELECT 1 AS x) t367 ON t1.x = t367.x JOIN (SELECT 1 AS x) t368 ON t1.x = t368.x JOIN (SELECT 1 AS x) t369 ON t1.x = t369.x JOIN (SELECT 1 AS x) t370 ON t1.x = t370.x JOIN (SELECT 1 AS x) t371 ON t1.x = t371.x JOIN (SELECT 1 AS x) t372 ON t1.x = t372.x JOIN (SELECT 1 AS x) t373 ON t1.x = t373.x JOIN (SELECT 1 AS x) t374 ON t1.x = t374.x JOIN (SELECT 1 AS x) t375 ON t1.x = t375.x JOIN (SELECT 1 AS x) t376 ON t1.x = t376.x JOIN (SELECT 1 AS x) t377 ON t1.x = t377.x JOIN (SELECT 1 AS x) t378 ON t1.x = t378.x JOIN (SELECT 1 AS x) t379 ON t1.x = t379.x JOIN (SELECT 1 AS x) t380 ON t1.x = t380.x JOIN (SELECT 1 AS x) t381 ON t1.x = t381.x JOIN (SELECT 1 AS x) t382 ON t1.x = t382.x JOIN (SELECT 1 AS x) t383 ON t1.x = t383.x JOIN (SELECT 1 AS x) t384 ON t1.x = t384.x JOIN (SELECT 1 AS x) t385 ON t1.x = t385.x JOIN (SELECT 1 AS x) t386 ON t1.x = t386.x JOIN (SELECT 1 AS x) t387 ON t1.x = t387.x JOIN (SELECT 1 AS x) t388 ON t1.x = t388.x JOIN (SELECT 1 AS x) t389 ON t1.x = t389.x JOIN (SELECT 1 AS x) t390 ON t1.x = t390.x JOIN (SELECT 1 AS x) t391 ON t1.x = t391.x JOIN (SELECT 1 AS x) t392 ON t1.x = t392.x JOIN (SELECT 1 AS x) t393 ON t1.x = t393.x JOIN (SELECT 1 AS x) t394 ON t1.x = t394.x JOIN (SELECT 1 AS x) t395 ON t1.x = t395.x JOIN (SELECT 1 AS x) t396 ON t1.x = t396.x JOIN (SELECT 1 AS x) t397 ON t1.x = t397.x JOIN (SELECT 1 AS x) t398 ON t1.x = t398.x JOIN (SELECT 1 AS x) t399 ON t1.x = t399.x JOIN (SELECT 1 AS x) t400 ON t1.x = t400.x JOIN (SELECT 1 AS x) t401 ON t1.x = t401.x JOIN (SELECT 1 AS x) t402 ON t1.x = t402.x JOIN (SELECT 1 AS x) t403 ON t1.x = t403.x JOIN (SELECT 1 AS x) t404 ON t1.x = t404.x JOIN (SELECT 1 AS x) t405 ON t1.x = t405.x JOIN (SELECT 1 AS x) t406 ON t1.x = t406.x JOIN (SELECT 1 AS x) t407 ON t1.x = t407.x JOIN (SELECT 1 AS x) t408 ON t1.x = t408.x JOIN (SELECT 1 AS x) t409 ON t1.x = t409.x JOIN (SELECT 1 AS x) t410 ON t1.x = t410.x JOIN (SELECT 1 AS x) t411 ON t1.x = t411.x JOIN (SELECT 1 AS x) t412 ON t1.x = t412.x JOIN (SELECT 1 AS x) t413 ON t1.x = t413.x JOIN (SELECT 1 AS x) t414 ON t1.x = t414.x JOIN (SELECT 1 AS x) t415 ON t1.x = t415.x JOIN (SELECT 1 AS x) t416 ON t1.x = t416.x JOIN (SELECT 1 AS x) t417 ON t1.x = t417.x JOIN (SELECT 1 AS x) t418 ON t1.x = t418.x JOIN (SELECT 1 AS x) t419 ON t1.x = t419.x JOIN (SELECT 1 AS x) t420 ON t1.x = t420.x JOIN (SELECT 1 AS x) t421 ON t1.x = t421.x JOIN (SELECT 1 AS x) t422 ON t1.x = t422.x JOIN (SELECT 1 AS x) t423 ON t1.x = t423.x JOIN (SELECT 1 AS x) t424 ON t1.x = t424.x JOIN (SELECT 1 AS x) t425 ON t1.x = t425.x JOIN (SELECT 1 AS x) t426 ON t1.x = t426.x JOIN (SELECT 1 AS x) t427 ON t1.x = t427.x JOIN (SELECT 1 AS x) t428 ON t1.x = t428.x JOIN (SELECT 1 AS x) t429 ON t1.x = t429.x JOIN (SELECT 1 AS x) t430 ON t1.x = t430.x JOIN (SELECT 1 AS x) t431 ON t1.x = t431.x JOIN (SELECT 1 AS x) t432 ON t1.x = t432.x JOIN (SELECT 1 AS x) t433 ON t1.x = t433.x JOIN (SELECT 1 AS x) t434 ON t1.x = t434.x JOIN (SELECT 1 AS x) t435 ON t1.x = t435.x JOIN (SELECT 1 AS x) t436 ON t1.x = t436.x JOIN (SELECT 1 AS x) t437 ON t1.x = t437.x JOIN (SELECT 1 AS x) t438 ON t1.x = t438.x JOIN (SELECT 1 AS x) t439 ON t1.x = t439.x JOIN (SELECT 1 AS x) t440 ON t1.x = t440.x JOIN (SELECT 1 AS x) t441 ON t1.x = t441.x JOIN (SELECT 1 AS x) t442 ON t1.x = t442.x JOIN (SELECT 1 AS x) t443 ON t1.x = t443.x JOIN (SELECT 1 AS x) t444 ON t1.x = t444.x JOIN (SELECT 1 AS x) t445 ON t1.x = t445.x JOIN (SELECT 1 AS x) t446 ON t1.x = t446.x JOIN (SELECT 1 AS x) t447 ON t1.x = t447.x JOIN (SELECT 1 AS x) t448 ON t1.x = t448.x JOIN (SELECT 1 AS x) t449 ON t1.x = t449.x JOIN (SELECT 1 AS x) t450 ON t1.x = t450.x JOIN (SELECT 1 AS x) t451 ON t1.x = t451.x JOIN (SELECT 1 AS x) t452 ON t1.x = t452.x JOIN (SELECT 1 AS x) t453 ON t1.x = t453.x JOIN (SELECT 1 AS x) t454 ON t1.x = t454.x JOIN (SELECT 1 AS x) t455 ON t1.x = t455.x JOIN (SELECT 1 AS x) t456 ON t1.x = t456.x JOIN (SELECT 1 AS x) t457 ON t1.x = t457.x JOIN (SELECT 1 AS x) t458 ON t1.x = t458.x JOIN (SELECT 1 AS x) t459 ON t1.x = t459.x JOIN (SELECT 1 AS x) t460 ON t1.x = t460.x JOIN (SELECT 1 AS x) t461 ON t1.x = t461.x JOIN (SELECT 1 AS x) t462 ON t1.x = t462.x JOIN (SELECT 1 AS x) t463 ON t1.x = t463.x JOIN (SELECT 1 AS x) t464 ON t1.x = t464.x JOIN (SELECT 1 AS x) t465 ON t1.x = t465.x JOIN (SELECT 1 AS x) t466 ON t1.x = t466.x JOIN (SELECT 1 AS x) t467 ON t1.x = t467.x JOIN (SELECT 1 AS x) t468 ON t1.x = t468.x JOIN (SELECT 1 AS x) t469 ON t1.x = t469.x JOIN (SELECT 1 AS x) t470 ON t1.x = t470.x JOIN (SELECT 1 AS x) t471 ON t1.x = t471.x JOIN (SELECT 1 AS x) t472 ON t1.x = t472.x JOIN (SELECT 1 AS x) t473 ON t1.x = t473.x JOIN (SELECT 1 AS x) t474 ON t1.x = t474.x JOIN (SELECT 1 AS x) t475 ON t1.x = t475.x JOIN (SELECT 1 AS x) t476 ON t1.x = t476.x JOIN (SELECT 1 AS x) t477 ON t1.x = t477.x JOIN (SELECT 1 AS x) t478 ON t1.x = t478.x JOIN (SELECT 1 AS x) t479 ON t1.x = t479.x JOIN (SELECT 1 AS x) t480 ON t1.x = t480.x JOIN (SELECT 1 AS x) t481 ON t1.x = t481.x JOIN (SELECT 1 AS x) t482 ON t1.x = t482.x JOIN (SELECT 1 AS x) t483 ON t1.x = t483.x JOIN (SELECT 1 AS x) t484 ON t1.x = t484.x JOIN (SELECT 1 AS x) t485 ON t1.x = t485.x JOIN (SELECT 1 AS x) t486 ON t1.x = t486.x JOIN (SELECT 1 AS x) t487 ON t1.x = t487.x JOIN (SELECT 1 AS x) t488 ON t1.x = t488.x JOIN (SELECT 1 AS x) t489 ON t1.x = t489.x JOIN (SELECT 1 AS x) t490 ON t1.x = t490.x JOIN (SELECT 1 AS x) t491 ON t1.x = t491.x JOIN (SELECT 1 AS x) t492 ON t1.x = t492.x JOIN (SELECT 1 AS x) t493 ON t1.x = t493.x JOIN (SELECT 1 AS x) t494 ON t1.x = t494.x JOIN (SELECT 1 AS x) t495 ON t1.x = t495.x JOIN (SELECT 1 AS x) t496 ON t1.x = t496.x JOIN (SELECT 1 AS x) t497 ON t1.x = t497.x JOIN (SELECT 1 AS x) t498 ON t1.x = t498.x JOIN (SELECT 1 AS x) t499 ON t1.x = t499.x JOIN (SELECT 1 AS x) t500 ON t1.x = t500.x JOIN (SELECT 1 AS x) t501 ON t1.x = t501.x JOIN (SELECT 1 AS x) t502 ON t1.x = t502.x JOIN (SELECT 1 AS x) t503 ON t1.x = t503.x JOIN (SELECT 1 AS x) t504 ON t1.x = t504.x JOIN (SELECT 1 AS x) t505 ON t1.x = t505.x JOIN (SELECT 1 AS x) t506 ON t1.x = t506.x JOIN (SELECT 1 AS x) t507 ON t1.x = t507.x JOIN (SELECT 1 AS x) t508 ON t1.x = t508.x JOIN (SELECT 1 AS x) t509 ON t1.x = t509.x JOIN (SELECT 1 AS x) t510 ON t1.x = t510.x JOIN (SELECT 1 AS x) t511 ON t1.x = t511.x JOIN (SELECT 1 AS x) t512 ON t1.x = t512.x JOIN (SELECT 1 AS x) t513 ON t1.x = t513.x JOIN (SELECT 1 AS x) t514 ON t1.x = t514.x JOIN (SELECT 1 AS x) t515 ON t1.x = t515.x JOIN (SELECT 1 AS x) t516 ON t1.x = t516.x JOIN (SELECT 1 AS x) t517 ON t1.x = t517.x JOIN (SELECT 1 AS x) t518 ON t1.x = t518.x JOIN (SELECT 1 AS x) t519 ON t1.x = t519.x JOIN (SELECT 1 AS x) t520 ON t1.x = t520.x JOIN (SELECT 1 AS x) t521 ON t1.x = t521.x JOIN (SELECT 1 AS x) t522 ON t1.x = t522.x JOIN (SELECT 1 AS x) t523 ON t1.x = t523.x JOIN (SELECT 1 AS x) t524 ON t1.x = t524.x JOIN (SELECT 1 AS x) t525 ON t1.x = t525.x JOIN (SELECT 1 AS x) t526 ON t1.x = t526.x JOIN (SELECT 1 AS x) t527 ON t1.x = t527.x JOIN (SELECT 1 AS x) t528 ON t1.x = t528.x JOIN (SELECT 1 AS x) t529 ON t1.x = t529.x JOIN (SELECT 1 AS x) t530 ON t1.x = t530.x JOIN (SELECT 1 AS x) t531 ON t1.x = t531.x JOIN (SELECT 1 AS x) t532 ON t1.x = t532.x JOIN (SELECT 1 AS x) t533 ON t1.x = t533.x JOIN (SELECT 1 AS x) t534 ON t1.x = t534.x JOIN (SELECT 1 AS x) t535 ON t1.x = t535.x JOIN (SELECT 1 AS x) t536 ON t1.x = t536.x JOIN (SELECT 1 AS x) t537 ON t1.x = t537.x JOIN (SELECT 1 AS x) t538 ON t1.x = t538.x JOIN (SELECT 1 AS x) t539 ON t1.x = t539.x JOIN (SELECT 1 AS x) t540 ON t1.x = t540.x JOIN (SELECT 1 AS x) t541 ON t1.x = t541.x JOIN (SELECT 1 AS x) t542 ON t1.x = t542.x JOIN (SELECT 1 AS x) t543 ON t1.x = t543.x JOIN (SELECT 1 AS x) t544 ON t1.x = t544.x JOIN (SELECT 1 AS x) t545 ON t1.x = t545.x JOIN (SELECT 1 AS x) t546 ON t1.x = t546.x JOIN (SELECT 1 AS x) t547 ON t1.x = t547.x JOIN (SELECT 1 AS x) t548 ON t1.x = t548.x JOIN (SELECT 1 AS x) t549 ON t1.x = t549.x JOIN (SELECT 1 AS x) t550 ON t1.x = t550.x JOIN (SELECT 1 AS x) t551 ON t1.x = t551.x JOIN (SELECT 1 AS x) t552 ON t1.x = t552.x JOIN (SELECT 1 AS x) t553 ON t1.x = t553.x JOIN (SELECT 1 AS x) t554 ON t1.x = t554.x JOIN (SELECT 1 AS x) t555 ON t1.x = t555.x JOIN (SELECT 1 AS x) t556 ON t1.x = t556.x JOIN (SELECT 1 AS x) t557 ON t1.x = t557.x JOIN (SELECT 1 AS x) t558 ON t1.x = t558.x JOIN (SELECT 1 AS x) t559 ON t1.x = t559.x JOIN (SELECT 1 AS x) t560 ON t1.x = t560.x JOIN (SELECT 1 AS x) t561 ON t1.x = t561.x JOIN (SELECT 1 AS x) t562 ON t1.x = t562.x JOIN (SELECT 1 AS x) t563 ON t1.x = t563.x JOIN (SELECT 1 AS x) t564 ON t1.x = t564.x JOIN (SELECT 1 AS x) t565 ON t1.x = t565.x JOIN (SELECT 1 AS x) t566 ON t1.x = t566.x JOIN (SELECT 1 AS x) t567 ON t1.x = t567.x JOIN (SELECT 1 AS x) t568 ON t1.x = t568.x JOIN (SELECT 1 AS x) t569 ON t1.x = t569.x JOIN (SELECT 1 AS x) t570 ON t1.x = t570.x JOIN (SELECT 1 AS x) t571 ON t1.x = t571.x JOIN (SELECT 1 AS x) t572 ON t1.x = t572.x JOIN (SELECT 1 AS x) t573 ON t1.x = t573.x JOIN (SELECT 1 AS x) t574 ON t1.x = t574.x JOIN (SELECT 1 AS x) t575 ON t1.x = t575.x JOIN (SELECT 1 AS x) t576 ON t1.x = t576.x JOIN (SELECT 1 AS x) t577 ON t1.x = t577.x JOIN (SELECT 1 AS x) t578 ON t1.x = t578.x JOIN (SELECT 1 AS x) t579 ON t1.x = t579.x JOIN (SELECT 1 AS x) t580 ON t1.x = t580.x JOIN (SELECT 1 AS x) t581 ON t1.x = t581.x JOIN (SELECT 1 AS x) t582 ON t1.x = t582.x JOIN (SELECT 1 AS x) t583 ON t1.x = t583.x JOIN (SELECT 1 AS x) t584 ON t1.x = t584.x JOIN (SELECT 1 AS x) t585 ON t1.x = t585.x JOIN (SELECT 1 AS x) t586 ON t1.x = t586.x JOIN (SELECT 1 AS x) t587 ON t1.x = t587.x JOIN (SELECT 1 AS x) t588 ON t1.x = t588.x JOIN (SELECT 1 AS x) t589 ON t1.x = t589.x JOIN (SELECT 1 AS x) t590 ON t1.x = t590.x JOIN (SELECT 1 AS x) t591 ON t1.x = t591.x JOIN (SELECT 1 AS x) t592 ON t1.x = t592.x JOIN (SELECT 1 AS x) t593 ON t1.x = t593.x JOIN (SELECT 1 AS x) t594 ON t1.x = t594.x JOIN (SELECT 1 AS x) t595 ON t1.x = t595.x JOIN (SELECT 1 AS x) t596 ON t1.x = t596.x JOIN (SELECT 1 AS x) t597 ON t1.x = t597.x JOIN (SELECT 1 AS x) t598 ON t1.x = t598.x JOIN (SELECT 1 AS x) t599 ON t1.x = t599.x JOIN (SELECT 1 AS x) t600 ON t1.x = t600.x JOIN (SELECT 1 AS x) t601 ON t1.x = t601.x JOIN (SELECT 1 AS x) t602 ON t1.x = t602.x JOIN (SELECT 1 AS x) t603 ON t1.x = t603.x JOIN (SELECT 1 AS x) t604 ON t1.x = t604.x JOIN (SELECT 1 AS x) t605 ON t1.x = t605.x JOIN (SELECT 1 AS x) t606 ON t1.x = t606.x JOIN (SELECT 1 AS x) t607 ON t1.x = t607.x JOIN (SELECT 1 AS x) t608 ON t1.x = t608.x JOIN (SELECT 1 AS x) t609 ON t1.x = t609.x JOIN (SELECT 1 AS x) t610 ON t1.x = t610.x JOIN (SELECT 1 AS x) t611 ON t1.x = t611.x JOIN (SELECT 1 AS x) t612 ON t1.x = t612.x JOIN (SELECT 1 AS x) t613 ON t1.x = t613.x JOIN (SELECT 1 AS x) t614 ON t1.x = t614.x JOIN (SELECT 1 AS x) t615 ON t1.x = t615.x JOIN (SELECT 1 AS x) t616 ON t1.x = t616.x JOIN (SELECT 1 AS x) t617 ON t1.x = t617.x JOIN (SELECT 1 AS x) t618 ON t1.x = t618.x JOIN (SELECT 1 AS x) t619 ON t1.x = t619.x JOIN (SELECT 1 AS x) t620 ON t1.x = t620.x JOIN (SELECT 1 AS x) t621 ON t1.x = t621.x JOIN (SELECT 1 AS x) t622 ON t1.x = t622.x JOIN (SELECT 1 AS x) t623 ON t1.x = t623.x JOIN (SELECT 1 AS x) t624 ON t1.x = t624.x JOIN (SELECT 1 AS x) t625 ON t1.x = t625.x JOIN (SELECT 1 AS x) t626 ON t1.x = t626.x JOIN (SELECT 1 AS x) t627 ON t1.x = t627.x JOIN (SELECT 1 AS x) t628 ON t1.x = t628.x JOIN (SELECT 1 AS x) t629 ON t1.x = t629.x JOIN (SELECT 1 AS x) t630 ON t1.x = t630.x JOIN (SELECT 1 AS x) t631 ON t1.x = t631.x JOIN (SELECT 1 AS x) t632 ON t1.x = t632.x JOIN (SELECT 1 AS x) t633 ON t1.x = t633.x JOIN (SELECT 1 AS x) t634 ON t1.x = t634.x JOIN (SELECT 1 AS x) t635 ON t1.x = t635.x JOIN (SELECT 1 AS x) t636 ON t1.x = t636.x JOIN (SELECT 1 AS x) t637 ON t1.x = t637.x JOIN (SELECT 1 AS x) t638 ON t1.x = t638.x JOIN (SELECT 1 AS x) t639 ON t1.x = t639.x JOIN (SELECT 1 AS x) t640 ON t1.x = t640.x JOIN (SELECT 1 AS x) t641 ON t1.x = t641.x JOIN (SELECT 1 AS x) t642 ON t1.x = t642.x JOIN (SELECT 1 AS x) t643 ON t1.x = t643.x JOIN (SELECT 1 AS x) t644 ON t1.x = t644.x JOIN (SELECT 1 AS x) t645 ON t1.x = t645.x JOIN (SELECT 1 AS x) t646 ON t1.x = t646.x JOIN (SELECT 1 AS x) t647 ON t1.x = t647.x JOIN (SELECT 1 AS x) t648 ON t1.x = t648.x JOIN (SELECT 1 AS x) t649 ON t1.x = t649.x JOIN (SELECT 1 AS x) t650 ON t1.x = t650.x JOIN (SELECT 1 AS x) t651 ON t1.x = t651.x JOIN (SELECT 1 AS x) t652 ON t1.x = t652.x JOIN (SELECT 1 AS x) t653 ON t1.x = t653.x JOIN (SELECT 1 AS x) t654 ON t1.x = t654.x JOIN (SELECT 1 AS x) t655 ON t1.x = t655.x JOIN (SELECT 1 AS x) t656 ON t1.x = t656.x JOIN (SELECT 1 AS x) t657 ON t1.x = t657.x JOIN (SELECT 1 AS x) t658 ON t1.x = t658.x JOIN (SELECT 1 AS x) t659 ON t1.x = t659.x JOIN (SELECT 1 AS x) t660 ON t1.x = t660.x JOIN (SELECT 1 AS x) t661 ON t1.x = t661.x JOIN (SELECT 1 AS x) t662 ON t1.x = t662.x JOIN (SELECT 1 AS x) t663 ON t1.x = t663.x JOIN (SELECT 1 AS x) t664 ON t1.x = t664.x JOIN (SELECT 1 AS x) t665 ON t1.x = t665.x JOIN (SELECT 1 AS x) t666 ON t1.x = t666.x diff --git a/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.reference b/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.reference index 5fde4f80c5d..d35bdeff98b 100644 --- a/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.reference +++ b/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.reference @@ -5,18 +5,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 0 1 2 + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 0 2 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: INNER Strictness: ALL Algorithm: HashJoin @@ -75,18 +75,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 0 1 2 + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 0 2 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: INNER Strictness: ALL Algorithm: HashJoin @@ -145,18 +145,18 @@ Header: id UInt64 rhs.value String Actions: INPUT : 0 -> __table1.id UInt64 : 0 INPUT : 1 -> __table1.value String : 1 - INPUT : 2 -> __table2.id UInt64 : 2 - INPUT : 3 -> __table2.value String : 3 + INPUT : 2 -> __table2.value String : 2 + INPUT : 3 -> __table2.id UInt64 : 3 ALIAS __table1.id :: 0 -> id UInt64 : 4 ALIAS __table1.value :: 1 -> value String : 0 - ALIAS __table2.id :: 2 -> rhs.id UInt64 : 1 - ALIAS __table2.value :: 3 -> rhs.value String : 2 -Positions: 4 0 1 2 + ALIAS __table2.value :: 2 -> rhs.value String : 1 + ALIAS __table2.id :: 3 -> rhs.id UInt64 : 2 +Positions: 4 0 2 1 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String - __table2.id UInt64 __table2.value String + __table2.id UInt64 Type: INNER Strictness: ALL Algorithm: HashJoin diff --git a/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql b/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql index ddefc322b4f..b3d1827d98f 100644 --- a/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql +++ b/tests/queries/0_stateless/03130_convert_outer_join_to_inner_join.sql @@ -22,10 +22,7 @@ SETTINGS index_granularity = 16 INSERT INTO test_table_1 VALUES (1, 'Value_1'), (2, 'Value_2'); INSERT INTO test_table_2 VALUES (2, 'Value_2'), (3, 'Value_3'); - -EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE rhs.id != 0 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE rhs.id != 0; SELECT '--'; @@ -33,9 +30,7 @@ SELECT * FROM test_table_1 AS lhs LEFT JOIN test_table_2 AS rhs ON lhs.id = rhs. SELECT '--'; -EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE lhs.id != 0 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE lhs.id != 0; SELECT '--'; @@ -43,9 +38,7 @@ SELECT * FROM test_table_1 AS lhs RIGHT JOIN test_table_2 AS rhs ON lhs.id = rhs SELECT '--'; -EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE lhs.id != 0 AND rhs.id != 0 -SETTINGS query_plan_join_inner_table_selection = 'right' -; +EXPLAIN header = 1, actions = 1 SELECT * FROM test_table_1 AS lhs FULL JOIN test_table_2 AS rhs ON lhs.id = rhs.id WHERE lhs.id != 0 AND rhs.id != 0; SELECT '--'; diff --git a/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.reference b/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.reference index 1c82e76cc65..7058d36aaf9 100644 --- a/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.reference +++ b/tests/queries/0_stateless/03152_join_filter_push_down_equivalent_columns.reference @@ -65,7 +65,8 @@ SELECT name FROM users RIGHT JOIN users2 USING name WHERE users2.name ='Alice'; Expression ((Project names + (Projection + ))) Header: name String Join (JOIN FillRightFirst) - Header: __table2.name String + Header: __table1.name String + __table2.name String Filter (( + Change column names to column identifiers)) Header: __table1.name String ReadFromMergeTree (default.users) diff --git a/tests/queries/0_stateless/03236_squashing_high_memory.sql b/tests/queries/0_stateless/03236_squashing_high_memory.sql index eeb3ae85e84..f6e5dbdef03 100644 --- a/tests/queries/0_stateless/03236_squashing_high_memory.sql +++ b/tests/queries/0_stateless/03236_squashing_high_memory.sql @@ -11,7 +11,6 @@ CREATE TABLE id_values ENGINE MergeTree ORDER BY id1 AS SELECT arrayJoin(range(500000)) AS id1, arrayJoin(range(1000)) AS id2; SET max_memory_usage = '1G'; -SET query_plan_join_inner_table_selection = 'right'; CREATE TABLE test_table ENGINE MergeTree ORDER BY id AS SELECT id_values.id1 AS id, From 4e30cf7e333312968bebe57dc0f6dd381cbccff5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Wed, 6 Nov 2024 16:30:16 +0100 Subject: [PATCH 480/680] Cleanup SettingsChangesHistory for revert --- src/Core/SettingsChangesHistory.cpp | 1 - 1 file changed, 1 deletion(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index ed87fde8b7e..64964f294bd 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -73,7 +73,6 @@ static std::initializer_list Date: Wed, 6 Nov 2024 10:50:45 +0100 Subject: [PATCH 481/680] Upgrade clickhouse-server and keeper base images --- docker/keeper/Dockerfile | 10 +++++++--- docker/server/Dockerfile.ubuntu | 2 +- 2 files changed, 8 insertions(+), 4 deletions(-) diff --git a/docker/keeper/Dockerfile b/docker/keeper/Dockerfile index bc76bdbb619..4ecc087afb4 100644 --- a/docker/keeper/Dockerfile +++ b/docker/keeper/Dockerfile @@ -1,7 +1,7 @@ # The Dockerfile.ubuntu exists for the tests/ci/docker_server.py script # If the image is built from Dockerfile.alpine, then the `-alpine` suffix is added automatically, # so the only purpose of Dockerfile.ubuntu is to push `latest`, `head` and so on w/o suffixes -FROM ubuntu:20.04 AS glibc-donor +FROM ubuntu:22.04 AS glibc-donor ARG TARGETARCH RUN arch=${TARGETARCH:-amd64} \ @@ -9,7 +9,11 @@ RUN arch=${TARGETARCH:-amd64} \ amd64) rarch=x86_64 ;; \ arm64) rarch=aarch64 ;; \ esac \ - && ln -s "${rarch}-linux-gnu" /lib/linux-gnu + && ln -s "${rarch}-linux-gnu" /lib/linux-gnu \ + && case $arch in \ + amd64) ln /lib/linux-gnu/ld-linux-x86-64.so.2 /lib/linux-gnu/ld-2.35.so ;; \ + arm64) ln /lib/linux-gnu/ld-linux-aarch64.so.1 /lib/linux-gnu/ld-2.35.so ;; \ + esac FROM alpine @@ -20,7 +24,7 @@ ENV LANG=en_US.UTF-8 \ TZ=UTC \ CLICKHOUSE_CONFIG=/etc/clickhouse-server/config.xml -COPY --from=glibc-donor /lib/linux-gnu/libc.so.6 /lib/linux-gnu/libdl.so.2 /lib/linux-gnu/libm.so.6 /lib/linux-gnu/libpthread.so.0 /lib/linux-gnu/librt.so.1 /lib/linux-gnu/libnss_dns.so.2 /lib/linux-gnu/libnss_files.so.2 /lib/linux-gnu/libresolv.so.2 /lib/linux-gnu/ld-2.31.so /lib/ +COPY --from=glibc-donor /lib/linux-gnu/libc.so.6 /lib/linux-gnu/libdl.so.2 /lib/linux-gnu/libm.so.6 /lib/linux-gnu/libpthread.so.0 /lib/linux-gnu/librt.so.1 /lib/linux-gnu/libnss_dns.so.2 /lib/linux-gnu/libnss_files.so.2 /lib/linux-gnu/libresolv.so.2 /lib/linux-gnu/ld-2.35.so /lib/ COPY --from=glibc-donor /etc/nsswitch.conf /etc/ COPY entrypoint.sh /entrypoint.sh diff --git a/docker/server/Dockerfile.ubuntu b/docker/server/Dockerfile.ubuntu index 506a627b11c..0d5c983f5e6 100644 --- a/docker/server/Dockerfile.ubuntu +++ b/docker/server/Dockerfile.ubuntu @@ -1,4 +1,4 @@ -FROM ubuntu:20.04 +FROM ubuntu:22.04 # see https://github.com/moby/moby/issues/4032#issuecomment-192327844 # It could be removed after we move on a version 23:04+ From 2903227143360795fc4912322de9963ec7f8c3ef Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Wed, 6 Nov 2024 10:58:21 +0100 Subject: [PATCH 482/680] Remove strange wrong named dockerfile --- .../clickhouse-statelest-test-runner.Dockerfile | 16 ---------------- 1 file changed, 16 deletions(-) delete mode 100644 docker/test/stateless/clickhouse-statelest-test-runner.Dockerfile diff --git a/docker/test/stateless/clickhouse-statelest-test-runner.Dockerfile b/docker/test/stateless/clickhouse-statelest-test-runner.Dockerfile deleted file mode 100644 index a9802f6f1da..00000000000 --- a/docker/test/stateless/clickhouse-statelest-test-runner.Dockerfile +++ /dev/null @@ -1,16 +0,0 @@ -# Since right now we can't set volumes to the docker during build, we split building container in stages: -# 1. build base container -# 2. run base conatiner with mounted volumes -# 3. commit container as image -FROM ubuntu:20.04 as clickhouse-test-runner-base - -# A volume where directory with clickhouse packages to be mounted, -# for later installing. -VOLUME /packages - -CMD apt-get update ;\ - DEBIAN_FRONTEND=noninteractive \ - apt install -y /packages/clickhouse-common-static_*.deb \ - /packages/clickhouse-client_*.deb \ - && apt-get clean \ - && rm -rf /var/lib/apt/lists/* /var/cache/debconf /tmp/* From 7b1de3fcf792aeae2cc2b197e841afcda9092654 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Wed, 6 Nov 2024 11:12:26 +0100 Subject: [PATCH 483/680] We use `aarch64` everywhere in code, so the vars should reflect it --- tests/ci/ci_config.py | 54 ++++++++++++++++----------------- tests/ci/ci_definitions.py | 30 +++++++++--------- tests/ci/compatibility_check.py | 2 +- tests/ci/test_ci_config.py | 8 ++--- tests/ci/test_ci_options.py | 4 +-- 5 files changed, 49 insertions(+), 49 deletions(-) diff --git a/tests/ci/ci_config.py b/tests/ci/ci_config.py index 6d23b594b24..67cdbbdcf6d 100644 --- a/tests/ci/ci_config.py +++ b/tests/ci/ci_config.py @@ -51,11 +51,11 @@ class CI: TAG_CONFIGS = { Tags.DO_NOT_TEST_LABEL: LabelConfig(run_jobs=[JobNames.STYLE_CHECK]), - Tags.CI_SET_ARM: LabelConfig( + Tags.CI_SET_AARCH64: LabelConfig( run_jobs=[ JobNames.STYLE_CHECK, BuildNames.PACKAGE_AARCH64, - JobNames.INTEGRATION_TEST_ARM, + JobNames.INTEGRATION_TEST_AARCH64, ] ), Tags.CI_SET_REQUIRED: LabelConfig( @@ -95,16 +95,16 @@ class CI: static_binary_name="aarch64", additional_pkgs=True, ), - runner_type=Runners.BUILDER_ARM, + runner_type=Runners.BUILDER_AARCH64, ), - BuildNames.PACKAGE_ARM_ASAN: CommonJobConfigs.BUILD.with_properties( + BuildNames.PACKAGE_AARCH64_ASAN: CommonJobConfigs.BUILD.with_properties( build_config=BuildConfig( - name=BuildNames.PACKAGE_ARM_ASAN, + name=BuildNames.PACKAGE_AARCH64_ASAN, compiler="clang-18-aarch64", sanitizer="address", package_type="deb", ), - runner_type=Runners.BUILDER_ARM, + runner_type=Runners.BUILDER_AARCH64, ), BuildNames.PACKAGE_ASAN: CommonJobConfigs.BUILD.with_properties( build_config=BuildConfig( @@ -276,16 +276,16 @@ class CI: JobNames.INSTALL_TEST_AMD: CommonJobConfigs.INSTALL_TEST.with_properties( required_builds=[BuildNames.PACKAGE_RELEASE] ), - JobNames.INSTALL_TEST_ARM: CommonJobConfigs.INSTALL_TEST.with_properties( + JobNames.INSTALL_TEST_AARCH64: CommonJobConfigs.INSTALL_TEST.with_properties( required_builds=[BuildNames.PACKAGE_AARCH64], - runner_type=Runners.STYLE_CHECKER_ARM, + runner_type=Runners.STYLE_CHECKER_AARCH64, ), JobNames.STATEFUL_TEST_ASAN: CommonJobConfigs.STATEFUL_TEST.with_properties( required_builds=[BuildNames.PACKAGE_ASAN] ), - JobNames.STATEFUL_TEST_ARM_ASAN: CommonJobConfigs.STATEFUL_TEST.with_properties( - required_builds=[BuildNames.PACKAGE_ARM_ASAN], - runner_type=Runners.FUNC_TESTER_ARM, + JobNames.STATEFUL_TEST_AARCH64_ASAN: CommonJobConfigs.STATEFUL_TEST.with_properties( + required_builds=[BuildNames.PACKAGE_AARCH64_ASAN], + runner_type=Runners.FUNC_TESTER_AARCH64, ), JobNames.STATEFUL_TEST_TSAN: CommonJobConfigs.STATEFUL_TEST.with_properties( required_builds=[BuildNames.PACKAGE_TSAN] @@ -307,7 +307,7 @@ class CI: ), JobNames.STATEFUL_TEST_AARCH64: CommonJobConfigs.STATEFUL_TEST.with_properties( required_builds=[BuildNames.PACKAGE_AARCH64], - runner_type=Runners.FUNC_TESTER_ARM, + runner_type=Runners.FUNC_TESTER_AARCH64, ), JobNames.STATEFUL_TEST_PARALLEL_REPL_RELEASE: CommonJobConfigs.STATEFUL_TEST.with_properties( required_builds=[BuildNames.PACKAGE_RELEASE] @@ -335,10 +335,10 @@ class CI: JobNames.STATELESS_TEST_ASAN: CommonJobConfigs.STATELESS_TEST.with_properties( required_builds=[BuildNames.PACKAGE_ASAN], num_batches=2 ), - JobNames.STATELESS_TEST_ARM_ASAN: CommonJobConfigs.STATELESS_TEST.with_properties( - required_builds=[BuildNames.PACKAGE_ARM_ASAN], + JobNames.STATELESS_TEST_AARCH64_ASAN: CommonJobConfigs.STATELESS_TEST.with_properties( + required_builds=[BuildNames.PACKAGE_AARCH64_ASAN], num_batches=2, - runner_type=Runners.FUNC_TESTER_ARM, + runner_type=Runners.FUNC_TESTER_AARCH64, ), JobNames.STATELESS_TEST_TSAN: CommonJobConfigs.STATELESS_TEST.with_properties( required_builds=[BuildNames.PACKAGE_TSAN], num_batches=4 @@ -360,7 +360,7 @@ class CI: ), JobNames.STATELESS_TEST_AARCH64: CommonJobConfigs.STATELESS_TEST.with_properties( required_builds=[BuildNames.PACKAGE_AARCH64], - runner_type=Runners.FUNC_TESTER_ARM, + runner_type=Runners.FUNC_TESTER_AARCH64, ), JobNames.STATELESS_TEST_OLD_ANALYZER_S3_REPLICATED_RELEASE: CommonJobConfigs.STATELESS_TEST.with_properties( required_builds=[BuildNames.PACKAGE_RELEASE], num_batches=2 @@ -432,10 +432,10 @@ class CI: num_batches=6, timeout=9000, # the job timed out with default value (7200) ), - JobNames.INTEGRATION_TEST_ARM: CommonJobConfigs.INTEGRATION_TEST.with_properties( + JobNames.INTEGRATION_TEST_AARCH64: CommonJobConfigs.INTEGRATION_TEST.with_properties( required_builds=[BuildNames.PACKAGE_AARCH64], num_batches=6, - runner_type=Runners.FUNC_TESTER_ARM, + runner_type=Runners.FUNC_TESTER_AARCH64, ), JobNames.INTEGRATION_TEST: CommonJobConfigs.INTEGRATION_TEST.with_properties( required_builds=[BuildNames.PACKAGE_RELEASE], @@ -453,10 +453,10 @@ class CI: required_builds=[BuildNames.PACKAGE_RELEASE], required_on_release_branch=True, ), - JobNames.COMPATIBILITY_TEST_ARM: CommonJobConfigs.COMPATIBILITY_TEST.with_properties( + JobNames.COMPATIBILITY_TEST_AARCH64: CommonJobConfigs.COMPATIBILITY_TEST.with_properties( required_builds=[BuildNames.PACKAGE_AARCH64], required_on_release_branch=True, - runner_type=Runners.STYLE_CHECKER_ARM, + runner_type=Runners.STYLE_CHECKER_AARCH64, ), JobNames.UNIT_TEST: CommonJobConfigs.UNIT_TEST.with_properties( required_builds=[BuildNames.BINARY_RELEASE], @@ -499,22 +499,22 @@ class CI: required_builds=[BuildNames.BINARY_RELEASE], run_by_labels=[Labels.JEPSEN_TEST], run_command="jepsen_check.py keeper", - runner_type=Runners.STYLE_CHECKER_ARM, + runner_type=Runners.STYLE_CHECKER_AARCH64, ), JobNames.JEPSEN_SERVER: JobConfig( required_builds=[BuildNames.BINARY_RELEASE], run_by_labels=[Labels.JEPSEN_TEST], run_command="jepsen_check.py server", - runner_type=Runners.STYLE_CHECKER_ARM, + runner_type=Runners.STYLE_CHECKER_AARCH64, ), JobNames.PERFORMANCE_TEST_AMD64: CommonJobConfigs.PERF_TESTS.with_properties( required_builds=[BuildNames.PACKAGE_RELEASE], num_batches=4 ), - JobNames.PERFORMANCE_TEST_ARM64: CommonJobConfigs.PERF_TESTS.with_properties( + JobNames.PERFORMANCE_TEST_AARCH64: CommonJobConfigs.PERF_TESTS.with_properties( required_builds=[BuildNames.PACKAGE_AARCH64], num_batches=4, run_by_labels=[Labels.PR_PERFORMANCE], - runner_type=Runners.FUNC_TESTER_ARM, + runner_type=Runners.FUNC_TESTER_AARCH64, ), JobNames.SQLANCER: CommonJobConfigs.SQLLANCER_TEST.with_properties( required_builds=[BuildNames.PACKAGE_RELEASE], @@ -532,9 +532,9 @@ class CI: JobNames.CLICKBENCH_TEST: CommonJobConfigs.CLICKBENCH_TEST.with_properties( required_builds=[BuildNames.PACKAGE_RELEASE], ), - JobNames.CLICKBENCH_TEST_ARM: CommonJobConfigs.CLICKBENCH_TEST.with_properties( + JobNames.CLICKBENCH_TEST_AARCH64: CommonJobConfigs.CLICKBENCH_TEST.with_properties( required_builds=[BuildNames.PACKAGE_AARCH64], - runner_type=Runners.FUNC_TESTER_ARM, + runner_type=Runners.FUNC_TESTER_AARCH64, ), JobNames.LIBFUZZER_TEST: JobConfig( required_builds=[BuildNames.FUZZERS], @@ -572,7 +572,7 @@ class CI: ), JobNames.STYLE_CHECK: JobConfig( run_always=True, - runner_type=Runners.STYLE_CHECKER_ARM, + runner_type=Runners.STYLE_CHECKER_AARCH64, ), JobNames.BUGFIX_VALIDATE: JobConfig( run_by_labels=[Labels.PR_BUGFIX, Labels.PR_CRITICAL_BUGFIX], diff --git a/tests/ci/ci_definitions.py b/tests/ci/ci_definitions.py index dd86dc320c2..fb3e55fdbe3 100644 --- a/tests/ci/ci_definitions.py +++ b/tests/ci/ci_definitions.py @@ -58,11 +58,11 @@ class Runners(metaclass=WithIter): """ BUILDER = "builder" - BUILDER_ARM = "builder-aarch64" + BUILDER_AARCH64 = "builder-aarch64" STYLE_CHECKER = "style-checker" - STYLE_CHECKER_ARM = "style-checker-aarch64" + STYLE_CHECKER_AARCH64 = "style-checker-aarch64" FUNC_TESTER = "func-tester" - FUNC_TESTER_ARM = "func-tester-aarch64" + FUNC_TESTER_AARCH64 = "func-tester-aarch64" FUZZER_UNIT_TESTER = "fuzzer-unit-tester" @@ -78,7 +78,7 @@ class Tags(metaclass=WithIter): # to upload all binaries from build jobs UPLOAD_ALL_ARTIFACTS = "upload_all" CI_SET_SYNC = "ci_set_sync" - CI_SET_ARM = "ci_set_arm" + CI_SET_AARCH64 = "ci_set_aarch64" CI_SET_REQUIRED = "ci_set_required" CI_SET_BUILDS = "ci_set_builds" @@ -106,7 +106,7 @@ class BuildNames(metaclass=WithIter): PACKAGE_MSAN = "package_msan" PACKAGE_DEBUG = "package_debug" PACKAGE_AARCH64 = "package_aarch64" - PACKAGE_ARM_ASAN = "package_aarch64_asan" + PACKAGE_AARCH64_ASAN = "package_aarch64_asan" PACKAGE_RELEASE_COVERAGE = "package_release_coverage" BINARY_RELEASE = "binary_release" BINARY_TIDY = "binary_tidy" @@ -134,14 +134,14 @@ class JobNames(metaclass=WithIter): DOCKER_SERVER = "Docker server image" DOCKER_KEEPER = "Docker keeper image" INSTALL_TEST_AMD = "Install packages (release)" - INSTALL_TEST_ARM = "Install packages (aarch64)" + INSTALL_TEST_AARCH64 = "Install packages (aarch64)" STATELESS_TEST_DEBUG = "Stateless tests (debug)" STATELESS_TEST_RELEASE = "Stateless tests (release)" STATELESS_TEST_RELEASE_COVERAGE = "Stateless tests (coverage)" STATELESS_TEST_AARCH64 = "Stateless tests (aarch64)" STATELESS_TEST_ASAN = "Stateless tests (asan)" - STATELESS_TEST_ARM_ASAN = "Stateless tests (aarch64, asan)" + STATELESS_TEST_AARCH64_ASAN = "Stateless tests (aarch64, asan)" STATELESS_TEST_TSAN = "Stateless tests (tsan)" STATELESS_TEST_MSAN = "Stateless tests (msan)" STATELESS_TEST_UBSAN = "Stateless tests (ubsan)" @@ -158,7 +158,7 @@ class JobNames(metaclass=WithIter): STATEFUL_TEST_RELEASE_COVERAGE = "Stateful tests (coverage)" STATEFUL_TEST_AARCH64 = "Stateful tests (aarch64)" STATEFUL_TEST_ASAN = "Stateful tests (asan)" - STATEFUL_TEST_ARM_ASAN = "Stateful tests (aarch64, asan)" + STATEFUL_TEST_AARCH64_ASAN = "Stateful tests (aarch64, asan)" STATEFUL_TEST_TSAN = "Stateful tests (tsan)" STATEFUL_TEST_MSAN = "Stateful tests (msan)" STATEFUL_TEST_UBSAN = "Stateful tests (ubsan)" @@ -181,7 +181,7 @@ class JobNames(metaclass=WithIter): INTEGRATION_TEST_ASAN = "Integration tests (asan)" INTEGRATION_TEST_ASAN_OLD_ANALYZER = "Integration tests (asan, old analyzer)" INTEGRATION_TEST_TSAN = "Integration tests (tsan)" - INTEGRATION_TEST_ARM = "Integration tests (aarch64)" + INTEGRATION_TEST_AARCH64 = "Integration tests (aarch64)" INTEGRATION_TEST_FLAKY = "Integration tests flaky check (asan)" UPGRADE_TEST_DEBUG = "Upgrade check (debug)" @@ -205,7 +205,7 @@ class JobNames(metaclass=WithIter): JEPSEN_SERVER = "ClickHouse Server Jepsen" PERFORMANCE_TEST_AMD64 = "Performance Comparison (release)" - PERFORMANCE_TEST_ARM64 = "Performance Comparison (aarch64)" + PERFORMANCE_TEST_AARCH64 = "Performance Comparison (aarch64)" # SQL_LOGIC_TEST = "Sqllogic test (release)" @@ -214,10 +214,10 @@ class JobNames(metaclass=WithIter): SQLTEST = "SQLTest" COMPATIBILITY_TEST = "Compatibility check (release)" - COMPATIBILITY_TEST_ARM = "Compatibility check (aarch64)" + COMPATIBILITY_TEST_AARCH64 = "Compatibility check (aarch64)" CLICKBENCH_TEST = "ClickBench (release)" - CLICKBENCH_TEST_ARM = "ClickBench (aarch64)" + CLICKBENCH_TEST_AARCH64 = "ClickBench (aarch64)" LIBFUZZER_TEST = "libFuzzer tests" @@ -387,7 +387,7 @@ class CommonJobConfigs: "./tests/ci/upload_result_helper.py", ], ), - runner_type=Runners.STYLE_CHECKER_ARM, + runner_type=Runners.STYLE_CHECKER_AARCH64, disable_await=True, ) COMPATIBILITY_TEST = JobConfig( @@ -634,8 +634,8 @@ REQUIRED_CHECKS = [ JobNames.STATEFUL_TEST_RELEASE, JobNames.STATELESS_TEST_RELEASE, JobNames.STATELESS_TEST_ASAN, - JobNames.STATELESS_TEST_ARM_ASAN, - JobNames.STATEFUL_TEST_ARM_ASAN, + JobNames.STATELESS_TEST_AARCH64_ASAN, + JobNames.STATEFUL_TEST_AARCH64_ASAN, JobNames.STATELESS_TEST_FLAKY_ASAN, JobNames.STATEFUL_TEST_ASAN, JobNames.STYLE_CHECK, diff --git a/tests/ci/compatibility_check.py b/tests/ci/compatibility_check.py index bb0c717160e..38fb2eceb28 100644 --- a/tests/ci/compatibility_check.py +++ b/tests/ci/compatibility_check.py @@ -131,7 +131,7 @@ def main(): check_name = args.check_name or os.getenv("CHECK_NAME") assert check_name check_glibc = True - # currently hardcoded to x86, don't enable for ARM + # currently hardcoded to x86, don't enable for AARCH64 check_distributions = ( "aarch64" not in check_name.lower() and "arm64" not in check_name.lower() ) diff --git a/tests/ci/test_ci_config.py b/tests/ci/test_ci_config.py index 0e396b827ea..03f28983262 100644 --- a/tests/ci/test_ci_config.py +++ b/tests/ci/test_ci_config.py @@ -36,11 +36,11 @@ class TestCIConfig(unittest.TestCase): elif "binary_" in job.lower() or "package_" in job.lower(): if job.lower() in ( CI.BuildNames.PACKAGE_AARCH64, - CI.BuildNames.PACKAGE_ARM_ASAN, + CI.BuildNames.PACKAGE_AARCH64_ASAN, ): self.assertTrue( - CI.JOB_CONFIGS[job].runner_type in (CI.Runners.BUILDER_ARM,), - f"Job [{job}] must have [{CI.Runners.BUILDER_ARM}] runner", + CI.JOB_CONFIGS[job].runner_type in (CI.Runners.BUILDER_AARCH64,), + f"Job [{job}] must have [{CI.Runners.BUILDER_AARCH64}] runner", ) else: self.assertTrue( @@ -96,7 +96,7 @@ class TestCIConfig(unittest.TestCase): else: self.assertTrue(CI.JOB_CONFIGS[job].build_config is None) if "asan" in job and "aarch" in job: - expected_builds = [CI.BuildNames.PACKAGE_ARM_ASAN] + expected_builds = [CI.BuildNames.PACKAGE_AARCH64_ASAN] elif "asan" in job: expected_builds = [CI.BuildNames.PACKAGE_ASAN] elif "msan" in job: diff --git a/tests/ci/test_ci_options.py b/tests/ci/test_ci_options.py index 536e18758f8..e1b780387e7 100644 --- a/tests/ci/test_ci_options.py +++ b/tests/ci/test_ci_options.py @@ -10,7 +10,7 @@ from ci_settings import CiSettings _TEST_BODY_1 = """ #### Run only: - [ ] Some Set -- [x] Integration tests (arm64) +- [x] Integration tests (aarch64) - [x] Integration tests - [x] Integration tests - [ ] Integration tests @@ -150,7 +150,7 @@ class TestCIOptions(unittest.TestCase): self.assertFalse(ci_options.no_ci_cache) self.assertTrue(ci_options.no_merge_commit) self.assertTrue(ci_options.woolen_wolfdog) - self.assertEqual(ci_options.ci_sets, ["ci_set_arm"]) + self.assertEqual(ci_options.ci_sets, ["ci_set_aarch64"]) self.assertCountEqual(ci_options.include_keywords, ["foo", "foo_bar"]) self.assertCountEqual(ci_options.exclude_keywords, ["foo", "foo_bar"]) From df632b6f1e4d825644138c77c2ae4a25943a7fe8 Mon Sep 17 00:00:00 2001 From: Sema Checherinda Date: Wed, 6 Nov 2024 16:44:52 +0100 Subject: [PATCH 484/680] clean up --- .../AzureBlobStorage/AzureObjectStorage.cpp | 13 ------------- .../AzureBlobStorage/AzureObjectStorage.h | 5 ----- .../ObjectStorages/Cached/CachedObjectStorage.cpp | 14 -------------- .../ObjectStorages/Cached/CachedObjectStorage.h | 4 ---- src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp | 10 ---------- src/Disks/ObjectStorages/S3/S3ObjectStorage.h | 7 ------- src/Disks/ObjectStorages/Web/WebObjectStorage.cpp | 10 ---------- src/Disks/ObjectStorages/Web/WebObjectStorage.h | 4 ---- 8 files changed, 67 deletions(-) diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp index 959afa65672..b8386bcf967 100644 --- a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp +++ b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.cpp @@ -277,19 +277,6 @@ void AzureObjectStorage::removeObjectImpl(const StoredObject & object, const Sha } } -/// Remove file. Throws exception if file doesn't exists or it's a directory. -// void AzureObjectStorage::removeObject(const StoredObject & object) -// { -// removeObjectImpl(object, client.get(), false); -// } - -// void AzureObjectStorage::removeObjects(const StoredObjects & objects) -// { -// auto client_ptr = client.get(); -// for (const auto & object : objects) -// removeObjectImpl(object, client_ptr, false); -// } - void AzureObjectStorage::removeObjectIfExists(const StoredObject & object) { removeObjectImpl(object, client.get(), true); diff --git a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h index 433fe7a852e..401493be367 100644 --- a/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h +++ b/src/Disks/ObjectStorages/AzureBlobStorage/AzureObjectStorage.h @@ -59,11 +59,6 @@ public: size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE, const WriteSettings & write_settings = {}) override; - /// Remove file. Throws exception if file doesn't exists or it's a directory. - //void removeObject(const StoredObject & object) override; - - //void removeObjects(const StoredObjects & objects) override; - void removeObjectIfExists(const StoredObject & object) override; void removeObjectsIfExist(const StoredObjects & objects) override; diff --git a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp index f2750e6814f..779b8830fab 100644 --- a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp +++ b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.cpp @@ -148,20 +148,6 @@ void CachedObjectStorage::removeCacheIfExists(const std::string & path_key_for_c cache->removeKeyIfExists(getCacheKey(path_key_for_cache), FileCache::getCommonUser().user_id); } -// void CachedObjectStorage::removeObject(const StoredObject & object) -// { -// removeCacheIfExists(object.remote_path); -// object_storage->removeObject(object); -// } - -// void CachedObjectStorage::removeObjects(const StoredObjects & objects) -// { -// for (const auto & object : objects) -// removeCacheIfExists(object.remote_path); - -// object_storage->removeObjects(objects); -// } - void CachedObjectStorage::removeObjectIfExists(const StoredObject & object) { removeCacheIfExists(object.remote_path); diff --git a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h index 7e10057e04c..77aa635b89b 100644 --- a/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h +++ b/src/Disks/ObjectStorages/Cached/CachedObjectStorage.h @@ -45,10 +45,6 @@ public: size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE, const WriteSettings & write_settings = {}) override; - // void removeObject(const StoredObject & object) override; - - // void removeObjects(const StoredObjects & objects) override; - void removeObjectIfExists(const StoredObject & object) override; void removeObjectsIfExist(const StoredObjects & objects) override; diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp index 7ed118c6b07..9fca3cad688 100644 --- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp +++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.cpp @@ -326,21 +326,11 @@ void S3ObjectStorage::removeObjectsImpl(const StoredObjects & objects, bool if_e ProfileEvents::DiskS3DeleteObjects); } -// void S3ObjectStorage::removeObject(const StoredObject & object) -// { -// removeObjectImpl(object, false); -// } - void S3ObjectStorage::removeObjectIfExists(const StoredObject & object) { removeObjectImpl(object, true); } -// void S3ObjectStorage::removeObjects(const StoredObjects & objects) -// { -// removeObjectsImpl(objects, false); -// } - void S3ObjectStorage::removeObjectsIfExist(const StoredObjects & objects) { removeObjectsImpl(objects, true); diff --git a/src/Disks/ObjectStorages/S3/S3ObjectStorage.h b/src/Disks/ObjectStorages/S3/S3ObjectStorage.h index a2aeaf8a43c..4b9c968ede9 100644 --- a/src/Disks/ObjectStorages/S3/S3ObjectStorage.h +++ b/src/Disks/ObjectStorages/S3/S3ObjectStorage.h @@ -101,13 +101,6 @@ public: ObjectStorageIteratorPtr iterate(const std::string & path_prefix, size_t max_keys) const override; - /// Uses `DeleteObjectRequest`. - //void removeObject(const StoredObject & object) override; - - /// Uses `DeleteObjectsRequest` if it is allowed by `s3_capabilities`, otherwise `DeleteObjectRequest`. - /// `DeleteObjectsRequest` is not supported on GCS, see https://issuetracker.google.com/issues/162653700 . - //void removeObjects(const StoredObjects & objects) override; - /// Uses `DeleteObjectRequest`. void removeObjectIfExists(const StoredObject & object) override; diff --git a/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp b/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp index 1503d5819eb..35abc0ed0df 100644 --- a/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp +++ b/src/Disks/ObjectStorages/Web/WebObjectStorage.cpp @@ -254,16 +254,6 @@ std::unique_ptr WebObjectStorage::writeObject( /// NOLI throwNotAllowed(); } -// void WebObjectStorage::removeObject(const StoredObject &) -// { -// throwNotAllowed(); -// } - -// void WebObjectStorage::removeObjects(const StoredObjects &) -// { -// throwNotAllowed(); -// } - void WebObjectStorage::removeObjectIfExists(const StoredObject &) { throwNotAllowed(); diff --git a/src/Disks/ObjectStorages/Web/WebObjectStorage.h b/src/Disks/ObjectStorages/Web/WebObjectStorage.h index ae52cc20f9b..1e612bd359c 100644 --- a/src/Disks/ObjectStorages/Web/WebObjectStorage.h +++ b/src/Disks/ObjectStorages/Web/WebObjectStorage.h @@ -47,10 +47,6 @@ public: size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE, const WriteSettings & write_settings = {}) override; - // void removeObject(const StoredObject & object) override; - - // void removeObjects(const StoredObjects & objects) override; - void removeObjectIfExists(const StoredObject & object) override; void removeObjectsIfExist(const StoredObjects & objects) override; From 52dfad190dc2bb938f68464d42f69bd80ea1b422 Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Wed, 6 Nov 2024 15:46:58 +0000 Subject: [PATCH 485/680] Automatic style fix --- tests/ci/test_ci_config.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/ci/test_ci_config.py b/tests/ci/test_ci_config.py index 03f28983262..65418310c31 100644 --- a/tests/ci/test_ci_config.py +++ b/tests/ci/test_ci_config.py @@ -39,7 +39,8 @@ class TestCIConfig(unittest.TestCase): CI.BuildNames.PACKAGE_AARCH64_ASAN, ): self.assertTrue( - CI.JOB_CONFIGS[job].runner_type in (CI.Runners.BUILDER_AARCH64,), + CI.JOB_CONFIGS[job].runner_type + in (CI.Runners.BUILDER_AARCH64,), f"Job [{job}] must have [{CI.Runners.BUILDER_AARCH64}] runner", ) else: From 8bb656ddec205c9836db55c8a459a6b9c2cbf3d1 Mon Sep 17 00:00:00 2001 From: divanik Date: Wed, 6 Nov 2024 15:55:41 +0000 Subject: [PATCH 486/680] Add context manager for partition manager --- tests/integration/test_quorum_inserts/test.py | 81 ++++++++++--------- 1 file changed, 43 insertions(+), 38 deletions(-) diff --git a/tests/integration/test_quorum_inserts/test.py b/tests/integration/test_quorum_inserts/test.py index a646319c5f9..5e4a960acdf 100644 --- a/tests/integration/test_quorum_inserts/test.py +++ b/tests/integration/test_quorum_inserts/test.py @@ -379,50 +379,55 @@ def test_insert_quorum_with_keeper_loss_connection(started_cluster): ) ) - pm = PartitionManager() - pm.drop_instance_zk_connections(zero) + with PartitionManager() as pm: + pm.drop_instance_zk_connections(zero) - retries = 0 - zk = cluster.get_kazoo_client("zoo1") - while True: - if ( - zk.exists(f"/clickhouse/tables/{table_name}/replicas/zero/is_active") - is None - ): - break - print("replica is still active") - time.sleep(1) - retries += 1 - if retries == 120: - raise Exception("Can not wait cluster replica inactive") + retries = 0 + zk = cluster.get_kazoo_client("zoo1") + while True: + if ( + zk.exists( + f"/clickhouse/tables/{table_name}/replicas/zero/is_active" + ) + is None + ): + break + print("replica is still active") + time.sleep(1) + retries += 1 + if retries == 120: + raise Exception("Can not wait cluster replica inactive") - first.query("SYSTEM ENABLE FAILPOINT finish_set_quorum_failed_parts") - quorum_fail_future = executor.submit( - lambda: first.query( - "SYSTEM WAIT FAILPOINT finish_set_quorum_failed_parts", timeout=300 + first.query("SYSTEM ENABLE FAILPOINT finish_set_quorum_failed_parts") + quorum_fail_future = executor.submit( + lambda: first.query( + "SYSTEM WAIT FAILPOINT finish_set_quorum_failed_parts", timeout=300 + ) ) - ) - first.query(f"SYSTEM START FETCHES {table_name}") + first.query(f"SYSTEM START FETCHES {table_name}") - concurrent.futures.wait([quorum_fail_future]) + concurrent.futures.wait([quorum_fail_future]) - assert quorum_fail_future.exception() is None + assert quorum_fail_future.exception() is None - zero.query("SYSTEM ENABLE FAILPOINT finish_clean_quorum_failed_parts") - clean_quorum_fail_parts_future = executor.submit( - lambda: first.query( - "SYSTEM WAIT FAILPOINT finish_clean_quorum_failed_parts", timeout=300 + zero.query("SYSTEM ENABLE FAILPOINT finish_clean_quorum_failed_parts") + clean_quorum_fail_parts_future = executor.submit( + lambda: first.query( + "SYSTEM WAIT FAILPOINT finish_clean_quorum_failed_parts", + timeout=300, + ) ) - ) - pm.restore_instance_zk_connections(zero) - concurrent.futures.wait([clean_quorum_fail_parts_future]) + pm.restore_instance_zk_connections(zero) + concurrent.futures.wait([clean_quorum_fail_parts_future]) - assert clean_quorum_fail_parts_future.exception() is None + assert clean_quorum_fail_parts_future.exception() is None - zero.query("SYSTEM DISABLE FAILPOINT replicated_merge_tree_insert_retry_pause") - concurrent.futures.wait([insert_future]) - assert insert_future.exception() is not None - assert not zero.contains_in_log("LOGICAL_ERROR") - assert zero.contains_in_log( - "fails to commit and will not retry or clean garbage" - ) + zero.query( + "SYSTEM DISABLE FAILPOINT replicated_merge_tree_insert_retry_pause" + ) + concurrent.futures.wait([insert_future]) + assert insert_future.exception() is not None + assert not zero.contains_in_log("LOGICAL_ERROR") + assert zero.contains_in_log( + "fails to commit and will not retry or clean garbage" + ) From e8a8a4f62eabf854ebabff367d500bcc52456e83 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Wed, 6 Nov 2024 17:31:57 +0100 Subject: [PATCH 487/680] Add test to check that accessing system.functions does not populate query_log used_functions --- ...nctions_should_not_fill_query_log_functions.reference | 1 + ...tem_functions_should_not_fill_query_log_functions.sql | 9 +++++++++ 2 files changed, 10 insertions(+) create mode 100644 tests/queries/0_stateless/03262_system_functions_should_not_fill_query_log_functions.reference create mode 100644 tests/queries/0_stateless/03262_system_functions_should_not_fill_query_log_functions.sql diff --git a/tests/queries/0_stateless/03262_system_functions_should_not_fill_query_log_functions.reference b/tests/queries/0_stateless/03262_system_functions_should_not_fill_query_log_functions.reference new file mode 100644 index 00000000000..021c06382c8 --- /dev/null +++ b/tests/queries/0_stateless/03262_system_functions_should_not_fill_query_log_functions.reference @@ -0,0 +1 @@ +[] ['equals'] [] diff --git a/tests/queries/0_stateless/03262_system_functions_should_not_fill_query_log_functions.sql b/tests/queries/0_stateless/03262_system_functions_should_not_fill_query_log_functions.sql new file mode 100644 index 00000000000..7e6f384c0a8 --- /dev/null +++ b/tests/queries/0_stateless/03262_system_functions_should_not_fill_query_log_functions.sql @@ -0,0 +1,9 @@ +SELECT * FROM system.functions WHERE name = 'bitShiftLeft' format Null; +SYSTEM FLUSH LOGS; +SELECT used_aggregate_functions, used_functions, used_table_functions +FROM system.query_log +WHERE + event_date >= yesterday() + AND type = 'QueryFinish' + AND current_database = currentDatabase() + AND query LIKE '%bitShiftLeft%'; From 530c04413eaf2839fb3fbdef3619628916e63405 Mon Sep 17 00:00:00 2001 From: Maksim Kita Date: Wed, 6 Nov 2024 19:59:41 +0300 Subject: [PATCH 488/680] Analyzer materialized view IN with CTE fix --- src/Analyzer/QueryNode.h | 12 ++++ src/Analyzer/Resolve/QueryAnalyzer.cpp | 48 +++++++++----- src/Analyzer/UnionNode.cpp | 21 +++++++ src/Analyzer/UnionNode.h | 3 + ...er_materialized_view_in_with_cte.reference | 1 + ...analyzer_materialized_view_in_with_cte.sql | 63 +++++++++++++++++++ ...zer_materialized_view_cte_nested.reference | 0 ..._analyzer_materialized_view_cte_nested.sql | 19 ++++++ 8 files changed, 150 insertions(+), 17 deletions(-) create mode 100644 tests/queries/0_stateless/03262_analyzer_materialized_view_in_with_cte.reference create mode 100644 tests/queries/0_stateless/03262_analyzer_materialized_view_in_with_cte.sql create mode 100644 tests/queries/0_stateless/03263_analyzer_materialized_view_cte_nested.reference create mode 100644 tests/queries/0_stateless/03263_analyzer_materialized_view_cte_nested.sql diff --git a/src/Analyzer/QueryNode.h b/src/Analyzer/QueryNode.h index aef0c8805bb..2333fc56218 100644 --- a/src/Analyzer/QueryNode.h +++ b/src/Analyzer/QueryNode.h @@ -602,9 +602,21 @@ public: return projection_columns; } + /// Returns true if query node is resolved, false otherwise + bool isResolved() const + { + return !projection_columns.empty(); + } + /// Resolve query node projection columns void resolveProjectionColumns(NamesAndTypes projection_columns_value); + /// Clear query node projection columns + void clearProjectionColumns() + { + projection_columns.clear(); + } + /// Remove unused projection columns void removeUnusedProjectionColumns(const std::unordered_set & used_projection_columns); diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index cb3087af707..c0a2de0f125 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -2958,27 +2958,28 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi /// Replace storage with values storage of insertion block if (StoragePtr storage = scope.context->getViewSource()) { - QueryTreeNodePtr table_expression; - /// Process possibly nested sub-selects - for (auto * query_node = in_second_argument->as(); query_node; query_node = table_expression->as()) - table_expression = extractLeftTableExpression(query_node->getJoinTree()); + QueryTreeNodePtr table_expression = in_second_argument; - if (table_expression) + /// Process possibly nested sub-selects + while (table_expression) { - if (auto * query_table_node = table_expression->as()) - { - if (query_table_node->getStorageID().getFullNameNotQuoted() == storage->getStorageID().getFullNameNotQuoted()) - { - auto replacement_table_expression = std::make_shared(storage, scope.context); - if (std::optional table_expression_modifiers = query_table_node->getTableExpressionModifiers()) - replacement_table_expression->setTableExpressionModifiers(*table_expression_modifiers); - in_second_argument = in_second_argument->cloneAndReplace(table_expression, std::move(replacement_table_expression)); - } - } + if (auto * query_node = table_expression->as()) + table_expression = extractLeftTableExpression(query_node->getJoinTree()); + else if (auto * union_node = table_expression->as()) + table_expression = union_node->getQueries().getNodes().at(0); + else + break; + } + + auto * table_expression_table_node = table_expression->as(); + if (table_expression_table_node && + table_expression_table_node->getStorageID().getFullNameNotQuoted() == storage->getStorageID().getFullNameNotQuoted()) + { + auto replacement_table_expression_table_node = table_expression_table_node->clone(); + replacement_table_expression_table_node->as().updateStorage(storage, scope.context); + in_second_argument = in_second_argument->cloneAndReplace(table_expression, std::move(replacement_table_expression_table_node)); } } - - resolveExpressionNode(in_second_argument, scope, false /*allow_lambda_expression*/, true /*allow_table_expression*/); } /// Edge case when the first argument of IN is scalar subquery. @@ -5310,6 +5311,16 @@ void QueryAnalyzer::resolveQuery(const QueryTreeNodePtr & query_node, Identifier auto & query_node_typed = query_node->as(); + /** It is unsafe to call resolveQuery on already resolved query node, because during identifier resolution process + * we replace identifiers with expressions without aliases, also at the end of resolveQuery all aliases from all nodes will be removed. + * For subsequent resolveQuery executions it is possible to have wrong projection header, because for nodes + * with aliases projection name is alias. + * + * If for client it is necessary to resolve query node after clone, client must clear projection columns from query node before resolve. + */ + if (query_node_typed.isResolved()) + return; + if (query_node_typed.isCTE()) ctes_in_resolve_process.insert(query_node_typed.getCTEName()); @@ -5675,6 +5686,9 @@ void QueryAnalyzer::resolveUnion(const QueryTreeNodePtr & union_node, Identifier { auto & union_node_typed = union_node->as(); + if (union_node_typed.isResolved()) + return; + if (union_node_typed.isCTE()) ctes_in_resolve_process.insert(union_node_typed.getCTEName()); diff --git a/src/Analyzer/UnionNode.cpp b/src/Analyzer/UnionNode.cpp index 6f70f01e519..545a6b2195b 100644 --- a/src/Analyzer/UnionNode.cpp +++ b/src/Analyzer/UnionNode.cpp @@ -35,6 +35,7 @@ namespace ErrorCodes { extern const int TYPE_MISMATCH; extern const int BAD_ARGUMENTS; + extern const int LOGICAL_ERROR; } UnionNode::UnionNode(ContextMutablePtr context_, SelectUnionMode union_mode_) @@ -50,6 +51,26 @@ UnionNode::UnionNode(ContextMutablePtr context_, SelectUnionMode union_mode_) children[queries_child_index] = std::make_shared(); } +bool UnionNode::isResolved() const +{ + for (const auto & query_node : getQueries().getNodes()) + { + bool is_resolved = false; + + if (auto * query_node_typed = query_node->as()) + is_resolved = query_node_typed->isResolved(); + else if (auto * union_node_typed = query_node->as()) + is_resolved = union_node_typed->isResolved(); + else + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected query tree node type in UNION node"); + + if (!is_resolved) + return false; + } + + return true; +} + NamesAndTypes UnionNode::computeProjectionColumns() const { if (recursive_cte_table) diff --git a/src/Analyzer/UnionNode.h b/src/Analyzer/UnionNode.h index 40baad1ad57..85d6afb1e47 100644 --- a/src/Analyzer/UnionNode.h +++ b/src/Analyzer/UnionNode.h @@ -163,6 +163,9 @@ public: return children[queries_child_index]; } + /// Returns true if union node is resolved, false otherwise + bool isResolved() const; + /// Compute union node projection columns NamesAndTypes computeProjectionColumns() const; diff --git a/tests/queries/0_stateless/03262_analyzer_materialized_view_in_with_cte.reference b/tests/queries/0_stateless/03262_analyzer_materialized_view_in_with_cte.reference new file mode 100644 index 00000000000..5ddf8439af5 --- /dev/null +++ b/tests/queries/0_stateless/03262_analyzer_materialized_view_in_with_cte.reference @@ -0,0 +1 @@ +1 2 \N test diff --git a/tests/queries/0_stateless/03262_analyzer_materialized_view_in_with_cte.sql b/tests/queries/0_stateless/03262_analyzer_materialized_view_in_with_cte.sql new file mode 100644 index 00000000000..4543d336d14 --- /dev/null +++ b/tests/queries/0_stateless/03262_analyzer_materialized_view_in_with_cte.sql @@ -0,0 +1,63 @@ +SET allow_experimental_analyzer = 1; + +DROP TABLE IF EXISTS mv_test; +DROP TABLE IF EXISTS mv_test_target; +DROP VIEW IF EXISTS mv_test_mv; + +CREATE TABLE mv_test +( + `id` UInt64, + `ref_id` UInt64, + `final_id` Nullable(UInt64), + `display` String +) +ENGINE = Log; + +CREATE TABLE mv_test_target +( + `id` UInt64, + `ref_id` UInt64, + `final_id` Nullable(UInt64), + `display` String +) +ENGINE = Log; + +CREATE MATERIALIZED VIEW mv_test_mv TO mv_test_target +( + `id` UInt64, + `ref_id` UInt64, + `final_id` Nullable(UInt64), + `display` String +) +AS WITH + tester AS + ( + SELECT + id, + ref_id, + final_id, + display + FROM mv_test + ), + id_set AS + ( + SELECT + display, + max(id) AS max_id + FROM mv_test + GROUP BY display + ) +SELECT * +FROM tester +WHERE id IN ( + SELECT max_id + FROM id_set +); + +INSERT INTO mv_test ( id, ref_id, display) values ( 1, 2, 'test'); + +SELECT * FROM mv_test_target; + +DROP VIEW mv_test_mv; +DROP TABLE mv_test_target; +DROP TABLE mv_test; diff --git a/tests/queries/0_stateless/03263_analyzer_materialized_view_cte_nested.reference b/tests/queries/0_stateless/03263_analyzer_materialized_view_cte_nested.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/03263_analyzer_materialized_view_cte_nested.sql b/tests/queries/0_stateless/03263_analyzer_materialized_view_cte_nested.sql new file mode 100644 index 00000000000..4ea853a7c22 --- /dev/null +++ b/tests/queries/0_stateless/03263_analyzer_materialized_view_cte_nested.sql @@ -0,0 +1,19 @@ +SET allow_experimental_analyzer = 1; + +DROP TABLE IF EXISTS test_table; +DROP VIEW IF EXISTS test_mv; + +CREATE TABLE test_table ENGINE = MergeTree ORDER BY tuple() AS SELECT 1 as col1; + +CREATE MATERIALIZED VIEW test_mv ENGINE = MergeTree ORDER BY tuple() AS +WITH + subquery_on_source AS (SELECT col1 AS aliased FROM test_table), + output AS (SELECT * FROM test_table WHERE col1 IN (SELECT aliased FROM subquery_on_source)) +SELECT * FROM output; + +INSERT INTO test_table VALUES (2); + +SELECT * FROM test_mv; + +DROP VIEW test_mv; +DROP TABLE test_table; From 4ad8273e5f3d16f5a95220824223800b4a356e26 Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Wed, 6 Nov 2024 17:31:24 +0000 Subject: [PATCH 489/680] Enable merge filters optimization. --- src/Core/Settings.cpp | 2 +- src/Core/SettingsChangesHistory.cpp | 1 + .../QueryPlanOptimizationSettings.h | 2 +- .../03262_filter_push_down_view.reference | 2 ++ .../03262_filter_push_down_view.sql | 36 +++++++++++++++++++ 5 files changed, 41 insertions(+), 2 deletions(-) create mode 100644 tests/queries/0_stateless/03262_filter_push_down_view.reference create mode 100644 tests/queries/0_stateless/03262_filter_push_down_view.sql diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 081e07ca2ce..6f8047bbdf8 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -4554,7 +4554,7 @@ Possible values: - 0 - Disable - 1 - Enable )", 0) \ - DECLARE(Bool, query_plan_merge_filters, false, R"( + DECLARE(Bool, query_plan_merge_filters, true, R"( Allow to merge filters in the query plan )", 0) \ DECLARE(Bool, query_plan_filter_push_down, true, R"( diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index ed87fde8b7e..12350b6cdaf 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -74,6 +74,7 @@ static std::initializer_list Date: Wed, 6 Nov 2024 17:48:04 +0000 Subject: [PATCH 490/680] Add missing reference file --- .../0_stateless/02354_vector_search_multiple_indexes.reference | 0 1 file changed, 0 insertions(+), 0 deletions(-) create mode 100644 tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference diff --git a/tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference b/tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference new file mode 100644 index 00000000000..e69de29bb2d From de21dde4cfac2c2fcb7257d018afda9e99c19a11 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Wed, 6 Nov 2024 19:26:39 +0100 Subject: [PATCH 491/680] Avoid crash when using UDF in a constraint --- .../UserDefinedSQLFunctionVisitor.cpp | 99 +++---------------- src/Parsers/ASTColumnDeclaration.cpp | 10 ++ src/Parsers/ASTColumnDeclaration.h | 3 + .../03262_udf_in_constraint.reference | 2 + .../0_stateless/03262_udf_in_constraint.sh | 17 ++++ 5 files changed, 45 insertions(+), 86 deletions(-) create mode 100644 tests/queries/0_stateless/03262_udf_in_constraint.reference create mode 100755 tests/queries/0_stateless/03262_udf_in_constraint.sh diff --git a/src/Functions/UserDefined/UserDefinedSQLFunctionVisitor.cpp b/src/Functions/UserDefined/UserDefinedSQLFunctionVisitor.cpp index ebd65471449..a04b8d7b998 100644 --- a/src/Functions/UserDefined/UserDefinedSQLFunctionVisitor.cpp +++ b/src/Functions/UserDefined/UserDefinedSQLFunctionVisitor.cpp @@ -24,92 +24,7 @@ namespace ErrorCodes void UserDefinedSQLFunctionVisitor::visit(ASTPtr & ast) { - if (!ast) - { - chassert(false); - return; - } - - /// FIXME: this helper should use updatePointerToChild(), but - /// forEachPointerToChild() is not implemented for ASTColumnDeclaration - /// (and also some members should be adjusted for this). - const auto visit_child_with_shared_ptr = [&](ASTPtr & child) - { - if (!child) - return; - - auto * old_value = child.get(); - visit(child); - - // child did not change - if (old_value == child.get()) - return; - - // child changed, we need to modify it in the list of children of the parent also - for (auto & current_child : ast->children) - { - if (current_child.get() == old_value) - current_child = child; - } - }; - - if (auto * col_decl = ast->as()) - { - visit_child_with_shared_ptr(col_decl->default_expression); - visit_child_with_shared_ptr(col_decl->ttl); - return; - } - - if (auto * storage = ast->as()) - { - const auto visit_child = [&](IAST * & child) - { - if (!child) - return; - - if (const auto * function = child->template as()) - { - std::unordered_set udf_in_replace_process; - auto replace_result = tryToReplaceFunction(*function, udf_in_replace_process); - if (replace_result) - ast->setOrReplace(child, replace_result); - } - - visit(child); - }; - - visit_child(storage->partition_by); - visit_child(storage->primary_key); - visit_child(storage->order_by); - visit_child(storage->sample_by); - visit_child(storage->ttl_table); - - return; - } - - if (auto * alter = ast->as()) - { - /// It is OK to use updatePointerToChild() because ASTAlterCommand implements forEachPointerToChild() - const auto visit_child_update_parent = [&](ASTPtr & child) - { - if (!child) - return; - - auto * old_ptr = child.get(); - visit(child); - auto * new_ptr = child.get(); - - /// Some AST classes have naked pointers to children elements as members. - /// We have to replace them if the child was replaced. - if (new_ptr != old_ptr) - ast->updatePointerToChild(old_ptr, new_ptr); - }; - - for (auto & children : alter->children) - visit_child_update_parent(children); - - return; - } + chassert(ast); if (const auto * function = ast->template as()) { @@ -120,7 +35,19 @@ void UserDefinedSQLFunctionVisitor::visit(ASTPtr & ast) } for (auto & child : ast->children) + { + if (!child) + return; + + auto * old_ptr = child.get(); visit(child); + auto * new_ptr = child.get(); + + /// Some AST classes have naked pointers to children elements as members. + /// We have to replace them if the child was replaced. + if (new_ptr != old_ptr) + ast->updatePointerToChild(old_ptr, new_ptr); + } } void UserDefinedSQLFunctionVisitor::visit(IAST * ast) diff --git a/src/Parsers/ASTColumnDeclaration.cpp b/src/Parsers/ASTColumnDeclaration.cpp index e7c3fdbb548..1c7d72bafcc 100644 --- a/src/Parsers/ASTColumnDeclaration.cpp +++ b/src/Parsers/ASTColumnDeclaration.cpp @@ -128,4 +128,14 @@ void ASTColumnDeclaration::formatImpl(const FormatSettings & format_settings, Fo } } +void ASTColumnDeclaration::forEachPointerToChild(std::function f) +{ + f(reinterpret_cast(&default_expression)); + f(reinterpret_cast(&comment)); + f(reinterpret_cast(&codec)); + f(reinterpret_cast(&statistics_desc)); + f(reinterpret_cast(&ttl)); + f(reinterpret_cast(&collation)); + f(reinterpret_cast(&settings)); +} } diff --git a/src/Parsers/ASTColumnDeclaration.h b/src/Parsers/ASTColumnDeclaration.h index 914916d5074..0c5076f0201 100644 --- a/src/Parsers/ASTColumnDeclaration.h +++ b/src/Parsers/ASTColumnDeclaration.h @@ -29,6 +29,9 @@ public: ASTPtr clone() const override; void formatImpl(const FormatSettings & format_settings, FormatState & state, FormatStateStacked frame) const override; + +protected: + void forEachPointerToChild(std::function f) override; }; } diff --git a/tests/queries/0_stateless/03262_udf_in_constraint.reference b/tests/queries/0_stateless/03262_udf_in_constraint.reference new file mode 100644 index 00000000000..29d403b85a8 --- /dev/null +++ b/tests/queries/0_stateless/03262_udf_in_constraint.reference @@ -0,0 +1,2 @@ +CREATE TABLE default.t0\n(\n `c0` Int32,\n CONSTRAINT c1 CHECK c0 > 5\n)\nENGINE = MergeTree\nORDER BY tuple()\nSETTINGS index_granularity = 8192 +10 diff --git a/tests/queries/0_stateless/03262_udf_in_constraint.sh b/tests/queries/0_stateless/03262_udf_in_constraint.sh new file mode 100755 index 00000000000..3c36e7caeb4 --- /dev/null +++ b/tests/queries/0_stateless/03262_udf_in_constraint.sh @@ -0,0 +1,17 @@ +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +$CLICKHOUSE_CLIENT -q " + CREATE FUNCTION ${CLICKHOUSE_DATABASE}_function AS (x) -> x > 5; + CREATE TABLE t0 (c0 Int, CONSTRAINT c1 CHECK ${CLICKHOUSE_DATABASE}_function(c0)) ENGINE = MergeTree() ORDER BY tuple(); + SHOW CREATE TABLE t0; + INSERT INTO t0(c0) VALUES (10); + INSERT INTO t0(c0) VALUES (3); -- {serverError VIOLATED_CONSTRAINT} + SELECT * FROM t0; + + DROP TABLE t0; + DROP FUNCTION ${CLICKHOUSE_DATABASE}_function; +" From c55840794195689299ccb1b9f838fdb3d1a7edfa Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Wed, 6 Nov 2024 19:53:01 +0000 Subject: [PATCH 492/680] Remove duplicate test (same as 02354_vector_search_bugs_multiple_indexes.sql) --- ...02354_vector_search_multiple_indexes.reference | 0 .../02354_vector_search_multiple_indexes.sql | 15 --------------- 2 files changed, 15 deletions(-) delete mode 100644 tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference delete mode 100644 tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql diff --git a/tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference b/tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference deleted file mode 100644 index e69de29bb2d..00000000000 diff --git a/tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql b/tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql deleted file mode 100644 index aedba286a9f..00000000000 --- a/tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql +++ /dev/null @@ -1,15 +0,0 @@ --- Tags: no-fasttest, no-ordinary-database - --- Tests that multiple vector similarity indexes can be created on the same column (even if that makes no sense) - -SET allow_experimental_vector_similarity_index = 1; - -DROP TABLE IF EXISTS tab; - -CREATE TABLE tab (id Int32, vec Array(Float32), PRIMARY KEY id, INDEX vec_idx(vec) TYPE vector_similarity('hnsw', 'L2Distance')); - -ALTER TABLE tab ADD INDEX idx(vec) TYPE minmax; -ALTER TABLE tab ADD INDEX vec_idx1(vec) TYPE vector_similarity('hnsw', 'cosineDistance'); -ALTER TABLE tab ADD INDEX vec_idx2(vec) TYPE vector_similarity('hnsw', 'L2Distance'); -- silly but creating the same index also works for non-vector indexes ... - -DROP TABLE tab; From 9cdd56abbc4c75936fb4456d741b012fd710f98a Mon Sep 17 00:00:00 2001 From: Pablo Marcos Date: Tue, 5 Nov 2024 14:02:43 +0000 Subject: [PATCH 493/680] Reduce the general critical section for query_metric_log - Use a separate mutex for each query to reduce the contention period for queries_mutex. - Refactor to use std::mutex instead of std::recursive_mutex for queries_mutex. - In case we're running late to schedule the next task, schedule it immediately. - Fix LockGuard because unlocking twice is undefined behavior. --- base/base/defines.h | 1 + src/Common/LockGuard.h | 32 +++++- src/Interpreters/QueryMetricLog.cpp | 165 +++++++++++++++++++--------- src/Interpreters/QueryMetricLog.h | 43 ++++++-- 4 files changed, 179 insertions(+), 62 deletions(-) diff --git a/base/base/defines.h b/base/base/defines.h index 5685a6d9833..a0c3c0d1de5 100644 --- a/base/base/defines.h +++ b/base/base/defines.h @@ -145,6 +145,7 @@ #define TSA_TRY_ACQUIRE_SHARED(...) __attribute__((try_acquire_shared_capability(__VA_ARGS__))) /// function tries to acquire a shared capability and returns a boolean value indicating success or failure #define TSA_RELEASE_SHARED(...) __attribute__((release_shared_capability(__VA_ARGS__))) /// function releases the given shared capability #define TSA_SCOPED_LOCKABLE __attribute__((scoped_lockable)) /// object of a class has scoped lockable capability +#define TSA_RETURN_CAPABILITY(...) __attribute__((lock_returned(__VA_ARGS__))) /// to return capabilities in functions /// Macros for suppressing TSA warnings for specific reads/writes (instead of suppressing it for the whole function) /// They use a lambda function to apply function attribute to a single statement. This enable us to suppress warnings locally instead of diff --git a/src/Common/LockGuard.h b/src/Common/LockGuard.h index 8a98c5f553a..03c8a3e7617 100644 --- a/src/Common/LockGuard.h +++ b/src/Common/LockGuard.h @@ -1,23 +1,47 @@ #pragma once -#include #include +#include +#include namespace DB { +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +}; + /** LockGuard provides RAII-style locking mechanism for a mutex. - ** It's intended to be used like std::unique_ptr but with TSA annotations + ** It's intended to be used like std::unique_lock but with TSA annotations */ template class TSA_SCOPED_LOCKABLE LockGuard { public: - explicit LockGuard(Mutex & mutex_) TSA_ACQUIRE(mutex_) : mutex(mutex_) { mutex.lock(); } - ~LockGuard() TSA_RELEASE() { mutex.unlock(); } + explicit LockGuard(Mutex & mutex_) TSA_ACQUIRE(mutex_) : mutex(mutex_) { lock(); } + ~LockGuard() TSA_RELEASE() { if (locked) unlock(); } + + void lock() TSA_ACQUIRE() + { + /// Don't allow recursive_mutex for now. + if (locked) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Can't lock twice the same mutex"); + mutex.lock(); + locked = true; + } + + void unlock() TSA_RELEASE() + { + if (!locked) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Can't unlock the mutex without locking it first"); + mutex.unlock(); + locked = false; + } private: Mutex & mutex; + bool locked = false; }; template typename TLockGuard, typename Mutex> diff --git a/src/Interpreters/QueryMetricLog.cpp b/src/Interpreters/QueryMetricLog.cpp index 5ab3fe590e0..e784c357b29 100644 --- a/src/Interpreters/QueryMetricLog.cpp +++ b/src/Interpreters/QueryMetricLog.cpp @@ -1,6 +1,7 @@ #include #include #include +#include #include #include #include @@ -16,7 +17,6 @@ #include #include -#include namespace DB @@ -24,6 +24,20 @@ namespace DB static auto logger = getLogger("QueryMetricLog"); +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +}; + +String timePointToString(QueryMetricLog::TimePoint time) +{ + /// fmtlib supports subsecond formatting in 10.0.0. We're in 9.1.0, so we need to add the milliseconds ourselves. + auto seconds = std::chrono::time_point_cast(time); + auto microseconds = std::chrono::duration_cast(time - seconds).count(); + + return fmt::format("{:%Y.%m.%d %H:%M:%S}.{:06}", seconds, microseconds); +} + ColumnsDescription QueryMetricLogElement::getColumnsDescription() { ColumnsDescription result; @@ -87,36 +101,69 @@ void QueryMetricLog::shutdown() Base::shutdown(); } -void QueryMetricLog::startQuery(const String & query_id, TimePoint start_time, UInt64 interval_milliseconds) +void QueryMetricLog::collectMetric(const ProcessList & process_list, String query_id) { - QueryMetricLogStatus status; - status.interval_milliseconds = interval_milliseconds; - status.next_collect_time = start_time + std::chrono::milliseconds(interval_milliseconds); + auto current_time = std::chrono::system_clock::now(); + const auto query_info = process_list.getQueryInfo(query_id, false, true, false); + if (!query_info) + { + LOG_TRACE(logger, "Query {} is not running anymore, so we couldn't get its QueryStatusInfo", query_id); + return; + } + + LockGuard global_lock(queries_mutex); + auto it = queries.find(query_id); + + /// The query might have finished while the scheduled task is running. + if (it == queries.end()) + { + global_lock.unlock(); + LOG_TRACE(logger, "Query {} not found in the list. Finished while this collecting task was running", query_id); + return; + } + + auto & query_status = it->second; + if (!query_status.mutex) + { + global_lock.unlock(); + LOG_TRACE(logger, "Query {} finished while this collecting task was running", query_id); + return; + } + + LockGuard query_lock(query_status.getMutex()); + global_lock.unlock(); + + auto elem = query_status.createLogMetricElement(query_id, *query_info, current_time); + if (elem) + add(std::move(elem.value())); +} + +/// We use TSA_NO_THREAD_SAFETY_ANALYSIS to prevent TSA complaining that we're modifying the query_status fields +/// without locking the mutex. Since we're building it from scratch, there's no harm in not holding it. +/// If we locked it to make TSA happy, TSAN build would falsely complain about +/// lock-order-inversion (potential deadlock) +/// which is not a real issue since QueryMetricLogStatus's mutex cannot be locked by anything else +/// until we add it to the queries map. +void QueryMetricLog::startQuery(const String & query_id, TimePoint start_time, UInt64 interval_milliseconds) TSA_NO_THREAD_SAFETY_ANALYSIS +{ + QueryMetricLogStatus query_status; + query_status.interval_milliseconds = interval_milliseconds; + query_status.next_collect_time = start_time + std::chrono::milliseconds(interval_milliseconds); auto context = getContext(); const auto & process_list = context->getProcessList(); - status.task = context->getSchedulePool().createTask("QueryMetricLog", [this, &process_list, query_id] { - auto current_time = std::chrono::system_clock::now(); - const auto query_info = process_list.getQueryInfo(query_id, false, true, false); - if (!query_info) - { - LOG_TRACE(logger, "Query {} is not running anymore, so we couldn't get its QueryStatusInfo", query_id); - return; - } - - auto elem = createLogMetricElement(query_id, *query_info, current_time); - if (elem) - add(std::move(elem.value())); + query_status.task = context->getSchedulePool().createTask("QueryMetricLog", [this, &process_list, query_id] { + collectMetric(process_list, query_id); }); - std::lock_guard lock(queries_mutex); - status.task->scheduleAfter(interval_milliseconds); - queries.emplace(query_id, std::move(status)); + LockGuard global_lock(queries_mutex); + query_status.scheduleNext(query_id); + queries.emplace(query_id, std::move(query_status)); } void QueryMetricLog::finishQuery(const String & query_id, TimePoint finish_time, QueryStatusInfoPtr query_info) { - std::unique_lock lock(queries_mutex); + LockGuard global_lock(queries_mutex); auto it = queries.find(query_id); /// finishQuery may be called from logExceptionBeforeStart when the query has not even started @@ -124,9 +171,19 @@ void QueryMetricLog::finishQuery(const String & query_id, TimePoint finish_time, if (it == queries.end()) return; + auto & query_status = it->second; + decltype(query_status.mutex) query_mutex; + LockGuard query_lock(query_status.getMutex()); + + /// Move the query mutex here so that we hold it until the end, after removing the query from queries. + query_mutex = std::move(query_status.mutex); + query_status.mutex = {}; + + global_lock.unlock(); + if (query_info) { - auto elem = createLogMetricElement(query_id, *query_info, finish_time, false); + auto elem = query_status.createLogMetricElement(query_id, *query_info, finish_time, false); if (elem) add(std::move(elem.value())); } @@ -139,51 +196,62 @@ void QueryMetricLog::finishQuery(const String & query_id, TimePoint finish_time, /// that order. { /// Take ownership of the task so that we can destroy it in this scope after unlocking `queries_mutex`. - auto task = std::move(it->second.task); + auto task = std::move(query_status.task); /// Build an empty task for the old task to make sure it does not lock any mutex on its destruction. - it->second.task = {}; + query_status.task = {}; + query_lock.unlock(); + global_lock.lock(); queries.erase(query_id); /// Ensure `queries_mutex` is unlocked before calling task's destructor at the end of this /// scope which will lock `exec_mutex`. - lock.unlock(); + global_lock.unlock(); } } -std::optional QueryMetricLog::createLogMetricElement(const String & query_id, const QueryStatusInfo & query_info, TimePoint query_info_time, bool schedule_next) +void QueryMetricLogStatus::scheduleNext(String query_id) { - /// fmtlib supports subsecond formatting in 10.0.0. We're in 9.1.0, so we need to add the milliseconds ourselves. - auto seconds = std::chrono::time_point_cast(query_info_time); - auto microseconds = std::chrono::duration_cast(query_info_time - seconds).count(); - LOG_DEBUG(logger, "Collecting query_metric_log for query {} with QueryStatusInfo from {:%Y.%m.%d %H:%M:%S}.{:06}. Schedule next: {}", query_id, seconds, microseconds, schedule_next); - - std::unique_lock lock(queries_mutex); - auto query_status_it = queries.find(query_id); - - /// The query might have finished while the scheduled task is running. - if (query_status_it == queries.end()) + const auto now = std::chrono::system_clock::now(); + if (next_collect_time > now) { - lock.unlock(); - LOG_TRACE(logger, "Query {} finished already while this collecting task was running", query_id); - return {}; + const auto wait_time = std::chrono::duration_cast(next_collect_time - now).count(); + task->scheduleAfter(wait_time); } - - auto & query_status = query_status_it->second; - if (query_info_time <= query_status.last_collect_time) + else + { + LOG_TRACE(logger, "The next collecting task for query {} should have already run at {}. Scheduling it right now", + query_id, timePointToString(next_collect_time)); + task->schedule(); + } +} + +std::optional QueryMetricLogStatus::createLogMetricElement(const String & query_id, const QueryStatusInfo & query_info, TimePoint query_info_time, bool schedule_next) +{ + LOG_TRACE(logger, "Collecting query_metric_log for query {} and interval {} ms with QueryStatusInfo from {}. Schedule next: {}", + query_id, interval_milliseconds, timePointToString(query_info_time), schedule_next); + + if (query_info_time <= last_collect_time) { - lock.unlock(); LOG_TRACE(logger, "Query {} has a more recent metrics collected. Skipping this one", query_id); return {}; } - query_status.last_collect_time = query_info_time; + /// Leave some margin because task->scheduleAfter takes a value in milliseconds. + /// So, we can expect up to 1ms of drift since BackgroundSchedulePool will compare + /// time points in milliseconds. + static auto error_margin = std::chrono::milliseconds(1); + if (schedule_next && query_info_time + error_margin < next_collect_time) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Task to collect metric for query {} scheduled at {} but run at {}", + query_id, timePointToString(next_collect_time), timePointToString(query_info_time)); + + last_collect_time = query_info_time; QueryMetricLogElement elem; elem.event_time = timeInSeconds(query_info_time); elem.event_time_microseconds = timeInMicroseconds(query_info_time); - elem.query_id = query_status_it->first; + elem.query_id = query_id; elem.memory_usage = query_info.memory_usage > 0 ? query_info.memory_usage : 0; elem.peak_memory_usage = query_info.peak_memory_usage > 0 ? query_info.peak_memory_usage : 0; @@ -192,7 +260,7 @@ std::optional QueryMetricLog::createLogMetricElement(cons for (ProfileEvents::Event i = ProfileEvents::Event(0), end = ProfileEvents::end(); i < end; ++i) { const auto & new_value = (*(query_info.profile_counters))[i]; - auto & old_value = query_status.last_profile_events[i]; + auto & old_value = last_profile_events[i]; /// Profile event counters are supposed to be monotonic. However, at least the `NetworkReceiveBytes` can be inaccurate. /// So, since in the future the counter should always have a bigger value than in the past, we skip this event. @@ -214,9 +282,8 @@ std::optional QueryMetricLog::createLogMetricElement(cons if (schedule_next) { - query_status.next_collect_time += std::chrono::milliseconds(query_status.interval_milliseconds); - const auto wait_time = std::chrono::duration_cast(query_status.next_collect_time - std::chrono::system_clock::now()).count(); - query_status.task->scheduleAfter(wait_time); + next_collect_time += std::chrono::milliseconds(interval_milliseconds); + scheduleNext(query_id); } return elem; diff --git a/src/Interpreters/QueryMetricLog.h b/src/Interpreters/QueryMetricLog.h index 802cee7bf26..65764229b0a 100644 --- a/src/Interpreters/QueryMetricLog.h +++ b/src/Interpreters/QueryMetricLog.h @@ -1,5 +1,6 @@ #pragma once +#include #include #include #include @@ -11,11 +12,17 @@ #include #include +#include namespace DB { +namespace ErrorCodes +{ + extern const int LOGICAL_ERROR; +}; + /** QueryMetricLogElement is a log of query metric values measured at regular time interval. */ @@ -36,31 +43,49 @@ struct QueryMetricLogElement struct QueryMetricLogStatus { + using TimePoint = std::chrono::system_clock::time_point; + using Mutex = std::mutex; + UInt64 interval_milliseconds; - std::chrono::system_clock::time_point last_collect_time; - std::chrono::system_clock::time_point next_collect_time; - std::vector last_profile_events = std::vector(ProfileEvents::end()); - BackgroundSchedulePool::TaskHolder task; + std::chrono::system_clock::time_point last_collect_time TSA_GUARDED_BY(getMutex()); + std::chrono::system_clock::time_point next_collect_time TSA_GUARDED_BY(getMutex()); + std::vector last_profile_events TSA_GUARDED_BY(getMutex()) = std::vector(ProfileEvents::end()); + BackgroundSchedulePool::TaskHolder task TSA_GUARDED_BY(getMutex()); + + /// We need to be able to move it for the hash map, so we need to add an indirection here. + std::unique_ptr mutex = std::make_unique(); + + /// Return a reference to the mutex, used for Thread Sanitizer annotations. + Mutex & getMutex() const TSA_RETURN_CAPABILITY(mutex) + { + if (!mutex) + throw Exception(ErrorCodes::LOGICAL_ERROR, "Mutex cannot be NULL"); + return *mutex; + } + + void scheduleNext(String query_id) TSA_REQUIRES(getMutex()); + std::optional createLogMetricElement(const String & query_id, const QueryStatusInfo & query_info, TimePoint query_info_time, bool schedule_next = true) TSA_REQUIRES(getMutex()); }; class QueryMetricLog : public SystemLog { using SystemLog::SystemLog; - using TimePoint = std::chrono::system_clock::time_point; using Base = SystemLog; public: + using TimePoint = std::chrono::system_clock::time_point; + void shutdown() final; - // Both startQuery and finishQuery are called from the thread that executes the query + /// Both startQuery and finishQuery are called from the thread that executes the query. void startQuery(const String & query_id, TimePoint start_time, UInt64 interval_milliseconds); void finishQuery(const String & query_id, TimePoint finish_time, QueryStatusInfoPtr query_info = nullptr); private: - std::optional createLogMetricElement(const String & query_id, const QueryStatusInfo & query_info, TimePoint query_info_time, bool schedule_next = true); + void collectMetric(const ProcessList & process_list, String query_id); - std::recursive_mutex queries_mutex; - std::unordered_map queries; + std::mutex queries_mutex; + std::unordered_map queries TSA_GUARDED_BY(queries_mutex); }; } From 26f0ba2c4ceb4b6d52f159943de63d4f2ca10520 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Wed, 6 Nov 2024 21:23:06 +0100 Subject: [PATCH 494/680] Update compatibility section for clickhouse-server docker image --- docker/server/README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/docker/server/README.md b/docker/server/README.md index 65239126790..1dc636414ac 100644 --- a/docker/server/README.md +++ b/docker/server/README.md @@ -20,6 +20,7 @@ For more information and documentation see https://clickhouse.com/. - The amd64 image requires support for [SSE3 instructions](https://en.wikipedia.org/wiki/SSE3). Virtually all x86 CPUs after 2005 support SSE3. - The arm64 image requires support for the [ARMv8.2-A architecture](https://en.wikipedia.org/wiki/AArch64#ARMv8.2-A) and additionally the Load-Acquire RCpc register. The register is optional in version ARMv8.2-A and mandatory in [ARMv8.3-A](https://en.wikipedia.org/wiki/AArch64#ARMv8.3-A). Supported in Graviton >=2, Azure and GCP instances. Examples for unsupported devices are Raspberry Pi 4 (ARMv8.0-A) and Jetson AGX Xavier/Orin (ARMv8.2-A). +- Since the Clickhouse 24.11 Ubuntu images started using `ubuntu:22.04` as its base image. It requires docker version >= `20.10.10` containing [patch](https://github.com/moby/moby/commit/977283509f75303bc6612665a04abf76ff1d2468). As a workaround you could use `docker run [--privileged | --security-opt seccomp=unconfined]` instead, however that has security implications. ## How to use this image From 157f745136094eb2eaeae72f17d103928194fd52 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Wed, 6 Nov 2024 22:09:12 +0100 Subject: [PATCH 495/680] Write a simple troubleshooting for an old docker and clickhouse-server --- docs/en/operations/_troubleshooting.md | 28 ++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/docs/en/operations/_troubleshooting.md b/docs/en/operations/_troubleshooting.md index 77389782675..f0ee1ca1d29 100644 --- a/docs/en/operations/_troubleshooting.md +++ b/docs/en/operations/_troubleshooting.md @@ -65,6 +65,34 @@ sudo rm -f /etc/yum.repos.d/clickhouse.repo After that follow the [install guide](../getting-started/install.md#from-rpm-packages) +### You Can't Run Docker Container + +You are running a simple `docker run clickhouse/clickhouse-server` and it crashes with a stack trace similar to following: + +``` +$ docker run -it clickhouse/clickhouse-server +........ +2024.11.06 21:04:48.912036 [ 1 ] {} SentryWriter: Sending crash reports is disabled +Poco::Exception. Code: 1000, e.code() = 0, System exception: cannot start thread, Stack trace (when copying this message, always include the lines below): + +0. Poco::ThreadImpl::startImpl(Poco::SharedPtr>) @ 0x00000000157c7b34 +1. Poco::Thread::start(Poco::Runnable&) @ 0x00000000157c8a0e +2. BaseDaemon::initializeTerminationAndSignalProcessing() @ 0x000000000d267a14 +3. BaseDaemon::initialize(Poco::Util::Application&) @ 0x000000000d2652cb +4. DB::Server::initialize(Poco::Util::Application&) @ 0x000000000d128b38 +5. Poco::Util::Application::run() @ 0x000000001581cfda +6. DB::Server::run() @ 0x000000000d1288f0 +7. Poco::Util::ServerApplication::run(int, char**) @ 0x0000000015825e27 +8. mainEntryClickHouseServer(int, char**) @ 0x000000000d125b38 +9. main @ 0x0000000007ea4eee +10. ? @ 0x00007f67ff946d90 +11. ? @ 0x00007f67ff946e40 +12. _start @ 0x00000000062e802e + (version 24.10.1.2812 (official build)) +``` + +The reason is an old docker daemon with version lower than `20.10.10`. A way to fix it either upgrading it, or running `docker run [--privileged | --security-opt seccomp=unconfined]`. The latter has security implications. + ## Connecting to the Server {#troubleshooting-accepts-no-connections} Possible issues: From 29aed6a58629dadca25840e976a4e680ac55a963 Mon Sep 17 00:00:00 2001 From: Michael Kolupaev Date: Wed, 6 Nov 2024 23:38:56 +0000 Subject: [PATCH 496/680] Fix compatibility with refreshable materialized views created by old clickhouse servers --- src/Storages/StorageMaterializedView.cpp | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/src/Storages/StorageMaterializedView.cpp b/src/Storages/StorageMaterializedView.cpp index d047b28e076..d56b09eec67 100644 --- a/src/Storages/StorageMaterializedView.cpp +++ b/src/Storages/StorageMaterializedView.cpp @@ -228,10 +228,20 @@ StorageMaterializedView::StorageMaterializedView( if (!fixed_uuid) { - if (to_inner_uuid != UUIDHelpers::Nil) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "TO INNER UUID is not allowed for materialized views with REFRESH without APPEND"); - if (to_table_id.hasUUID()) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "explicit UUID is not allowed for target table of materialized view with REFRESH without APPEND"); + if (mode >= LoadingStrictnessLevel::ATTACH) + { + /// Old versions of ClickHouse (when refreshable MV was experimental) could add useless + /// UUIDs to attach queries. + to_table_id.uuid = UUIDHelpers::Nil; + to_inner_uuid = UUIDHelpers::Nil; + } + else + { + if (to_inner_uuid != UUIDHelpers::Nil) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "TO INNER UUID is not allowed for materialized views with REFRESH without APPEND"); + if (to_table_id.hasUUID()) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "explicit UUID is not allowed for target table of materialized view with REFRESH without APPEND"); + } } if (!has_inner_table) From 8fb52b72b5bc1a4324cedaf2171e1af4e777f1af Mon Sep 17 00:00:00 2001 From: cangyin Date: Fri, 14 Jun 2024 12:58:46 +0000 Subject: [PATCH 497/680] Fix use-after-dtor logic in hashtable destroyElements --- src/Common/HashTable/HashTable.h | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) diff --git a/src/Common/HashTable/HashTable.h b/src/Common/HashTable/HashTable.h index f4374a0f2ca..d379c3f6a87 100644 --- a/src/Common/HashTable/HashTable.h +++ b/src/Common/HashTable/HashTable.h @@ -658,16 +658,11 @@ protected: { if (!std::is_trivially_destructible_v) { - for (iterator it = begin(), it_end = end(); it != it_end; ++it) + for (iterator it = begin(), it_end = end(); it != it_end;) { - it.ptr->~Cell(); - /// In case of poison_in_dtor=1 it will be poisoned, - /// but it maybe used later, during iteration. - /// - /// NOTE, that technically this is UB [1], but OK for now. - /// - /// [1]: https://github.com/google/sanitizers/issues/854#issuecomment-329661378 - __msan_unpoison(it.ptr, sizeof(*it.ptr)); + auto ptr = it.ptr; + ++it; + ptr->~Cell(); } /// Everything had been destroyed in the loop above, reset the flag From 042e82c6a9cbfa97d68cebb10e88c412c435cd3b Mon Sep 17 00:00:00 2001 From: Maksim Kita Date: Thu, 7 Nov 2024 13:10:51 +0300 Subject: [PATCH 498/680] Fix tests --- src/Analyzer/Resolve/QueryAnalyzer.cpp | 3 ++- .../03263_analyzer_materialized_view_cte_nested.reference | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index c0a2de0f125..c2eac8d008b 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -2971,7 +2971,8 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi break; } - auto * table_expression_table_node = table_expression->as(); + TableNode * table_expression_table_node = table_expression ? table_expression->as() : nullptr; + if (table_expression_table_node && table_expression_table_node->getStorageID().getFullNameNotQuoted() == storage->getStorageID().getFullNameNotQuoted()) { diff --git a/tests/queries/0_stateless/03263_analyzer_materialized_view_cte_nested.reference b/tests/queries/0_stateless/03263_analyzer_materialized_view_cte_nested.reference index e69de29bb2d..0cfbf08886f 100644 --- a/tests/queries/0_stateless/03263_analyzer_materialized_view_cte_nested.reference +++ b/tests/queries/0_stateless/03263_analyzer_materialized_view_cte_nested.reference @@ -0,0 +1 @@ +2 From e7ad525e0033e1a42cfe6ba35e2a9f0ecd2088b0 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Wed, 6 Nov 2024 20:03:14 +0000 Subject: [PATCH 499/680] Re-introduce support for legacy index creation syntax --- .../table-engines/mergetree-family/annindexes.md | 6 +++--- .../MergeTree/MergeTreeIndexVectorSimilarity.cpp | 6 ++++-- ...or_search_legacy_index_creation_syntax.reference | 0 ...4_vector_search_legacy_index_creation_syntax.sql | 13 +++++++++++++ 4 files changed, 20 insertions(+), 5 deletions(-) create mode 100644 tests/queries/0_stateless/02354_vector_search_legacy_index_creation_syntax.reference create mode 100644 tests/queries/0_stateless/02354_vector_search_legacy_index_creation_syntax.sql diff --git a/docs/en/engines/table-engines/mergetree-family/annindexes.md b/docs/en/engines/table-engines/mergetree-family/annindexes.md index dc12a60e8ef..fcdc16637e6 100644 --- a/docs/en/engines/table-engines/mergetree-family/annindexes.md +++ b/docs/en/engines/table-engines/mergetree-family/annindexes.md @@ -54,7 +54,7 @@ Parameters: - `distance_function`: either `L2Distance` (the [Euclidean distance](https://en.wikipedia.org/wiki/Euclidean_distance) - the length of a line between two points in Euclidean space), or `cosineDistance` (the [cosine distance](https://en.wikipedia.org/wiki/Cosine_similarity#Cosine_distance)- the angle between two non-zero vectors). -- `quantization`: either `f64`, `f32`, `f16`, `bf16`, or `i8` for storing the vector with reduced precision (optional, default: `bf16`) +- `quantization`: either `f64`, `f32`, `f16`, `bf16`, or `i8` for storing vectors with reduced precision (optional, default: `bf16`) - `hnsw_max_connections_per_layer`: the number of neighbors per HNSW graph node, also known as `M` in the [HNSW paper](https://doi.org/10.1109/TPAMI.2018.2889473) (optional, default: 32) - `hnsw_candidate_list_size_for_construction`: the size of the dynamic candidate list when constructing the HNSW graph, also known as @@ -92,8 +92,8 @@ Vector similarity indexes currently support two distance functions: - `cosineDistance`, also called cosine similarity, is the cosine of the angle between two (non-zero) vectors ([Wikipedia](https://en.wikipedia.org/wiki/Cosine_similarity)). -Vector similarity indexes allows storing the vectors in reduced precision formats. Supported scalar kinds are `f64`, `f32`, `f16` or `i8`. -If no scalar kind was specified during index creation, `f16` is used as default. +Vector similarity indexes allows storing the vectors in reduced precision formats. Supported scalar kinds are `f64`, `f32`, `f16`, `bf16`, +and `i8`. If no scalar kind was specified during index creation, `bf16` is used as default. For normalized data, `L2Distance` is usually a better choice, otherwise `cosineDistance` is recommended to compensate for scale. If no distance function was specified during index creation, `L2Distance` is used as default. diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp index f95b840e223..cca3ca6ce3b 100644 --- a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp @@ -531,15 +531,17 @@ void vectorSimilarityIndexValidator(const IndexDescription & index, bool /* atta { const bool has_two_args = (index.arguments.size() == 2); const bool has_five_args = (index.arguments.size() == 5); + const bool has_six_args = (index.arguments.size() == 6); /// Legacy index creation syntax before #70616. Supported only to be able to load old tables, can be removed mid-2025. + /// The 6th argument (ef_search) is ignored. /// Check number and type of arguments - if (!has_two_args && !has_five_args) + if (!has_two_args && !has_five_args && !has_six_args) throw Exception(ErrorCodes::INCORRECT_QUERY, "Vector similarity index must have two or five arguments"); if (index.arguments[0].getType() != Field::Types::String) throw Exception(ErrorCodes::INCORRECT_QUERY, "First argument of vector similarity index (method) must be of type String"); if (index.arguments[1].getType() != Field::Types::String) throw Exception(ErrorCodes::INCORRECT_QUERY, "Second argument of vector similarity index (metric) must be of type String"); - if (has_five_args) + if (has_five_args || has_six_args) { if (index.arguments[2].getType() != Field::Types::String) throw Exception(ErrorCodes::INCORRECT_QUERY, "Third argument of vector similarity index (quantization) must be of type String"); diff --git a/tests/queries/0_stateless/02354_vector_search_legacy_index_creation_syntax.reference b/tests/queries/0_stateless/02354_vector_search_legacy_index_creation_syntax.reference new file mode 100644 index 00000000000..e69de29bb2d diff --git a/tests/queries/0_stateless/02354_vector_search_legacy_index_creation_syntax.sql b/tests/queries/0_stateless/02354_vector_search_legacy_index_creation_syntax.sql new file mode 100644 index 00000000000..e5dbc6aa6a9 --- /dev/null +++ b/tests/queries/0_stateless/02354_vector_search_legacy_index_creation_syntax.sql @@ -0,0 +1,13 @@ +-- Tags: no-fasttest, no-ordinary-database + +-- Tests the legacy syntax to create vector similarity indexes before #70616. +-- Support for this syntax can be removed after mid-2025. + +SET allow_experimental_vector_similarity_index = 1; + +DROP TABLE IF EXISTS tab; + +CREATE TABLE tab(id Int32, vec Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'cosineDistance', 'f32', 42, 99, 113)) ENGINE = MergeTree ORDER BY id; -- Note the 6th parameter: 133 + +DROP TABLE tab; + From cf594010c862a568b07a440c4d70f9d59319b1a7 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Thu, 7 Nov 2024 09:43:42 +0000 Subject: [PATCH 500/680] Rename some tests for more consistency --- ...=> 02354_vector_search_adaptive_index_granularity.reference} | 0 ...y.sql => 02354_vector_search_adaptive_index_granularity.sql} | 0 ...=> 02354_vector_search_and_other_skipping_indexes.reference} | 0 ...1.sql => 02354_vector_search_and_other_skipping_indexes.sql} | 2 +- ...ence => 02354_vector_search_different_array_sizes.reference} | 0 ..._sizes.sql => 02354_vector_search_different_array_sizes.sql} | 0 ...2354_vector_search_empty_arrays_or_default_values.reference} | 0 ...l => 02354_vector_search_empty_arrays_or_default_values.sql} | 2 +- ...reference => 02354_vector_search_multiple_indexes.reference} | 0 ...ple_indexes.sql => 02354_vector_search_multiple_indexes.sql} | 0 ...s.reference => 02354_vector_search_multiple_marks.reference} | 0 ...ultiple_marks.sql => 02354_vector_search_multiple_marks.sql} | 0 ...g_69085.reference => 02354_vector_search_subquery.reference} | 0 ...or_search_bug_69085.sql => 02354_vector_search_subquery.sql} | 2 +- 14 files changed, 3 insertions(+), 3 deletions(-) rename tests/queries/0_stateless/{02354_vector_search_bug_52282.reference => 02354_vector_search_adaptive_index_granularity.reference} (100%) rename tests/queries/0_stateless/{02354_vector_search_bug_adaptive_index_granularity.sql => 02354_vector_search_adaptive_index_granularity.sql} (100%) rename tests/queries/0_stateless/{02354_vector_search_bug_71381.reference => 02354_vector_search_and_other_skipping_indexes.reference} (100%) rename tests/queries/0_stateless/{02354_vector_search_bug_71381.sql => 02354_vector_search_and_other_skipping_indexes.sql} (79%) rename tests/queries/0_stateless/{02354_vector_search_bug_adaptive_index_granularity.reference => 02354_vector_search_different_array_sizes.reference} (100%) rename tests/queries/0_stateless/{02354_vector_search_bug_different_array_sizes.sql => 02354_vector_search_different_array_sizes.sql} (100%) rename tests/queries/0_stateless/{02354_vector_search_bug_different_array_sizes.reference => 02354_vector_search_empty_arrays_or_default_values.reference} (100%) rename tests/queries/0_stateless/{02354_vector_search_bug_52282.sql => 02354_vector_search_empty_arrays_or_default_values.sql} (80%) rename tests/queries/0_stateless/{02354_vector_search_bug_multiple_indexes.reference => 02354_vector_search_multiple_indexes.reference} (100%) rename tests/queries/0_stateless/{02354_vector_search_bug_multiple_indexes.sql => 02354_vector_search_multiple_indexes.sql} (100%) rename tests/queries/0_stateless/{02354_vector_search_bug_multiple_marks.reference => 02354_vector_search_multiple_marks.reference} (100%) rename tests/queries/0_stateless/{02354_vector_search_bug_multiple_marks.sql => 02354_vector_search_multiple_marks.sql} (100%) rename tests/queries/0_stateless/{02354_vector_search_bug_69085.reference => 02354_vector_search_subquery.reference} (100%) rename tests/queries/0_stateless/{02354_vector_search_bug_69085.sql => 02354_vector_search_subquery.sql} (93%) diff --git a/tests/queries/0_stateless/02354_vector_search_bug_52282.reference b/tests/queries/0_stateless/02354_vector_search_adaptive_index_granularity.reference similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_bug_52282.reference rename to tests/queries/0_stateless/02354_vector_search_adaptive_index_granularity.reference diff --git a/tests/queries/0_stateless/02354_vector_search_bug_adaptive_index_granularity.sql b/tests/queries/0_stateless/02354_vector_search_adaptive_index_granularity.sql similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_bug_adaptive_index_granularity.sql rename to tests/queries/0_stateless/02354_vector_search_adaptive_index_granularity.sql diff --git a/tests/queries/0_stateless/02354_vector_search_bug_71381.reference b/tests/queries/0_stateless/02354_vector_search_and_other_skipping_indexes.reference similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_bug_71381.reference rename to tests/queries/0_stateless/02354_vector_search_and_other_skipping_indexes.reference diff --git a/tests/queries/0_stateless/02354_vector_search_bug_71381.sql b/tests/queries/0_stateless/02354_vector_search_and_other_skipping_indexes.sql similarity index 79% rename from tests/queries/0_stateless/02354_vector_search_bug_71381.sql rename to tests/queries/0_stateless/02354_vector_search_and_other_skipping_indexes.sql index 9e3246700b8..386d3b6e26e 100644 --- a/tests/queries/0_stateless/02354_vector_search_bug_71381.sql +++ b/tests/queries/0_stateless/02354_vector_search_and_other_skipping_indexes.sql @@ -2,7 +2,7 @@ SET allow_experimental_vector_similarity_index = 1; --- Issue #71381: Usage of vector similarity index and further skipping indexes on the same table +-- Usage of vector similarity index and further skipping indexes on the same table (issue #71381) DROP TABLE IF EXISTS tab; diff --git a/tests/queries/0_stateless/02354_vector_search_bug_adaptive_index_granularity.reference b/tests/queries/0_stateless/02354_vector_search_different_array_sizes.reference similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_bug_adaptive_index_granularity.reference rename to tests/queries/0_stateless/02354_vector_search_different_array_sizes.reference diff --git a/tests/queries/0_stateless/02354_vector_search_bug_different_array_sizes.sql b/tests/queries/0_stateless/02354_vector_search_different_array_sizes.sql similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_bug_different_array_sizes.sql rename to tests/queries/0_stateless/02354_vector_search_different_array_sizes.sql diff --git a/tests/queries/0_stateless/02354_vector_search_bug_different_array_sizes.reference b/tests/queries/0_stateless/02354_vector_search_empty_arrays_or_default_values.reference similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_bug_different_array_sizes.reference rename to tests/queries/0_stateless/02354_vector_search_empty_arrays_or_default_values.reference diff --git a/tests/queries/0_stateless/02354_vector_search_bug_52282.sql b/tests/queries/0_stateless/02354_vector_search_empty_arrays_or_default_values.sql similarity index 80% rename from tests/queries/0_stateless/02354_vector_search_bug_52282.sql rename to tests/queries/0_stateless/02354_vector_search_empty_arrays_or_default_values.sql index b8066ce278a..e24b1a527be 100644 --- a/tests/queries/0_stateless/02354_vector_search_bug_52282.sql +++ b/tests/queries/0_stateless/02354_vector_search_empty_arrays_or_default_values.sql @@ -2,7 +2,7 @@ SET allow_experimental_vector_similarity_index = 1; --- Issue #52258: Vector similarity indexes must reject empty Arrays or Arrays with default values +-- Vector similarity indexes must reject empty Arrays or Arrays with default values (issue #52258) DROP TABLE IF EXISTS tab; diff --git a/tests/queries/0_stateless/02354_vector_search_bug_multiple_indexes.reference b/tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_bug_multiple_indexes.reference rename to tests/queries/0_stateless/02354_vector_search_multiple_indexes.reference diff --git a/tests/queries/0_stateless/02354_vector_search_bug_multiple_indexes.sql b/tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_bug_multiple_indexes.sql rename to tests/queries/0_stateless/02354_vector_search_multiple_indexes.sql diff --git a/tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.reference b/tests/queries/0_stateless/02354_vector_search_multiple_marks.reference similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.reference rename to tests/queries/0_stateless/02354_vector_search_multiple_marks.reference diff --git a/tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.sql b/tests/queries/0_stateless/02354_vector_search_multiple_marks.sql similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_bug_multiple_marks.sql rename to tests/queries/0_stateless/02354_vector_search_multiple_marks.sql diff --git a/tests/queries/0_stateless/02354_vector_search_bug_69085.reference b/tests/queries/0_stateless/02354_vector_search_subquery.reference similarity index 100% rename from tests/queries/0_stateless/02354_vector_search_bug_69085.reference rename to tests/queries/0_stateless/02354_vector_search_subquery.reference diff --git a/tests/queries/0_stateless/02354_vector_search_bug_69085.sql b/tests/queries/0_stateless/02354_vector_search_subquery.sql similarity index 93% rename from tests/queries/0_stateless/02354_vector_search_bug_69085.sql rename to tests/queries/0_stateless/02354_vector_search_subquery.sql index 4dbcdf66e36..65ad0dbcd97 100644 --- a/tests/queries/0_stateless/02354_vector_search_bug_69085.sql +++ b/tests/queries/0_stateless/02354_vector_search_subquery.sql @@ -3,7 +3,7 @@ SET allow_experimental_vector_similarity_index = 1; SET enable_analyzer = 0; --- Issue #69085: Reference vector for vector search is computed by a subquery +-- Reference vector for vector search is computed by a subquery (issue #69085) DROP TABLE IF EXISTS tab; From be10aba49aca0d3253e4c714eabed196fe6411e2 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Thu, 7 Nov 2024 10:42:51 +0000 Subject: [PATCH 501/680] Minor cleanup --- .../MergeTree/MergeTreeIndexVectorSimilarity.cpp | 9 +++------ src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h | 3 --- 2 files changed, 3 insertions(+), 9 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp index cca3ca6ce3b..0b17fa05072 100644 --- a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp @@ -178,23 +178,20 @@ String USearchIndexWithSerialization::Statistics::toString() const } MergeTreeIndexGranuleVectorSimilarity::MergeTreeIndexGranuleVectorSimilarity( const String & index_name_, - const Block & index_sample_block_, unum::usearch::metric_kind_t metric_kind_, unum::usearch::scalar_kind_t scalar_kind_, UsearchHnswParams usearch_hnsw_params_) - : MergeTreeIndexGranuleVectorSimilarity(index_name_, index_sample_block_, metric_kind_, scalar_kind_, usearch_hnsw_params_, nullptr) + : MergeTreeIndexGranuleVectorSimilarity(index_name_, metric_kind_, scalar_kind_, usearch_hnsw_params_, nullptr) { } MergeTreeIndexGranuleVectorSimilarity::MergeTreeIndexGranuleVectorSimilarity( const String & index_name_, - const Block & index_sample_block_, unum::usearch::metric_kind_t metric_kind_, unum::usearch::scalar_kind_t scalar_kind_, UsearchHnswParams usearch_hnsw_params_, USearchIndexWithSerializationPtr index_) : index_name(index_name_) - , index_sample_block(index_sample_block_) , metric_kind(metric_kind_) , scalar_kind(scalar_kind_) , usearch_hnsw_params(usearch_hnsw_params_) @@ -261,7 +258,7 @@ MergeTreeIndexAggregatorVectorSimilarity::MergeTreeIndexAggregatorVectorSimilari MergeTreeIndexGranulePtr MergeTreeIndexAggregatorVectorSimilarity::getGranuleAndReset() { - auto granule = std::make_shared(index_name, index_sample_block, metric_kind, scalar_kind, usearch_hnsw_params, index); + auto granule = std::make_shared(index_name, metric_kind, scalar_kind, usearch_hnsw_params, index); index = nullptr; return granule; } @@ -490,7 +487,7 @@ MergeTreeIndexVectorSimilarity::MergeTreeIndexVectorSimilarity( MergeTreeIndexGranulePtr MergeTreeIndexVectorSimilarity::createIndexGranule() const { - return std::make_shared(index.name, index.sample_block, metric_kind, scalar_kind, usearch_hnsw_params); + return std::make_shared(index.name, metric_kind, scalar_kind, usearch_hnsw_params); } MergeTreeIndexAggregatorPtr MergeTreeIndexVectorSimilarity::createIndexAggregator(const MergeTreeWriterSettings & /*settings*/) const diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h index 9a81e168393..fe5049daf77 100644 --- a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h @@ -69,14 +69,12 @@ struct MergeTreeIndexGranuleVectorSimilarity final : public IMergeTreeIndexGranu { MergeTreeIndexGranuleVectorSimilarity( const String & index_name_, - const Block & index_sample_block_, unum::usearch::metric_kind_t metric_kind_, unum::usearch::scalar_kind_t scalar_kind_, UsearchHnswParams usearch_hnsw_params_); MergeTreeIndexGranuleVectorSimilarity( const String & index_name_, - const Block & index_sample_block_, unum::usearch::metric_kind_t metric_kind_, unum::usearch::scalar_kind_t scalar_kind_, UsearchHnswParams usearch_hnsw_params_, @@ -90,7 +88,6 @@ struct MergeTreeIndexGranuleVectorSimilarity final : public IMergeTreeIndexGranu bool empty() const override { return !index || index->size() == 0; } const String index_name; - const Block index_sample_block; const unum::usearch::metric_kind_t metric_kind; const unum::usearch::scalar_kind_t scalar_kind; const UsearchHnswParams usearch_hnsw_params; From f229fc5b40bd0faa3f312bcdc3123cfdfb6a70fc Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Thu, 7 Nov 2024 12:14:09 +0100 Subject: [PATCH 502/680] Deprecate CLICKHOUSE_UID/CLICKHOUSE_GID docker ENV --- docker/keeper/entrypoint.sh | 14 ++++++++------ docker/server/entrypoint.sh | 14 ++++++++------ 2 files changed, 16 insertions(+), 12 deletions(-) diff --git a/docker/keeper/entrypoint.sh b/docker/keeper/entrypoint.sh index 68bd0ef9d87..c5d5d26ec11 100644 --- a/docker/keeper/entrypoint.sh +++ b/docker/keeper/entrypoint.sh @@ -9,13 +9,15 @@ if [ "${CLICKHOUSE_DO_NOT_CHOWN:-0}" = "1" ]; then DO_CHOWN=0 fi -CLICKHOUSE_UID="${CLICKHOUSE_UID:-"$(id -u clickhouse)"}" -CLICKHOUSE_GID="${CLICKHOUSE_GID:-"$(id -g clickhouse)"}" - -# support --user +# support `docker run --user=xxx:xxxx` if [ "$(id -u)" = "0" ]; then - USER=$CLICKHOUSE_UID - GROUP=$CLICKHOUSE_GID + # CLICKHOUSE_UID and CLICKHOUSE_GID are kept for backward compatibility + if [[ "${CLICKHOUSE_UID:-}" || "${CLICKHOUSE_GID:-}" ]]; then + echo 'WARNING: consider using a proper "--user=xxx:xxxx" running argument instead of CLICKHOUSE_UID/CLICKHOUSE_GID' >&2 + echo 'Support for CLICKHOUSE_UID/CLICKHOUSE_GID will be removed in a couple of releases' >&2 + fi + USER="${CLICKHOUSE_UID:-"$(id -u clickhouse)"}" + GROUP="${CLICKHOUSE_GID:-"$(id -g clickhouse)"}" if command -v gosu &> /dev/null; then gosu="gosu $USER:$GROUP" elif command -v su-exec &> /dev/null; then diff --git a/docker/server/entrypoint.sh b/docker/server/entrypoint.sh index 3102ab8297c..a60643c63f1 100755 --- a/docker/server/entrypoint.sh +++ b/docker/server/entrypoint.sh @@ -8,13 +8,15 @@ if [ "${CLICKHOUSE_DO_NOT_CHOWN:-0}" = "1" ]; then DO_CHOWN=0 fi -CLICKHOUSE_UID="${CLICKHOUSE_UID:-"$(id -u clickhouse)"}" -CLICKHOUSE_GID="${CLICKHOUSE_GID:-"$(id -g clickhouse)"}" - -# support --user +# support `docker run --user=xxx:xxxx` if [ "$(id -u)" = "0" ]; then - USER=$CLICKHOUSE_UID - GROUP=$CLICKHOUSE_GID + # CLICKHOUSE_UID and CLICKHOUSE_GID are kept for backward compatibility + if [[ "${CLICKHOUSE_UID:-}" || "${CLICKHOUSE_GID:-}" ]]; then + echo 'WARNING: consider using a proper "--user=xxx:xxxx" running argument instead of CLICKHOUSE_UID/CLICKHOUSE_GID' >&2 + echo 'Support for CLICKHOUSE_UID/CLICKHOUSE_GID will be removed in a couple of releases' >&2 + fi + USER="${CLICKHOUSE_UID:-"$(id -u clickhouse)"}" + GROUP="${CLICKHOUSE_GID:-"$(id -g clickhouse)"}" else USER="$(id -u)" GROUP="$(id -g)" From b82658a28524f47356ab63a3c367489e10c83791 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Thu, 7 Nov 2024 12:16:19 +0100 Subject: [PATCH 503/680] Remove processing of CLICKHOUSE_DOCKER_RESTART_ON_EXIT --- docker/server/entrypoint.sh | 14 ++------------ 1 file changed, 2 insertions(+), 12 deletions(-) diff --git a/docker/server/entrypoint.sh b/docker/server/entrypoint.sh index a60643c63f1..6aa031b1352 100755 --- a/docker/server/entrypoint.sh +++ b/docker/server/entrypoint.sh @@ -205,18 +205,8 @@ if [[ $# -lt 1 ]] || [[ "$1" == "--"* ]]; then CLICKHOUSE_WATCHDOG_ENABLE=${CLICKHOUSE_WATCHDOG_ENABLE:-0} export CLICKHOUSE_WATCHDOG_ENABLE - # An option for easy restarting and replacing clickhouse-server in a container, especially in Kubernetes. - # For example, you can replace the clickhouse-server binary to another and restart it while keeping the container running. - if [[ "${CLICKHOUSE_DOCKER_RESTART_ON_EXIT:-0}" -eq "1" ]]; then - while true; do - # This runs the server as a child process of the shell script: - /usr/bin/clickhouse su "${USER}:${GROUP}" /usr/bin/clickhouse-server --config-file="$CLICKHOUSE_CONFIG" "$@" ||: - echo >&2 'ClickHouse Server exited, and the environment variable CLICKHOUSE_DOCKER_RESTART_ON_EXIT is set to 1. Restarting the server.' - done - else - # This replaces the shell script with the server: - exec /usr/bin/clickhouse su "${USER}:${GROUP}" /usr/bin/clickhouse-server --config-file="$CLICKHOUSE_CONFIG" "$@" - fi + # This replaces the shell script with the server: + exec /usr/bin/clickhouse su "${USER}:${GROUP}" /usr/bin/clickhouse-server --config-file="$CLICKHOUSE_CONFIG" "$@" fi # Otherwise, we assume the user want to run his own process, for example a `bash` shell to explore this image From ae97149041d2c489617f242ce2c96648c98ae620 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Thu, 7 Nov 2024 12:18:11 +0100 Subject: [PATCH 504/680] Remove `/usr/bin` for clickhouse/clickhouse-server/clickhouse-keeper --- docker/keeper/entrypoint.sh | 4 ++-- docker/server/entrypoint.sh | 6 +++--- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/docker/keeper/entrypoint.sh b/docker/keeper/entrypoint.sh index c5d5d26ec11..92b91a0f8c3 100644 --- a/docker/keeper/entrypoint.sh +++ b/docker/keeper/entrypoint.sh @@ -84,11 +84,11 @@ if [[ $# -lt 1 ]] || [[ "$1" == "--"* ]]; then # There is a config file. It is already tested with gosu (if it is readably by keeper user) if [ -f "$KEEPER_CONFIG" ]; then - exec $gosu /usr/bin/clickhouse-keeper --config-file="$KEEPER_CONFIG" "$@" + exec $gosu clickhouse-keeper --config-file="$KEEPER_CONFIG" "$@" fi # There is no config file. Will use embedded one - exec $gosu /usr/bin/clickhouse-keeper --log-file="$LOG_PATH" --errorlog-file="$ERROR_LOG_PATH" "$@" + exec $gosu clickhouse-keeper --log-file="$LOG_PATH" --errorlog-file="$ERROR_LOG_PATH" "$@" fi # Otherwise, we assume the user want to run his own process, for example a `bash` shell to explore this image diff --git a/docker/server/entrypoint.sh b/docker/server/entrypoint.sh index 6aa031b1352..7a990e7d889 100755 --- a/docker/server/entrypoint.sh +++ b/docker/server/entrypoint.sh @@ -62,7 +62,7 @@ function create_directory_and_do_chown() { # if DO_CHOWN=0 it means that the system does not map root user to "admin" permissions # it mainly happens on NFS mounts where root==nobody for security reasons # thus mkdir MUST run with user id/gid and not from nobody that has zero permissions - mkdir="/usr/bin/clickhouse su "${USER}:${GROUP}" mkdir" + mkdir="clickhouse su ""${USER}:${GROUP}"" mkdir" fi if ! $mkdir -p "$dir"; then echo "Couldn't create necessary directory: $dir" @@ -145,7 +145,7 @@ if [ -n "${RUN_INITDB_SCRIPTS}" ]; then fi # Listen only on localhost until the initialization is done - /usr/bin/clickhouse su "${USER}:${GROUP}" /usr/bin/clickhouse-server --config-file="$CLICKHOUSE_CONFIG" -- --listen_host=127.0.0.1 & + clickhouse su "${USER}:${GROUP}" clickhouse-server --config-file="$CLICKHOUSE_CONFIG" -- --listen_host=127.0.0.1 & pid="$!" # check if clickhouse is ready to accept connections @@ -206,7 +206,7 @@ if [[ $# -lt 1 ]] || [[ "$1" == "--"* ]]; then export CLICKHOUSE_WATCHDOG_ENABLE # This replaces the shell script with the server: - exec /usr/bin/clickhouse su "${USER}:${GROUP}" /usr/bin/clickhouse-server --config-file="$CLICKHOUSE_CONFIG" "$@" + exec clickhouse su "${USER}:${GROUP}" clickhouse-server --config-file="$CLICKHOUSE_CONFIG" "$@" fi # Otherwise, we assume the user want to run his own process, for example a `bash` shell to explore this image From 1babb919c3450969b2ecc810705854e702458110 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Thu, 7 Nov 2024 12:18:57 +0100 Subject: [PATCH 505/680] Follow the DOI review recommendations/requirements --- docker/server/Dockerfile.ubuntu | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/docker/server/Dockerfile.ubuntu b/docker/server/Dockerfile.ubuntu index 0d5c983f5e6..2b023a9cf03 100644 --- a/docker/server/Dockerfile.ubuntu +++ b/docker/server/Dockerfile.ubuntu @@ -91,7 +91,6 @@ RUN if [ -n "${single_binary_location_url}" ]; then \ RUN if ! clickhouse local -q "SELECT ''" > /dev/null 2>&1; then \ apt-get update \ && apt-get install --yes --no-install-recommends \ - apt-transport-https \ dirmngr \ gnupg2 \ && mkdir -p /etc/apt/sources.list.d \ @@ -108,13 +107,12 @@ RUN if ! clickhouse local -q "SELECT ''" > /dev/null 2>&1; then \ && for package in ${PACKAGES}; do \ packages="${packages} ${package}=${VERSION}" \ ; done \ - && apt-get install --allow-unauthenticated --yes --no-install-recommends ${packages} || exit 1 \ + && apt-get install --yes --no-install-recommends ${packages} || exit 1 \ && rm -rf \ /var/lib/apt/lists/* \ /var/cache/debconf \ /tmp/* \ - && apt-get autoremove --purge -yq libksba8 \ - && apt-get autoremove -yq \ + && apt-get autoremove --purge -yq dirmngr gnupg2 \ ; fi # post install @@ -126,8 +124,6 @@ RUN clickhouse-local -q 'SELECT * FROM system.build_options' \ RUN locale-gen en_US.UTF-8 ENV LANG en_US.UTF-8 -ENV LANGUAGE en_US:en -ENV LC_ALL en_US.UTF-8 ENV TZ UTC RUN mkdir /docker-entrypoint-initdb.d From 552b0fc8d0f106db1a85805ab883debe7e491e9c Mon Sep 17 00:00:00 2001 From: kssenii Date: Thu, 7 Nov 2024 13:11:33 +0100 Subject: [PATCH 506/680] Rename a setting --- src/Core/Settings.cpp | 3 ++- src/IO/ReadSettings.h | 2 +- src/Interpreters/Cache/QueryLimit.cpp | 2 +- src/Interpreters/Context.cpp | 4 ++-- 4 files changed, 6 insertions(+), 5 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index c2ffc2ddf0e..d9668849fd2 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -4852,7 +4852,7 @@ Allows to record the filesystem caching log for each query DECLARE(Bool, read_from_filesystem_cache_if_exists_otherwise_bypass_cache, false, R"( Allow to use the filesystem cache in passive mode - benefit from the existing cache entries, but don't put more entries into the cache. If you set this setting for heavy ad-hoc queries and leave it disabled for short real-time queries, this will allows to avoid cache threshing by too heavy queries and to improve the overall system efficiency. )", 0) \ - DECLARE(Bool, skip_download_if_exceeds_query_cache, true, R"( + DECLARE(Bool, filesystem_cache_skip_download_if_exceeds_per_query_cache_write_limit, true, R"( Skip download from remote filesystem if exceeds query cache size )", 0) \ DECLARE(UInt64, filesystem_cache_max_download_size, (128UL * 1024 * 1024 * 1024), R"( @@ -5887,6 +5887,7 @@ Experimental data deduplication for SELECT queries based on part UUIDs MAKE_OBSOLETE(M, Bool, use_mysql_types_in_show_columns, false) \ MAKE_OBSOLETE(M, Bool, s3queue_allow_experimental_sharded_mode, false) \ MAKE_OBSOLETE(M, LightweightMutationProjectionMode, lightweight_mutation_projection_mode, LightweightMutationProjectionMode::THROW) \ + MAKE_OBSOLETE(M, Bool, skip_download_if_exceeds_query_cache, true) \ /* moved to config.xml: see also src/Core/ServerSettings.h */ \ MAKE_DEPRECATED_BY_SERVER_CONFIG(M, UInt64, background_buffer_flush_schedule_pool_size, 16) \ MAKE_DEPRECATED_BY_SERVER_CONFIG(M, UInt64, background_pool_size, 16) \ diff --git a/src/IO/ReadSettings.h b/src/IO/ReadSettings.h index 6ed02212095..103ce7df54b 100644 --- a/src/IO/ReadSettings.h +++ b/src/IO/ReadSettings.h @@ -68,7 +68,7 @@ struct ReadSettings std::shared_ptr page_cache; size_t filesystem_cache_max_download_size = (128UL * 1024 * 1024 * 1024); - bool skip_download_if_exceeds_query_cache = true; + bool filesystem_cache_skip_download_if_exceeds_per_query_cache_write_limit = true; size_t remote_read_min_bytes_for_seek = DBMS_DEFAULT_BUFFER_SIZE; diff --git a/src/Interpreters/Cache/QueryLimit.cpp b/src/Interpreters/Cache/QueryLimit.cpp index b18d23a5b7f..a7c964022a5 100644 --- a/src/Interpreters/Cache/QueryLimit.cpp +++ b/src/Interpreters/Cache/QueryLimit.cpp @@ -53,7 +53,7 @@ FileCacheQueryLimit::QueryContextPtr FileCacheQueryLimit::getOrSetQueryContext( { it->second = std::make_shared( settings.filesystem_cache_max_download_size, - !settings.skip_download_if_exceeds_query_cache); + !settings.filesystem_cache_skip_download_if_exceeds_per_query_cache_write_limit); } return it->second; diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp index c1fa2c8549a..7b7cdfa2104 100644 --- a/src/Interpreters/Context.cpp +++ b/src/Interpreters/Context.cpp @@ -236,7 +236,7 @@ namespace Setting extern const SettingsUInt64 remote_fs_read_backoff_max_tries; extern const SettingsUInt64 remote_read_min_bytes_for_seek; extern const SettingsBool throw_on_error_from_cache_on_write_operations; - extern const SettingsBool skip_download_if_exceeds_query_cache; + extern const SettingsBool filesystem_cache_skip_download_if_exceeds_per_query_cache_write_limit; extern const SettingsBool s3_allow_parallel_part_upload; extern const SettingsBool use_page_cache_for_disks_without_file_cache; extern const SettingsUInt64 use_structure_from_insertion_table_in_table_functions; @@ -5753,7 +5753,7 @@ ReadSettings Context::getReadSettings() const res.filesystem_cache_allow_background_download_during_fetch = settings_ref[Setting::filesystem_cache_enable_background_download_during_fetch]; res.filesystem_cache_max_download_size = settings_ref[Setting::filesystem_cache_max_download_size]; - res.skip_download_if_exceeds_query_cache = settings_ref[Setting::skip_download_if_exceeds_query_cache]; + res.filesystem_cache_skip_download_if_exceeds_per_query_cache_write_limit = settings_ref[Setting::filesystem_cache_skip_download_if_exceeds_per_query_cache_write_limit]; res.page_cache = getPageCache(); res.use_page_cache_for_disks_without_file_cache = settings_ref[Setting::use_page_cache_for_disks_without_file_cache]; From d8ff6f868fe6cb346ac751b468b462b857399480 Mon Sep 17 00:00:00 2001 From: Pablo Marcos Date: Thu, 7 Nov 2024 12:36:21 +0000 Subject: [PATCH 507/680] bitShift: return 0 instead of throwing an exception if overflow --- src/Functions/bitShiftLeft.cpp | 20 +++++++++++-------- src/Functions/bitShiftRight.cpp | 20 +++++++++++-------- .../02766_bitshift_with_const_arguments.sql | 2 +- ...t_throws_error_for_out_of_bounds.reference | 6 ++++++ ...t_shift_throws_error_for_out_of_bounds.sql | 12 +++++------ 5 files changed, 37 insertions(+), 23 deletions(-) diff --git a/src/Functions/bitShiftLeft.cpp b/src/Functions/bitShiftLeft.cpp index 0eb0d82ef0f..7fd0f7cf631 100644 --- a/src/Functions/bitShiftLeft.cpp +++ b/src/Functions/bitShiftLeft.cpp @@ -25,8 +25,10 @@ struct BitShiftLeftImpl { if constexpr (is_big_int_v) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "BitShiftLeft is not implemented for big integers as second argument"); - else if (b < 0 || static_cast(b) > 8 * sizeof(A)) - throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); + else if (b < 0) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value"); + else if (static_cast(b) > 8 * sizeof(A)) + return static_cast(0); else if constexpr (is_big_int_v) return static_cast(a) << static_cast(b); else @@ -43,9 +45,10 @@ struct BitShiftLeftImpl const UInt8 word_size = 8 * sizeof(*pos); size_t n = end - pos; const UInt128 bit_limit = static_cast(word_size) * n; - if (b < 0 || static_cast(b) > bit_limit) - throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); - if (b == bit_limit) + if (b < 0) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value"); + + if (b == bit_limit || static_cast(b) > bit_limit) { // insert default value out_vec.push_back(0); @@ -111,9 +114,10 @@ struct BitShiftLeftImpl const UInt8 word_size = 8; size_t n = end - pos; const UInt128 bit_limit = static_cast(word_size) * n; - if (b < 0 || static_cast(b) > bit_limit) - throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); - if (b == bit_limit) + if (b < 0) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value"); + + if (b == bit_limit || static_cast(b) > bit_limit) { // insert default value out_vec.resize_fill(out_vec.size() + n); diff --git a/src/Functions/bitShiftRight.cpp b/src/Functions/bitShiftRight.cpp index 16032b32f68..19ea7b8c751 100644 --- a/src/Functions/bitShiftRight.cpp +++ b/src/Functions/bitShiftRight.cpp @@ -26,8 +26,10 @@ struct BitShiftRightImpl { if constexpr (is_big_int_v) throw Exception(ErrorCodes::NOT_IMPLEMENTED, "BitShiftRight is not implemented for big integers as second argument"); - else if (b < 0 || static_cast(b) > 8 * sizeof(A)) - throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); + else if (b < 0) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value"); + else if (static_cast(b) > 8 * sizeof(A)) + return static_cast(0); else if constexpr (is_big_int_v) return static_cast(a) >> static_cast(b); else @@ -59,9 +61,10 @@ struct BitShiftRightImpl const UInt8 word_size = 8; size_t n = end - pos; const UInt128 bit_limit = static_cast(word_size) * n; - if (b < 0 || static_cast(b) > bit_limit) - throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); - if (b == bit_limit) + if (b < 0) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value"); + + if (b == bit_limit || static_cast(b) > bit_limit) { /// insert default value out_vec.push_back(0); @@ -99,9 +102,10 @@ struct BitShiftRightImpl const UInt8 word_size = 8; size_t n = end - pos; const UInt128 bit_limit = static_cast(word_size) * n; - if (b < 0 || static_cast(b) > bit_limit) - throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value and less or equal to the bit width of the value to shift"); - if (b == bit_limit) + if (b < 0) + throw Exception(ErrorCodes::ARGUMENT_OUT_OF_BOUND, "The number of shift positions needs to be a non-negative value"); + + if (b == bit_limit || static_cast(b) > bit_limit) { // insert default value out_vec.resize_fill(out_vec.size() + n); diff --git a/tests/queries/0_stateless/02766_bitshift_with_const_arguments.sql b/tests/queries/0_stateless/02766_bitshift_with_const_arguments.sql index 91e8624057c..6b2961f0555 100644 --- a/tests/queries/0_stateless/02766_bitshift_with_const_arguments.sql +++ b/tests/queries/0_stateless/02766_bitshift_with_const_arguments.sql @@ -10,7 +10,7 @@ DROP TABLE IF EXISTS t1; CREATE TABLE t0 (vkey UInt32, pkey UInt32, c0 UInt32) engine = TinyLog; CREATE TABLE t1 (vkey UInt32) ENGINE = AggregatingMergeTree ORDER BY vkey; INSERT INTO t0 VALUES (15, 25000, 58); -SELECT ref_5.pkey AS c_2_c2392_6 FROM t0 AS ref_5 WHERE 'J[' < multiIf(ref_5.pkey IN ( SELECT 1 ), bitShiftLeft(multiIf(ref_5.c0 > NULL, '1', ')'), 40), NULL); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT ref_5.pkey AS c_2_c2392_6 FROM t0 AS ref_5 WHERE 'J[' < multiIf(ref_5.pkey IN ( SELECT 1 ), bitShiftLeft(multiIf(ref_5.c0 > NULL, '1', ')'), 40), NULL); DROP TABLE t0; DROP TABLE t1; diff --git a/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.reference b/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.reference index 33b8cd6ee26..1fda82a9747 100644 --- a/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.reference +++ b/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.reference @@ -1,3 +1,9 @@ -- bitShiftRight +0 + +\0\0\0\0\0\0\0\0 -- bitShiftLeft +0 + +\0\0\0\0\0\0\0\0 OK diff --git a/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.sql b/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.sql index aec01753673..340cc1292e4 100644 --- a/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.sql +++ b/tests/queries/0_stateless/03198_bit_shift_throws_error_for_out_of_bounds.sql @@ -1,17 +1,17 @@ SELECT '-- bitShiftRight'; SELECT bitShiftRight(1, -1); -- { serverError ARGUMENT_OUT_OF_BOUND } -SELECT bitShiftRight(toUInt8(1), 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftRight(toUInt8(1), 8 + 1); SELECT bitShiftRight('hola', -1); -- { serverError ARGUMENT_OUT_OF_BOUND } -SELECT bitShiftRight('hola', 4 * 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftRight('hola', 4 * 8 + 1); SELECT bitShiftRight(toFixedString('hola', 8), -1); -- { serverError ARGUMENT_OUT_OF_BOUND } -SELECT bitShiftRight(toFixedString('hola', 8), 8 * 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftRight(toFixedString('hola', 8), 8 * 8 + 1); SELECT '-- bitShiftLeft'; SELECT bitShiftLeft(1, -1); -- { serverError ARGUMENT_OUT_OF_BOUND } -SELECT bitShiftLeft(toUInt8(1), 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftLeft(toUInt8(1), 8 + 1); SELECT bitShiftLeft('hola', -1); -- { serverError ARGUMENT_OUT_OF_BOUND } -SELECT bitShiftLeft('hola', 4 * 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftLeft('hola', 4 * 8 + 1); SELECT bitShiftLeft(toFixedString('hola', 8), -1); -- { serverError ARGUMENT_OUT_OF_BOUND } -SELECT bitShiftLeft(toFixedString('hola', 8), 8 * 8 + 1); -- { serverError ARGUMENT_OUT_OF_BOUND } +SELECT bitShiftLeft(toFixedString('hola', 8), 8 * 8 + 1); SELECT 'OK'; \ No newline at end of file From f727a3931bfa0d7b3945bfb8703665aef3fc0695 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Thu, 7 Nov 2024 12:41:48 +0000 Subject: [PATCH 508/680] Clarify query cache docs and remove obsolete setting --- docs/en/operations/query-cache.md | 23 +++++++++++------------ src/Core/Settings.cpp | 1 - 2 files changed, 11 insertions(+), 13 deletions(-) diff --git a/docs/en/operations/query-cache.md b/docs/en/operations/query-cache.md index 955cec0234e..f0941aa28aa 100644 --- a/docs/en/operations/query-cache.md +++ b/docs/en/operations/query-cache.md @@ -25,9 +25,10 @@ Query caches can generally be viewed as transactionally consistent or inconsiste slowly enough that the database only needs to compute the report once (represented by the first `SELECT` query). Further queries can be served directly from the query cache. In this example, a reasonable validity period could be 30 min. -Transactionally inconsistent caching is traditionally provided by client tools or proxy packages interacting with the database. As a result, -the same caching logic and configuration is often duplicated. With ClickHouse's query cache, the caching logic moves to the server side. -This reduces maintenance effort and avoids redundancy. +Transactionally inconsistent caching is traditionally provided by client tools or proxy packages (e.g. +[chproxy](https://www.chproxy.org/configuration/caching/)) interacting with the database. As a result, the same caching logic and +configuration is often duplicated. With ClickHouse's query cache, the caching logic moves to the server side. This reduces maintenance +effort and avoids redundancy. ## Configuration Settings and Usage @@ -138,7 +139,10 @@ is only cached if the query runs longer than 5 seconds. It is also possible to s cached - for that use setting [query_cache_min_query_runs](settings/settings.md#query-cache-min-query-runs). Entries in the query cache become stale after a certain time period (time-to-live). By default, this period is 60 seconds but a different -value can be specified at session, profile or query level using setting [query_cache_ttl](settings/settings.md#query-cache-ttl). +value can be specified at session, profile or query level using setting [query_cache_ttl](settings/settings.md#query-cache-ttl). The query +cache evicts entries "lazily", i.e. when an entry becomes stale, it is not immediately removed from the cache. Instead, when a new entry +is to be inserted into the query cache, the database checks whether the cache has enough free space for the new entry. If this is not the +case, the database tries to remove all stale entries. If the cache still has not enough free space, the new entry is not inserted. Entries in the query cache are compressed by default. This reduces the overall memory consumption at the cost of slower writes into / reads from the query cache. To disable compression, use setting [query_cache_compress_entries](settings/settings.md#query-cache-compress-entries). @@ -188,14 +192,9 @@ Also, results of queries with non-deterministic functions are not cached by defa To force caching of results of queries with non-deterministic functions regardless, use setting [query_cache_nondeterministic_function_handling](settings/settings.md#query-cache-nondeterministic-function-handling). -Results of queries that involve system tables, e.g. `system.processes` or `information_schema.tables`, are not cached by default. To force -caching of results of queries with system tables regardless, use setting -[query_cache_system_table_handling](settings/settings.md#query-cache-system-table-handling). - -:::note -Prior to ClickHouse v23.11, setting 'query_cache_store_results_of_queries_with_nondeterministic_functions = 0 / 1' controlled whether -results of queries with non-deterministic results were cached. In newer ClickHouse versions, this setting is obsolete and has no effect. -::: +Results of queries that involve system tables (e.g. [system.processes](system-tables/processes.md)` or +[information_schema.tables](system-tables/information_schema.md)) are not cached by default. To force caching of results of queries with +system tables regardless, use setting [query_cache_system_table_handling](settings/settings.md#query-cache-system-table-handling). Finally, entries in the query cache are not shared between users due to security reasons. For example, user A must not be able to bypass a row policy on a table by running the same query as another user B for whom no such policy exists. However, if necessary, cache entries can diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index c2ffc2ddf0e..3bfa58e4f98 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -5916,7 +5916,6 @@ Experimental data deduplication for SELECT queries based on part UUIDs MAKE_OBSOLETE(M, UInt64, parallel_replicas_min_number_of_granules_to_enable, 0) \ MAKE_OBSOLETE(M, ParallelReplicasCustomKeyFilterType, parallel_replicas_custom_key_filter_type, ParallelReplicasCustomKeyFilterType::DEFAULT) \ MAKE_OBSOLETE(M, Bool, query_plan_optimize_projection, true) \ - MAKE_OBSOLETE(M, Bool, query_cache_store_results_of_queries_with_nondeterministic_functions, false) \ MAKE_OBSOLETE(M, Bool, allow_experimental_annoy_index, false) \ MAKE_OBSOLETE(M, UInt64, max_threads_for_annoy_index_creation, 4) \ MAKE_OBSOLETE(M, Int64, annoy_index_search_k_nodes, -1) \ From ca23e5254c2cca5e6b3f4a9c7ccd65f70be42fc4 Mon Sep 17 00:00:00 2001 From: avogar Date: Thu, 7 Nov 2024 12:44:57 +0000 Subject: [PATCH 509/680] Fix for tmp parts --- src/Storages/MergeTree/IMergeTreeDataPart.cpp | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp index 20d7528d38a..fb934a77512 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp @@ -2501,7 +2501,7 @@ ColumnPtr IMergeTreeDataPart::getColumnSample(const NameAndTypePair & column) co { const size_t total_mark = getMarksCount(); /// If column doesn't have dynamic subcolumns or part has no data, just create column using it's type. - if (!column.type->hasDynamicSubcolumns() || !total_mark) + if (is_temp || !column.type->hasDynamicSubcolumns() || !total_mark) return column.type->createColumn(); /// Otherwise, read sample column with 0 rows from the part, so it will load dynamic structure. @@ -2510,22 +2510,24 @@ ColumnPtr IMergeTreeDataPart::getColumnSample(const NameAndTypePair & column) co StorageMetadataPtr metadata_ptr = storage.getInMemoryMetadataPtr(); StorageSnapshotPtr storage_snapshot_ptr = std::make_shared(storage, metadata_ptr); + MergeTreeReaderSettings settings; + settings.can_read_part_without_marks = true; MergeTreeReaderPtr reader = getReader( cols, storage_snapshot_ptr, - MarkRanges{MarkRange(0, 1)}, + MarkRanges{MarkRange(0, total_mark)}, /*virtual_fields=*/ {}, /*uncompressed_cache=*/{}, storage.getContext()->getMarkCache().get(), std::make_shared(), - MergeTreeReaderSettings{}, + settings, ValueSizeMap{}, ReadBufferFromFileBase::ProfileCallback{}); Columns result; result.resize(1); - reader->readRows(0, 1, false, 0, result); + reader->readRows(0, total_mark, false, 0, result); return result[0]; } From d43329f254eaaddaece94d4f96631b3307be23bb Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Tue, 5 Nov 2024 13:31:10 +0100 Subject: [PATCH 510/680] UX: slightly improve cache await interface --- tests/ci/ci_cache.py | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/tests/ci/ci_cache.py b/tests/ci/ci_cache.py index 6f2e3e70736..5ebed827926 100644 --- a/tests/ci/ci_cache.py +++ b/tests/ci/ci_cache.py @@ -795,11 +795,12 @@ class CiCache: # start waiting for the next TIMEOUT seconds if there are more than X(=4) jobs to wait # wait TIMEOUT seconds in rounds. Y(=5) is the max number of rounds expired_sec = 0 - start_at = int(time.time()) + start_at = time.time() while expired_sec < TIMEOUT and self.jobs_to_wait: await_finished: Set[str] = set() if not dry_run: - time.sleep(poll_interval_sec) + # Do not sleep longer than required + time.sleep(min(poll_interval_sec, TIMEOUT - expired_sec)) self.update() for job_name, job_config in self.jobs_to_wait.items(): num_batches = job_config.num_batches @@ -844,7 +845,8 @@ class CiCache: del self.jobs_to_wait[job] if not dry_run: - expired_sec = int(time.time()) - start_at + # Avoid `seconds left [-3]` + expired_sec = min(int(time.time() - start_at), TIMEOUT) print( f"...awaiting continues... seconds left [{TIMEOUT - expired_sec}]" ) From ccaa66963dfa937f6a2562ff22d9b90254fefea3 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Tue, 5 Nov 2024 13:37:35 +0100 Subject: [PATCH 511/680] Print a proper message for finished awaiting --- tests/ci/ci_cache.py | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/tests/ci/ci_cache.py b/tests/ci/ci_cache.py index 5ebed827926..c271339db8b 100644 --- a/tests/ci/ci_cache.py +++ b/tests/ci/ci_cache.py @@ -845,11 +845,12 @@ class CiCache: del self.jobs_to_wait[job] if not dry_run: - # Avoid `seconds left [-3]` - expired_sec = min(int(time.time() - start_at), TIMEOUT) - print( - f"...awaiting continues... seconds left [{TIMEOUT - expired_sec}]" - ) + expired_sec = int(time.time() - start_at) + msg = f"...awaiting continues... seconds left [{TIMEOUT - expired_sec}]" + if expired_sec >= TIMEOUT: + # Avoid `seconds left [-3]` + msg = f"awaiting for round {round_cnt} is finished" + print(msg) else: # make up for 2 iterations in dry_run expired_sec += int(TIMEOUT / 2) + 1 From 07b480c1e4e1f1fd647c4c9cf7d00e29b5619868 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Thu, 7 Nov 2024 14:44:31 +0100 Subject: [PATCH 512/680] Implicitly treat a file argument as --queries-file --- programs/main.cpp | 40 +++++++++---------- src/Client/ClientBaseOptimizedParts.cpp | 8 +++- ...al_arguments_implicit_query_file.reference | 11 +++++ ...ositional_arguments_implicit_query_file.sh | 34 ++++++++++++++++ 4 files changed, 71 insertions(+), 22 deletions(-) create mode 100644 tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.reference create mode 100755 tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.sh diff --git a/programs/main.cpp b/programs/main.cpp index 02ea1471108..ea8fbc1aece 100644 --- a/programs/main.cpp +++ b/programs/main.cpp @@ -1,27 +1,22 @@ -#include -#include +#include +#include +#include +#include -#include -#include -#include -#include -#include -#include /// pair - -#include +#if defined(SANITIZE_COVERAGE) +# include +#endif #include "config.h" #include "config_tools.h" -#include -#include -#include -#include -#include - -#include -#include - +#include +#include +#include +#include +#include +#include /// pair +#include /// Universal executable for various clickhouse applications int mainEntryClickHouseServer(int argc, char ** argv); @@ -238,9 +233,12 @@ int main(int argc_, char ** argv_) /// clickhouse # spawn local /// clickhouse local # spawn local /// clickhouse "select ..." # spawn local + /// clickhouse /tmp/repro --enable-analyzer /// - if (main_func == printHelp && !argv.empty() && (argv.size() == 1 || argv[1][0] == '-' - || std::string_view(argv[1]).contains(' '))) + std::error_code ec; + if (main_func == printHelp && !argv.empty() + && (argv.size() == 1 || argv[1][0] == '-' || std::string_view(argv[1]).contains(' ') + || std::filesystem::exists(std::filesystem::path{argv[1]}, ec))) { main_func = mainEntryClickHouseLocal; } diff --git a/src/Client/ClientBaseOptimizedParts.cpp b/src/Client/ClientBaseOptimizedParts.cpp index ac4d3417779..bc362288079 100644 --- a/src/Client/ClientBaseOptimizedParts.cpp +++ b/src/Client/ClientBaseOptimizedParts.cpp @@ -1,5 +1,7 @@ #include +#include + namespace DB { @@ -107,6 +109,7 @@ void ClientApplicationBase::parseAndCheckOptions(OptionsDescription & options_de && !op.original_tokens[0].empty() && !op.value.empty()) { /// Two special cases for better usability: + /// - if the option is a filesystem file, then it's likely a queries file (clickhouse repro.sql) /// - if the option contains a whitespace, it might be a query: clickhouse "SELECT 1" /// These are relevant for interactive usage - user-friendly, but questionable in general. /// In case of ambiguity or for scripts, prefer using proper options. @@ -115,7 +118,10 @@ void ClientApplicationBase::parseAndCheckOptions(OptionsDescription & options_de po::variable_value value(boost::any(op.value), false); const char * option; - if (token.contains(' ')) + std::error_code ec; + if (std::filesystem::exists(std::filesystem::path{token}, ec)) + option = "queries-file"; + else if (token.contains(' ')) option = "query"; else throw Exception(ErrorCodes::BAD_ARGUMENTS, "Positional option `{}` is not supported.", token); diff --git a/tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.reference b/tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.reference new file mode 100644 index 00000000000..fe2432a063f --- /dev/null +++ b/tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.reference @@ -0,0 +1,11 @@ +Hello from a file +Hello from a file +Hello from a file +Hello from a file +Hello from a file +Hello from a file +Hello from a file +Hello from a file +Hello from a file +max_local_read_bandwidth 1 100 +max_local_read_bandwidth 1 200 diff --git a/tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.sh b/tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.sh new file mode 100755 index 00000000000..14b6e735a9a --- /dev/null +++ b/tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.sh @@ -0,0 +1,34 @@ +# Tags: no-random-settings + +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +FILE=${CLICKHOUSE_TMP}/${CLICKHOUSE_DATABASE}_without_extension +echo "SELECT 'Hello from a file'" > ${FILE} + +# Queries can be read from a file. +${CLICKHOUSE_BINARY} --queries-file ${FILE} + +# Or from stdin. +${CLICKHOUSE_BINARY} < ${FILE} + +# Also the positional argument can be interpreted as a file. +${CLICKHOUSE_BINARY} ${FILE} + +${CLICKHOUSE_LOCAL} --queries-file ${FILE} +${CLICKHOUSE_LOCAL} < ${FILE} +${CLICKHOUSE_LOCAL} ${FILE} + +${CLICKHOUSE_CLIENT} --queries-file ${FILE} +${CLICKHOUSE_CLIENT} < ${FILE} +${CLICKHOUSE_CLIENT} ${FILE} + +# Check that positional arguments work in any place +echo "Select name, changed, value FROM system.settings where name = 'max_local_read_bandwidth'" > ${FILE} +${CLICKHOUSE_BINARY} ${FILE} --max-local-read-bandwidth 100 +${CLICKHOUSE_BINARY} --max-local-read-bandwidth 200 ${FILE} + +rm ${FILE} From 06b580777e6ee8ef95cfa261b0a745ddda2662f7 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Thu, 7 Nov 2024 15:08:05 +0100 Subject: [PATCH 513/680] Style --- .../03267_positional_arguments_implicit_query_file.sh | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.sh b/tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.sh index 14b6e735a9a..791aa3af0db 100755 --- a/tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.sh +++ b/tests/queries/0_stateless/03267_positional_arguments_implicit_query_file.sh @@ -1,6 +1,5 @@ -# Tags: no-random-settings - #!/usr/bin/env bash +# Tags: no-random-settings CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh From 5cc42571f326ac409abdf612278042c84c4e3a74 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Thu, 7 Nov 2024 14:57:24 +0000 Subject: [PATCH 514/680] Revert obsolete settings removal --- src/Core/Settings.cpp | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 3bfa58e4f98..0d322f107de 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -5859,7 +5859,7 @@ Experimental data deduplication for SELECT queries based on part UUIDs // Please add settings related to formats in Core/FormatFactorySettings.h, move obsolete settings to OBSOLETE_SETTINGS and obsolete format settings to OBSOLETE_FORMAT_SETTINGS. #define OBSOLETE_SETTINGS(M, ALIAS) \ - /** Obsolete settings that do nothing but left for compatibility reasons. Remove each one after half a year of obsolescence. */ \ + /** Obsolete settings which are kept around for compatibility reasons. They have no effect anymore. */ \ MAKE_OBSOLETE(M, Bool, update_insert_deduplication_token_in_dependent_materialized_views, 0) \ MAKE_OBSOLETE(M, UInt64, max_memory_usage_for_all_queries, 0) \ MAKE_OBSOLETE(M, UInt64, multiple_joins_rewriter_version, 0) \ @@ -5916,6 +5916,7 @@ Experimental data deduplication for SELECT queries based on part UUIDs MAKE_OBSOLETE(M, UInt64, parallel_replicas_min_number_of_granules_to_enable, 0) \ MAKE_OBSOLETE(M, ParallelReplicasCustomKeyFilterType, parallel_replicas_custom_key_filter_type, ParallelReplicasCustomKeyFilterType::DEFAULT) \ MAKE_OBSOLETE(M, Bool, query_plan_optimize_projection, true) \ + MAKE_OBSOLETE(M, Bool, query_cache_store_results_of_queries_with_nondeterministic_functions, false) \ MAKE_OBSOLETE(M, Bool, allow_experimental_annoy_index, false) \ MAKE_OBSOLETE(M, UInt64, max_threads_for_annoy_index_creation, 4) \ MAKE_OBSOLETE(M, Int64, annoy_index_search_k_nodes, -1) \ From de03a5dae75b06520ab19a5fd34a561f83ae74e2 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Thu, 7 Nov 2024 15:04:53 +0000 Subject: [PATCH 515/680] Fix test which used an obsolete setting --- tests/queries/0_stateless/02494_query_cache_normalize_ast.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/02494_query_cache_normalize_ast.sql b/tests/queries/0_stateless/02494_query_cache_normalize_ast.sql index 1dbb3ef8158..cb53c4db7de 100644 --- a/tests/queries/0_stateless/02494_query_cache_normalize_ast.sql +++ b/tests/queries/0_stateless/02494_query_cache_normalize_ast.sql @@ -7,7 +7,7 @@ SYSTEM DROP QUERY CACHE; -- Run query whose result gets cached in the query cache. -- Besides "use_query_cache", pass two more knobs (one QC-specific knob and one non-QC-specific knob). We just care -- *that* they are passed and not about their effect. -SELECT 1 SETTINGS use_query_cache = true, query_cache_store_results_of_queries_with_nondeterministic_functions = true, max_threads = 16; +SELECT 1 SETTINGS use_query_cache = true, query_cache_nondeterministic_function_handling = 'save', max_threads = 16; -- Check that entry in QC exists SELECT COUNT(*) FROM system.query_cache; From a01c2e3f8c265aceb3042cdee1abafeed4f68485 Mon Sep 17 00:00:00 2001 From: Pervakov Grigorii Date: Thu, 7 Nov 2024 16:51:53 +0300 Subject: [PATCH 516/680] Keep materialized view security overriden context until end of query --- src/Processors/Sinks/SinkToStorage.h | 4 ++++ src/Storages/StorageMaterializedView.cpp | 2 ++ ...67_materialized_view_keeps_security_context.reference | 1 + .../03267_materialized_view_keeps_security_context.sql | 9 +++++++++ 4 files changed, 16 insertions(+) create mode 100644 tests/queries/0_stateless/03267_materialized_view_keeps_security_context.reference create mode 100644 tests/queries/0_stateless/03267_materialized_view_keeps_security_context.sql diff --git a/src/Processors/Sinks/SinkToStorage.h b/src/Processors/Sinks/SinkToStorage.h index c728fa87b1e..4bdcb2fe855 100644 --- a/src/Processors/Sinks/SinkToStorage.h +++ b/src/Processors/Sinks/SinkToStorage.h @@ -5,6 +5,8 @@ namespace DB { +class Context; + /// Sink which is returned from Storage::write. class SinkToStorage : public ExceptionKeepingTransform { @@ -16,12 +18,14 @@ public: const Block & getHeader() const { return inputs.front().getHeader(); } void addTableLock(const TableLockHolder & lock) { table_locks.push_back(lock); } + void addInterpreterContext(std::shared_ptr context) { interpreter_context.emplace_back(std::move(context)); } protected: virtual void consume(Chunk & chunk) = 0; private: std::vector table_locks; + std::vector> interpreter_context; void onConsume(Chunk chunk) override; GenerateResult onGenerate() override; diff --git a/src/Storages/StorageMaterializedView.cpp b/src/Storages/StorageMaterializedView.cpp index d047b28e076..3289ff1ae25 100644 --- a/src/Storages/StorageMaterializedView.cpp +++ b/src/Storages/StorageMaterializedView.cpp @@ -382,6 +382,7 @@ void StorageMaterializedView::read( } query_plan.addStorageHolder(storage); + query_plan.addInterpreterContext(context); query_plan.addTableLock(std::move(lock)); } } @@ -405,6 +406,7 @@ SinkToStoragePtr StorageMaterializedView::write(const ASTPtr & query, const Stor auto sink = storage->write(query, metadata_snapshot, context, async_insert); + sink->addInterpreterContext(context); sink->addTableLock(lock); return sink; } diff --git a/tests/queries/0_stateless/03267_materialized_view_keeps_security_context.reference b/tests/queries/0_stateless/03267_materialized_view_keeps_security_context.reference new file mode 100644 index 00000000000..d00491fd7e5 --- /dev/null +++ b/tests/queries/0_stateless/03267_materialized_view_keeps_security_context.reference @@ -0,0 +1 @@ +1 diff --git a/tests/queries/0_stateless/03267_materialized_view_keeps_security_context.sql b/tests/queries/0_stateless/03267_materialized_view_keeps_security_context.sql new file mode 100644 index 00000000000..bb44e4920af --- /dev/null +++ b/tests/queries/0_stateless/03267_materialized_view_keeps_security_context.sql @@ -0,0 +1,9 @@ +DROP TABLE IF EXISTS {CLICKHOUSE_DATABASE:Identifier}.rview; +DROP TABLE IF EXISTS {CLICKHOUSE_DATABASE:Identifier}.wview; + +-- Read from view +CREATE MATERIALIZED VIEW rview ENGINE = File(CSV) POPULATE AS SELECT 1 AS c0; +SELECT 1 FROM rview; + +-- Write through view populate +CREATE MATERIALIZED VIEW wview ENGINE = Join(ALL, INNER, c0) POPULATE AS SELECT 1 AS c0; From 96b59a2ef679b6b23ffcecafd59c05a0ea784ada Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Thu, 7 Nov 2024 13:43:58 +0100 Subject: [PATCH 517/680] Avoid port clash in CoordinationTest/0.TestSummingRaft1 --- src/Coordination/tests/gtest_coordination.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Coordination/tests/gtest_coordination.cpp b/src/Coordination/tests/gtest_coordination.cpp index 9648fdd4530..c56e698766a 100644 --- a/src/Coordination/tests/gtest_coordination.cpp +++ b/src/Coordination/tests/gtest_coordination.cpp @@ -330,7 +330,7 @@ TYPED_TEST(CoordinationTest, TestSummingRaft1) this->setLogDirectory("./logs"); this->setStateFileDirectory("."); - SummingRaftServer s1(1, "localhost", 44444, this->keeper_context); + SummingRaftServer s1(1, "localhost", 0, this->keeper_context); SCOPE_EXIT(if (std::filesystem::exists("./state")) std::filesystem::remove("./state");); /// Single node is leader From e5fc37bc7e6c707cd7ea14bb3c4888f94118a126 Mon Sep 17 00:00:00 2001 From: kssenii Date: Thu, 7 Nov 2024 17:27:51 +0100 Subject: [PATCH 518/680] Add alias --- src/Core/Settings.cpp | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index d9668849fd2..328f950da1d 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -4854,7 +4854,7 @@ Allow to use the filesystem cache in passive mode - benefit from the existing ca )", 0) \ DECLARE(Bool, filesystem_cache_skip_download_if_exceeds_per_query_cache_write_limit, true, R"( Skip download from remote filesystem if exceeds query cache size -)", 0) \ +)", 0) ALIAS(skip_download_if_exceeds_query_cache) \ DECLARE(UInt64, filesystem_cache_max_download_size, (128UL * 1024 * 1024 * 1024), R"( Max remote filesystem cache size that can be downloaded by a single query )", 0) \ @@ -5887,7 +5887,6 @@ Experimental data deduplication for SELECT queries based on part UUIDs MAKE_OBSOLETE(M, Bool, use_mysql_types_in_show_columns, false) \ MAKE_OBSOLETE(M, Bool, s3queue_allow_experimental_sharded_mode, false) \ MAKE_OBSOLETE(M, LightweightMutationProjectionMode, lightweight_mutation_projection_mode, LightweightMutationProjectionMode::THROW) \ - MAKE_OBSOLETE(M, Bool, skip_download_if_exceeds_query_cache, true) \ /* moved to config.xml: see also src/Core/ServerSettings.h */ \ MAKE_DEPRECATED_BY_SERVER_CONFIG(M, UInt64, background_buffer_flush_schedule_pool_size, 16) \ MAKE_DEPRECATED_BY_SERVER_CONFIG(M, UInt64, background_pool_size, 16) \ From bfad05ac60b90bf7b4000cf6f87b54730ce108a5 Mon Sep 17 00:00:00 2001 From: alesapin Date: Thu, 7 Nov 2024 17:35:10 +0100 Subject: [PATCH 519/680] Shrink to fit index granularity array in memory to reduce memory footprint --- src/Storages/MergeTree/IMergeTreeDataPart.cpp | 2 ++ src/Storages/MergeTree/MergeTreeIndexGranularity.cpp | 6 ++++++ src/Storages/MergeTree/MergeTreeIndexGranularity.h | 2 ++ 3 files changed, 10 insertions(+) diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp index 41783ffddb0..7453d609fa9 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp @@ -735,7 +735,9 @@ void IMergeTreeDataPart::loadColumnsChecksumsIndexes(bool require_columns_checks loadUUID(); loadColumns(require_columns_checksums); loadChecksums(require_columns_checksums); + loadIndexGranularity(); + index_granularity.shrinkToFitInMemory(); if (!(*storage.getSettings())[MergeTreeSetting::primary_key_lazy_load]) getIndex(); diff --git a/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp b/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp index d69a00643f0..c3e740bde84 100644 --- a/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp @@ -122,4 +122,10 @@ std::string MergeTreeIndexGranularity::describe() const { return fmt::format("initialized: {}, marks_rows_partial_sums: [{}]", initialized, fmt::join(marks_rows_partial_sums, ", ")); } + +void MergeTreeIndexGranularity::shrinkToFitInMemory() +{ + marks_rows_partial_sums.shrink_to_fit(); +} + } diff --git a/src/Storages/MergeTree/MergeTreeIndexGranularity.h b/src/Storages/MergeTree/MergeTreeIndexGranularity.h index f66e721ec1e..9b8375dd2d8 100644 --- a/src/Storages/MergeTree/MergeTreeIndexGranularity.h +++ b/src/Storages/MergeTree/MergeTreeIndexGranularity.h @@ -100,6 +100,8 @@ public: void resizeWithFixedGranularity(size_t size, size_t fixed_granularity); std::string describe() const; + + void shrinkToFitInMemory(); }; } From 95d821549106ecff95e6e42e19b014aa6ac0e669 Mon Sep 17 00:00:00 2001 From: kssenii Date: Thu, 7 Nov 2024 17:34:52 +0100 Subject: [PATCH 520/680] Fix --- src/Interpreters/Cache/FileCache.cpp | 21 +++++++++++++++++++-- tests/config/config.d/storage_conf.xml | 1 + 2 files changed, 20 insertions(+), 2 deletions(-) diff --git a/src/Interpreters/Cache/FileCache.cpp b/src/Interpreters/Cache/FileCache.cpp index f7b7ffc5aea..7de3f7af78d 100644 --- a/src/Interpreters/Cache/FileCache.cpp +++ b/src/Interpreters/Cache/FileCache.cpp @@ -37,6 +37,11 @@ namespace ProfileEvents extern const Event FilesystemCacheFailToReserveSpaceBecauseOfCacheResize; } +namespace CurrentMetrics +{ + extern const Metric FilesystemCacheDownloadQueueElements; +} + namespace DB { @@ -918,7 +923,13 @@ bool FileCache::tryReserve( if (!query_priority->collectCandidatesForEviction( size, required_elements_num, reserve_stat, eviction_candidates, {}, user.user_id, cache_lock)) { - failure_reason = "cannot evict enough space for query limit"; + const auto & stat = reserve_stat.total_stat; + failure_reason = fmt::format( + "cannot evict enough space for query limit " + "(non-releasable count: {}, non-releasable size: {}, " + "releasable count: {}, releasable size: {}, background download elements: {})", + stat.non_releasable_count, stat.non_releasable_size, stat.releasable_count, stat.releasable_size, + CurrentMetrics::get(CurrentMetrics::FilesystemCacheDownloadQueueElements)); return false; } @@ -933,7 +944,13 @@ bool FileCache::tryReserve( if (!main_priority->collectCandidatesForEviction( size, required_elements_num, reserve_stat, eviction_candidates, queue_iterator, user.user_id, cache_lock)) { - failure_reason = "cannot evict enough space"; + const auto & stat = reserve_stat.total_stat; + failure_reason = fmt::format( + "cannot evict enough space " + "(non-releasable count: {}, non-releasable size: {}, " + "releasable count: {}, releasable size: {}, background download elements: {})", + stat.non_releasable_count, stat.non_releasable_size, stat.releasable_count, stat.releasable_size, + CurrentMetrics::get(CurrentMetrics::FilesystemCacheDownloadQueueElements)); return false; } diff --git a/tests/config/config.d/storage_conf.xml b/tests/config/config.d/storage_conf.xml index 74bad7528c8..fee7ce841a6 100644 --- a/tests/config/config.d/storage_conf.xml +++ b/tests/config/config.d/storage_conf.xml @@ -27,6 +27,7 @@ 0.3 0.15 0.15 + 50 0 From 2c59fce5b488c9ddd2d99e0dcbaaf84d2f36ef04 Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Thu, 7 Nov 2024 17:44:41 +0100 Subject: [PATCH 521/680] Update test.py --- tests/integration/test_storage_s3_queue/test.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/integration/test_storage_s3_queue/test.py b/tests/integration/test_storage_s3_queue/test.py index c495fc1d44f..284b304c632 100644 --- a/tests/integration/test_storage_s3_queue/test.py +++ b/tests/integration/test_storage_s3_queue/test.py @@ -1403,8 +1403,8 @@ def test_shards_distributed(started_cluster, mode, processing_threads): # A unique path is necessary for repeatable tests keeper_path = f"/clickhouse/test_{table_name}_{generate_random_string()}" files_path = f"{table_name}_data" - files_to_generate = 300 - row_num = 300 + files_to_generate = 600 + row_num = 1000 total_rows = row_num * files_to_generate shards_num = 2 From 45aaebc41a73131c4ceee63214afbc88104dd59f Mon Sep 17 00:00:00 2001 From: alesapin Date: Thu, 7 Nov 2024 18:24:36 +0100 Subject: [PATCH 522/680] Review fix --- src/Storages/MergeTree/MergedBlockOutputStream.cpp | 2 ++ src/Storages/MergeTree/MutateTask.cpp | 2 ++ 2 files changed, 4 insertions(+) diff --git a/src/Storages/MergeTree/MergedBlockOutputStream.cpp b/src/Storages/MergeTree/MergedBlockOutputStream.cpp index 77c34aae30a..39096718b5c 100644 --- a/src/Storages/MergeTree/MergedBlockOutputStream.cpp +++ b/src/Storages/MergeTree/MergedBlockOutputStream.cpp @@ -207,6 +207,8 @@ MergedBlockOutputStream::Finalizer MergedBlockOutputStream::finalizePartAsync( new_part->setBytesOnDisk(checksums.getTotalSizeOnDisk()); new_part->setBytesUncompressedOnDisk(checksums.getTotalSizeUncompressedOnDisk()); new_part->index_granularity = writer->getIndexGranularity(); + /// Just in case + new_part->index_granularity.shrinkToFitInMemory(); new_part->calculateColumnsAndSecondaryIndicesSizesOnDisk(); /// In mutation, existing_rows_count is already calculated in PartMergerWriter diff --git a/src/Storages/MergeTree/MutateTask.cpp b/src/Storages/MergeTree/MutateTask.cpp index 936df7b0275..7f6588fc632 100644 --- a/src/Storages/MergeTree/MutateTask.cpp +++ b/src/Storages/MergeTree/MutateTask.cpp @@ -984,6 +984,8 @@ void finalizeMutatedPart( new_data_part->rows_count = source_part->rows_count; new_data_part->index_granularity = source_part->index_granularity; + /// Just in case + new_data_part->index_granularity.shrinkToFitInMemory(); new_data_part->setIndex(*source_part->getIndex()); new_data_part->minmax_idx = source_part->minmax_idx; new_data_part->modification_time = time(nullptr); From 4fb38411c128e3a293c93d6f1d5f9b71c961e8db Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Thu, 7 Nov 2024 19:06:36 +0100 Subject: [PATCH 523/680] Only accept regular files --- programs/main.cpp | 2 +- src/Client/ClientBaseOptimizedParts.cpp | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/programs/main.cpp b/programs/main.cpp index ea8fbc1aece..d15c20867d1 100644 --- a/programs/main.cpp +++ b/programs/main.cpp @@ -238,7 +238,7 @@ int main(int argc_, char ** argv_) std::error_code ec; if (main_func == printHelp && !argv.empty() && (argv.size() == 1 || argv[1][0] == '-' || std::string_view(argv[1]).contains(' ') - || std::filesystem::exists(std::filesystem::path{argv[1]}, ec))) + || std::filesystem::is_regular_file(std::filesystem::path{argv[1]}, ec))) { main_func = mainEntryClickHouseLocal; } diff --git a/src/Client/ClientBaseOptimizedParts.cpp b/src/Client/ClientBaseOptimizedParts.cpp index bc362288079..afffe775029 100644 --- a/src/Client/ClientBaseOptimizedParts.cpp +++ b/src/Client/ClientBaseOptimizedParts.cpp @@ -119,7 +119,7 @@ void ClientApplicationBase::parseAndCheckOptions(OptionsDescription & options_de const char * option; std::error_code ec; - if (std::filesystem::exists(std::filesystem::path{token}, ec)) + if (std::filesystem::is_regular_file(std::filesystem::path{token}, ec)) option = "queries-file"; else if (token.contains(' ')) option = "query"; From 0ac6ce56bd08e25fc9c22022fec21f3346a753c5 Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Thu, 7 Nov 2024 18:19:26 +0000 Subject: [PATCH 524/680] Trying to fix short-circuit for FilterStep. --- src/Processors/QueryPlan/FilterStep.cpp | 98 ++++++++++++++++++++++++- 1 file changed, 97 insertions(+), 1 deletion(-) diff --git a/src/Processors/QueryPlan/FilterStep.cpp b/src/Processors/QueryPlan/FilterStep.cpp index 862e03d74f2..64c46332c34 100644 --- a/src/Processors/QueryPlan/FilterStep.cpp +++ b/src/Processors/QueryPlan/FilterStep.cpp @@ -5,6 +5,9 @@ #include #include #include +#include +#include +#include namespace DB { @@ -24,6 +27,78 @@ static ITransformingStep::Traits getTraits() }; } +static bool isTrivialSubtree(const ActionsDAG::Node * node) +{ + while (node->type == ActionsDAG::ActionType::ALIAS) + node = node->children.at(0); + + return node->type != ActionsDAG::ActionType::FUNCTION && node->type != ActionsDAG::ActionType::ARRAY_JOIN; +} + +struct ActionsAndName +{ + ActionsDAG dag; + std::string name; +}; + +static ActionsAndName splitSingleAndFilter(ActionsDAG & dag, const ActionsDAG::Node * filter_node) +{ + auto name = filter_node->result_name; + auto split_result = dag.split({filter_node}, true); + dag = std::move(split_result.second); + split_result.first.getOutputs().emplace(split_result.first.getOutputs().begin(), split_result.split_nodes_mapping[filter_node]); + return ActionsAndName{std::move(split_result.first), std::move(name)}; +} + +static std::optional trySplitSingleAndFilter(ActionsDAG & dag, const std::string & filter_name) +{ + const auto * filter = &dag.findInOutputs(filter_name); + while (filter->type == ActionsDAG::ActionType::ALIAS) + filter = filter->children.at(0); + + if (filter->type != ActionsDAG::ActionType::FUNCTION || filter->function_base->getName() != "and") + return {}; + + const ActionsDAG::Node * condition_to_split = nullptr; + std::stack nodes; + nodes.push(filter); + while (!nodes.empty()) + { + const auto * node = nodes.top(); + nodes.pop(); + + if (node->type == ActionsDAG::ActionType::FUNCTION && node->function_base->getName() == "and") + { + for (const auto * child : node->children | std::ranges::views::reverse) + nodes.push(child); + + continue; + } + + if (isTrivialSubtree(node)) + continue; + + /// Do not split subtree if it's the last non-trivial one. + /// So, split the first found condition only when there is a another one found. + if (condition_to_split) + return splitSingleAndFilter(dag, condition_to_split); + + condition_to_split = node; + } + + return {}; +} + +std::vector splitAndChainIntoMultipleFilters(ActionsDAG & dag, const std::string & filter_name) +{ + std::vector res; + + while (auto condition = trySplitSingleAndFilter(dag, filter_name)) + res.push_back(std::move(*condition)); + + return res; +} + FilterStep::FilterStep( const Header & input_header_, ActionsDAG actions_dag_, @@ -50,6 +125,17 @@ FilterStep::FilterStep( void FilterStep::transformPipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings & settings) { + auto and_atoms = splitAndChainIntoMultipleFilters(actions_dag, filter_column_name); + for (auto & and_atom : and_atoms) + { + auto expression = std::make_shared(std::move(and_atom.dag), settings.getActionsSettings()); + pipeline.addSimpleTransform([&](const Block & header, QueryPipelineBuilder::StreamType stream_type) + { + bool on_totals = stream_type == QueryPipelineBuilder::StreamType::Totals; + return std::make_shared(header, expression, and_atom.name, true, on_totals); + }); + } + auto expression = std::make_shared(std::move(actions_dag), settings.getActionsSettings()); pipeline.addSimpleTransform([&](const Block & header, QueryPipelineBuilder::StreamType stream_type) @@ -76,13 +162,23 @@ void FilterStep::transformPipeline(QueryPipelineBuilder & pipeline, const BuildQ void FilterStep::describeActions(FormatSettings & settings) const { String prefix(settings.offset, settings.indent_char); + + auto cloned_dag = actions_dag.clone(); + auto and_atoms = splitAndChainIntoMultipleFilters(cloned_dag, filter_column_name); + for (auto & and_atom : and_atoms) + { + auto expression = std::make_shared(std::move(and_atom.dag)); + settings.out << prefix << "AND column: " << and_atom.name; + expression->describeActions(settings.out, prefix); + } + settings.out << prefix << "Filter column: " << filter_column_name; if (remove_filter_column) settings.out << " (removed)"; settings.out << '\n'; - auto expression = std::make_shared(actions_dag.clone()); + auto expression = std::make_shared(std::move(cloned_dag)); expression->describeActions(settings.out, prefix); } From 4e53dda5801cf797a85ad07b9fb55e08aa0cdcf8 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Thu, 7 Nov 2024 20:45:31 +0100 Subject: [PATCH 525/680] Use array for conditional mkdir --- docker/server/entrypoint.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/docker/server/entrypoint.sh b/docker/server/entrypoint.sh index 7a990e7d889..5a91d54d32b 100755 --- a/docker/server/entrypoint.sh +++ b/docker/server/entrypoint.sh @@ -57,14 +57,14 @@ function create_directory_and_do_chown() { [ -z "$dir" ] && return # ensure directories exist if [ "$DO_CHOWN" = "1" ]; then - mkdir="mkdir" + mkdir=( mkdir ) else # if DO_CHOWN=0 it means that the system does not map root user to "admin" permissions # it mainly happens on NFS mounts where root==nobody for security reasons # thus mkdir MUST run with user id/gid and not from nobody that has zero permissions - mkdir="clickhouse su ""${USER}:${GROUP}"" mkdir" + mkdir=( clickhouse su "${USER}:${GROUP}" mkdir ) fi - if ! $mkdir -p "$dir"; then + if ! "${mkdir[@]}" -p "$dir"; then echo "Couldn't create necessary directory: $dir" exit 1 fi From 2fa357f3747a9436acdeefd4c255e5333c461c3f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Thu, 7 Nov 2024 20:51:39 +0100 Subject: [PATCH 526/680] Revert "Enable enable_job_stack_trace by default" --- src/Core/Settings.cpp | 2 +- src/Core/SettingsChangesHistory.cpp | 1 - 2 files changed, 1 insertion(+), 2 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 01339226c2d..6f0109fa300 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -2869,7 +2869,7 @@ Limit on size of multipart/form-data content. This setting cannot be parsed from DECLARE(Bool, calculate_text_stack_trace, true, R"( Calculate text stack trace in case of exceptions during query execution. This is the default. It requires symbol lookups that may slow down fuzzing tests when a huge amount of wrong queries are executed. In normal cases, you should not disable this option. )", 0) \ - DECLARE(Bool, enable_job_stack_trace, true, R"( + DECLARE(Bool, enable_job_stack_trace, false, R"( Output stack trace of a job creator when job results in exception )", 0) \ DECLARE(Bool, allow_ddl, true, R"( diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index edf4e60706b..c6223bef2b2 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -80,7 +80,6 @@ static std::initializer_list Date: Thu, 7 Nov 2024 19:53:30 +0000 Subject: [PATCH 527/680] Fix getting column sample for not finalized part --- src/Columns/ColumnVariant.cpp | 2 +- src/Storages/MergeTree/IMergeTreeDataPart.cpp | 11 ++++++----- src/Storages/MergeTree/IMergeTreeDataPart.h | 6 +++--- src/Storages/MergeTree/IMergeTreeDataPartWriter.h | 2 ++ src/Storages/MergeTree/MergeTreeDataPartCompact.cpp | 2 +- src/Storages/MergeTree/MergeTreeDataPartCompact.h | 2 +- src/Storages/MergeTree/MergeTreeDataPartWide.cpp | 8 ++++---- src/Storages/MergeTree/MergeTreeDataPartWide.h | 4 ++-- .../MergeTree/MergeTreeDataPartWriterOnDisk.h | 2 ++ src/Storages/MergeTree/MergedBlockOutputStream.cpp | 2 +- 10 files changed, 23 insertions(+), 18 deletions(-) diff --git a/src/Columns/ColumnVariant.cpp b/src/Columns/ColumnVariant.cpp index 54f0421fc4b..2fa59b8e33c 100644 --- a/src/Columns/ColumnVariant.cpp +++ b/src/Columns/ColumnVariant.cpp @@ -952,7 +952,7 @@ ColumnPtr ColumnVariant::permute(const Permutation & perm, size_t limit) const if (hasOnlyNulls()) { if (limit) - return cloneResized(limit); + return cloneResized(limit ? std::min(size(), limit) : size()); /// If no limit, we can just return current immutable column. return this->getPtr(); diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp index b631d991e90..f73b52dbafd 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp @@ -2252,18 +2252,18 @@ void IMergeTreeDataPart::checkConsistencyWithProjections(bool require_part_metad proj_part->checkConsistency(require_part_metadata); } -void IMergeTreeDataPart::calculateColumnsAndSecondaryIndicesSizesOnDisk() +void IMergeTreeDataPart::calculateColumnsAndSecondaryIndicesSizesOnDisk(std::optional columns_sample) { - calculateColumnsSizesOnDisk(); + calculateColumnsSizesOnDisk(columns_sample); calculateSecondaryIndicesSizesOnDisk(); } -void IMergeTreeDataPart::calculateColumnsSizesOnDisk() +void IMergeTreeDataPart::calculateColumnsSizesOnDisk(std::optional columns_sample) { if (getColumns().empty() || checksums.empty()) throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot calculate columns sizes when columns or checksums are not initialized"); - calculateEachColumnSizes(columns_sizes, total_columns_size); + calculateEachColumnSizes(columns_sizes, total_columns_size, columns_sample); } void IMergeTreeDataPart::calculateSecondaryIndicesSizesOnDisk() @@ -2501,7 +2501,7 @@ ColumnPtr IMergeTreeDataPart::getColumnSample(const NameAndTypePair & column) co { const size_t total_mark = getMarksCount(); /// If column doesn't have dynamic subcolumns or part has no data, just create column using it's type. - if (is_temp || !column.type->hasDynamicSubcolumns() || !total_mark) + if (!column.type->hasDynamicSubcolumns() || !total_mark) return column.type->createColumn(); /// Otherwise, read sample column with 0 rows from the part, so it will load dynamic structure. @@ -2527,6 +2527,7 @@ ColumnPtr IMergeTreeDataPart::getColumnSample(const NameAndTypePair & column) co Columns result; result.resize(1); + LOG_DEBUG(getLogger("IMergeTreeDataPart"), "getColumnSample"); reader->readRows(0, total_mark, false, 0, result); return result[0]; } diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.h b/src/Storages/MergeTree/IMergeTreeDataPart.h index b41a1d840e1..a7051a2491a 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.h +++ b/src/Storages/MergeTree/IMergeTreeDataPart.h @@ -426,7 +426,7 @@ public: bool shallParticipateInMerges(const StoragePolicyPtr & storage_policy) const; /// Calculate column and secondary indices sizes on disk. - void calculateColumnsAndSecondaryIndicesSizesOnDisk(); + void calculateColumnsAndSecondaryIndicesSizesOnDisk(std::optional columns_sample = std::nullopt); std::optional getRelativePathForPrefix(const String & prefix, bool detached = false, bool broken = false) const; @@ -631,7 +631,7 @@ protected: /// Fill each_columns_size and total_size with sizes from columns files on /// disk using columns and checksums. - virtual void calculateEachColumnSizes(ColumnSizeByName & each_columns_size, ColumnSize & total_size) const = 0; + virtual void calculateEachColumnSizes(ColumnSizeByName & each_columns_size, ColumnSize & total_size, std::optional columns_sample) const = 0; std::optional getRelativePathForDetachedPart(const String & prefix, bool broken) const; @@ -713,7 +713,7 @@ private: void loadPartitionAndMinMaxIndex(); - void calculateColumnsSizesOnDisk(); + void calculateColumnsSizesOnDisk(std::optional columns_sample = std::nullopt); void calculateSecondaryIndicesSizesOnDisk(); diff --git a/src/Storages/MergeTree/IMergeTreeDataPartWriter.h b/src/Storages/MergeTree/IMergeTreeDataPartWriter.h index d1c76505d7c..8923f6a59ca 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPartWriter.h +++ b/src/Storages/MergeTree/IMergeTreeDataPartWriter.h @@ -54,6 +54,8 @@ public: const MergeTreeIndexGranularity & getIndexGranularity() const { return index_granularity; } + virtual Block getColumnsSample() const = 0; + protected: SerializationPtr getSerialization(const String & column_name) const; diff --git a/src/Storages/MergeTree/MergeTreeDataPartCompact.cpp b/src/Storages/MergeTree/MergeTreeDataPartCompact.cpp index 14c2da82de1..8856f467b90 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartCompact.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartCompact.cpp @@ -80,7 +80,7 @@ MergeTreeDataPartWriterPtr createMergeTreeDataPartCompactWriter( } -void MergeTreeDataPartCompact::calculateEachColumnSizes(ColumnSizeByName & /*each_columns_size*/, ColumnSize & total_size) const +void MergeTreeDataPartCompact::calculateEachColumnSizes(ColumnSizeByName & /*each_columns_size*/, ColumnSize & total_size, std::optional /*columns_sample*/) const { auto bin_checksum = checksums.files.find(DATA_FILE_NAME_WITH_EXTENSION); if (bin_checksum != checksums.files.end()) diff --git a/src/Storages/MergeTree/MergeTreeDataPartCompact.h b/src/Storages/MergeTree/MergeTreeDataPartCompact.h index 8e279571578..c394de0d7c1 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartCompact.h +++ b/src/Storages/MergeTree/MergeTreeDataPartCompact.h @@ -70,7 +70,7 @@ private: void loadIndexGranularity() override; /// Compact parts don't support per column size, only total size - void calculateEachColumnSizes(ColumnSizeByName & each_columns_size, ColumnSize & total_size) const override; + void calculateEachColumnSizes(ColumnSizeByName & each_columns_size, ColumnSize & total_size, std::optional columns_sample) const override; }; } diff --git a/src/Storages/MergeTree/MergeTreeDataPartWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWide.cpp index b3b6a0dded6..39f96ba06ad 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWide.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWide.cpp @@ -82,7 +82,7 @@ MergeTreeDataPartWriterPtr createMergeTreeDataPartWideWriter( /// Takes into account the fact that several columns can e.g. share their .size substreams. /// When calculating totals these should be counted only once. ColumnSize MergeTreeDataPartWide::getColumnSizeImpl( - const NameAndTypePair & column, std::unordered_set * processed_substreams) const + const NameAndTypePair & column, std::unordered_set * processed_substreams, std::optional columns_sample) const { ColumnSize size; if (checksums.empty()) @@ -108,7 +108,7 @@ ColumnSize MergeTreeDataPartWide::getColumnSizeImpl( auto mrk_checksum = checksums.files.find(*stream_name + getMarksFileExtension()); if (mrk_checksum != checksums.files.end()) size.marks += mrk_checksum->second.file_size; - }, column.type, getColumnSample(column)); + }, column.type, columns_sample && columns_sample->has(column.name) ? columns_sample->getByName(column.name).column : getColumnSample(column)); return size; } @@ -374,12 +374,12 @@ std::optional MergeTreeDataPartWide::getFileNameForColumn(const NameAndT return filename; } -void MergeTreeDataPartWide::calculateEachColumnSizes(ColumnSizeByName & each_columns_size, ColumnSize & total_size) const +void MergeTreeDataPartWide::calculateEachColumnSizes(ColumnSizeByName & each_columns_size, ColumnSize & total_size, std::optional columns_sample) const { std::unordered_set processed_substreams; for (const auto & column : columns) { - ColumnSize size = getColumnSizeImpl(column, &processed_substreams); + ColumnSize size = getColumnSizeImpl(column, &processed_substreams, columns_sample); each_columns_size[column.name] = size; total_size.add(size); diff --git a/src/Storages/MergeTree/MergeTreeDataPartWide.h b/src/Storages/MergeTree/MergeTreeDataPartWide.h index 022a5fb746c..a6d4897ed87 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWide.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWide.h @@ -64,9 +64,9 @@ private: /// Loads marks index granularity into memory void loadIndexGranularity() override; - ColumnSize getColumnSizeImpl(const NameAndTypePair & column, std::unordered_set * processed_substreams) const; + ColumnSize getColumnSizeImpl(const NameAndTypePair & column, std::unordered_set * processed_substreams, std::optional columns_sample) const; - void calculateEachColumnSizes(ColumnSizeByName & each_columns_size, ColumnSize & total_size) const override; + void calculateEachColumnSizes(ColumnSizeByName & each_columns_size, ColumnSize & total_size, std::optional columns_sample) const override; }; diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h index 49d654c15e1..b22d58ba51e 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h @@ -123,6 +123,8 @@ public: written_offset_columns = written_offset_columns_; } + Block getColumnsSample() const override { return block_sample; } + protected: /// Count index_granularity for block and store in `index_granularity` size_t computeIndexGranularity(const Block & block) const; diff --git a/src/Storages/MergeTree/MergedBlockOutputStream.cpp b/src/Storages/MergeTree/MergedBlockOutputStream.cpp index 77c34aae30a..604b2fda20a 100644 --- a/src/Storages/MergeTree/MergedBlockOutputStream.cpp +++ b/src/Storages/MergeTree/MergedBlockOutputStream.cpp @@ -207,7 +207,7 @@ MergedBlockOutputStream::Finalizer MergedBlockOutputStream::finalizePartAsync( new_part->setBytesOnDisk(checksums.getTotalSizeOnDisk()); new_part->setBytesUncompressedOnDisk(checksums.getTotalSizeUncompressedOnDisk()); new_part->index_granularity = writer->getIndexGranularity(); - new_part->calculateColumnsAndSecondaryIndicesSizesOnDisk(); + new_part->calculateColumnsAndSecondaryIndicesSizesOnDisk(writer->getColumnsSample()); /// In mutation, existing_rows_count is already calculated in PartMergerWriter /// In merge situation, lightweight deleted rows was physically deleted, existing_rows_count equals rows_count From 3525954fa3cd116bf0b7ec70dc70be3999cf0090 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Thu, 7 Nov 2024 20:55:04 +0100 Subject: [PATCH 528/680] Implicit SELECT in clickhouse-local --- programs/local/LocalServer.cpp | 6 +++--- src/Client/ClientBase.cpp | 5 ++++- src/Client/ClientBaseHelpers.cpp | 11 +++++++++-- src/Client/ClientBaseHelpers.h | 4 +++- src/Core/Settings.cpp | 2 ++ .../0_stateless/03267_implicit_select.reference | 5 +++++ tests/queries/0_stateless/03267_implicit_select.sh | 11 +++++++++++ 7 files changed, 37 insertions(+), 7 deletions(-) create mode 100644 tests/queries/0_stateless/03267_implicit_select.reference create mode 100755 tests/queries/0_stateless/03267_implicit_select.sh diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp index 1dcef5eb25e..145cac02a3c 100644 --- a/programs/local/LocalServer.cpp +++ b/programs/local/LocalServer.cpp @@ -22,7 +22,6 @@ #include #include #include -#include #include #include #include @@ -31,7 +30,6 @@ #include #include #include -#include #include #include #include @@ -50,7 +48,6 @@ #include #include #include -#include #include #include #include @@ -71,9 +68,11 @@ namespace CurrentMetrics namespace DB { + namespace Setting { extern const SettingsBool allow_introspection_functions; + extern const SettingsBool implicit_select; extern const SettingsLocalFSReadMethod storage_file_read_method; } @@ -126,6 +125,7 @@ void applySettingsOverridesForLocal(ContextMutablePtr context) settings[Setting::allow_introspection_functions] = true; settings[Setting::storage_file_read_method] = LocalFSReadMethod::mmap; + settings[Setting::implicit_select] = true; context->setSettings(settings); } diff --git a/src/Client/ClientBase.cpp b/src/Client/ClientBase.cpp index 0a824753dc0..29abed7e52d 100644 --- a/src/Client/ClientBase.cpp +++ b/src/Client/ClientBase.cpp @@ -2674,7 +2674,10 @@ void ClientBase::runInteractive() #if USE_REPLXX replxx::Replxx::highlighter_callback_t highlight_callback{}; if (getClientConfiguration().getBool("highlight", true)) - highlight_callback = highlight; + highlight_callback = [this](const String & query, std::vector & colors) + { + highlight(query, colors, *client_context); + }; ReplxxLineReader lr( *suggest, diff --git a/src/Client/ClientBaseHelpers.cpp b/src/Client/ClientBaseHelpers.cpp index 156c0c87fb6..ea2a5fd42f5 100644 --- a/src/Client/ClientBaseHelpers.cpp +++ b/src/Client/ClientBaseHelpers.cpp @@ -5,6 +5,8 @@ #include #include #include +#include +#include #include @@ -12,6 +14,11 @@ namespace DB { +namespace Setting +{ + extern const SettingsBool implicit_select; +} + /// Should we celebrate a bit? bool isNewYearMode() { @@ -95,7 +102,7 @@ bool isChineseNewYearMode(const String & local_tz) } #if USE_REPLXX -void highlight(const String & query, std::vector & colors) +void highlight(const String & query, std::vector & colors, const Context & context) { using namespace replxx; @@ -135,7 +142,7 @@ void highlight(const String & query, std::vector & colors /// Currently we highlight only the first query in the multi-query mode. - ParserQuery parser(end); + ParserQuery parser(end, false, context.getSettingsRef()[Setting::implicit_select]); ASTPtr ast; bool parse_res = false; diff --git a/src/Client/ClientBaseHelpers.h b/src/Client/ClientBaseHelpers.h index adc1c81b3c5..dcfac21c500 100644 --- a/src/Client/ClientBaseHelpers.h +++ b/src/Client/ClientBaseHelpers.h @@ -11,13 +11,15 @@ namespace DB { +class Context; + /// Should we celebrate a bit? bool isNewYearMode(); bool isChineseNewYearMode(const String & local_tz); #if USE_REPLXX -void highlight(const String & query, std::vector & colors); +void highlight(const String & query, std::vector & colors, const Context & context); #endif } diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index f3ada33cb37..049e29dc8d8 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -5708,6 +5708,8 @@ If enabled, MongoDB tables will return an error when a MongoDB query cannot be b )", 0) \ DECLARE(Bool, implicit_select, false, R"( Allow writing simple SELECT queries without the leading SELECT keyword, which makes it simple for calculator-style usage, e.g. `1 + 2` becomes a valid query. + +In `clickhouse-local` it is enabled by default and can be explicitly disabled. )", 0) \ \ \ diff --git a/tests/queries/0_stateless/03267_implicit_select.reference b/tests/queries/0_stateless/03267_implicit_select.reference new file mode 100644 index 00000000000..97c1fd4333b --- /dev/null +++ b/tests/queries/0_stateless/03267_implicit_select.reference @@ -0,0 +1,5 @@ +3 +3 +3 +Syntax error +3 diff --git a/tests/queries/0_stateless/03267_implicit_select.sh b/tests/queries/0_stateless/03267_implicit_select.sh new file mode 100755 index 00000000000..068fb457bb1 --- /dev/null +++ b/tests/queries/0_stateless/03267_implicit_select.sh @@ -0,0 +1,11 @@ +#!/usr/bin/env bash + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + +$CLICKHOUSE_LOCAL "1 + 2" +$CLICKHOUSE_LOCAL -q "1 + 2" +$CLICKHOUSE_LOCAL --query "1 + 2" +$CLICKHOUSE_LOCAL --implicit_select 0 --query "1 + 2" 2>&1 | grep -oF 'Syntax error' +$CLICKHOUSE_LOCAL --implicit_select 0 --query "SELECT 1 + 2" From 8f98f2333f21566ab62430a8bc9379e6b24f6062 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Thu, 7 Nov 2024 20:49:06 +0100 Subject: [PATCH 529/680] Make `clickhouse local` fuse in the repository install RUN --- docker/server/Dockerfile.ubuntu | 52 ++++++++++++++++----------------- 1 file changed, 26 insertions(+), 26 deletions(-) diff --git a/docker/server/Dockerfile.ubuntu b/docker/server/Dockerfile.ubuntu index 2b023a9cf03..0fe9a409ee4 100644 --- a/docker/server/Dockerfile.ubuntu +++ b/docker/server/Dockerfile.ubuntu @@ -88,32 +88,32 @@ RUN if [ -n "${single_binary_location_url}" ]; then \ #docker-official-library:on # A fallback to installation from ClickHouse repository -RUN if ! clickhouse local -q "SELECT ''" > /dev/null 2>&1; then \ - apt-get update \ - && apt-get install --yes --no-install-recommends \ - dirmngr \ - gnupg2 \ - && mkdir -p /etc/apt/sources.list.d \ - && GNUPGHOME=$(mktemp -d) \ - && GNUPGHOME="$GNUPGHOME" gpg --batch --no-default-keyring \ - --keyring /usr/share/keyrings/clickhouse-keyring.gpg \ - --keyserver hkp://keyserver.ubuntu.com:80 \ - --recv-keys 3a9ea1193a97b548be1457d48919f6bd2b48d754 \ - && rm -rf "$GNUPGHOME" \ - && chmod +r /usr/share/keyrings/clickhouse-keyring.gpg \ - && echo "${REPOSITORY}" > /etc/apt/sources.list.d/clickhouse.list \ - && echo "installing from repository: ${REPOSITORY}" \ - && apt-get update \ - && for package in ${PACKAGES}; do \ - packages="${packages} ${package}=${VERSION}" \ - ; done \ - && apt-get install --yes --no-install-recommends ${packages} || exit 1 \ - && rm -rf \ - /var/lib/apt/lists/* \ - /var/cache/debconf \ - /tmp/* \ - && apt-get autoremove --purge -yq dirmngr gnupg2 \ - ; fi +# It works unless the clickhouse binary already exists +RUN clickhouse local -q 'SELECT 1' >/dev/null 2>&1 && exit 0 || : \ + ; apt-get update \ + && apt-get install --yes --no-install-recommends \ + dirmngr \ + gnupg2 \ + && mkdir -p /etc/apt/sources.list.d \ + && GNUPGHOME=$(mktemp -d) \ + && GNUPGHOME="$GNUPGHOME" gpg --batch --no-default-keyring \ + --keyring /usr/share/keyrings/clickhouse-keyring.gpg \ + --keyserver hkp://keyserver.ubuntu.com:80 \ + --recv-keys 3a9ea1193a97b548be1457d48919f6bd2b48d754 \ + && rm -rf "$GNUPGHOME" \ + && chmod +r /usr/share/keyrings/clickhouse-keyring.gpg \ + && echo "${REPOSITORY}" > /etc/apt/sources.list.d/clickhouse.list \ + && echo "installing from repository: ${REPOSITORY}" \ + && apt-get update \ + && for package in ${PACKAGES}; do \ + packages="${packages} ${package}=${VERSION}" \ + ; done \ + && apt-get install --yes --no-install-recommends ${packages} || exit 1 \ + && rm -rf \ + /var/lib/apt/lists/* \ + /var/cache/debconf \ + /tmp/* \ + && apt-get autoremove --purge -yq dirmngr gnupg2 # post install # we need to allow "others" access to clickhouse folder, because docker container From 0ff0c96b007108ab222a264e4a3bf8aa7cb7a18e Mon Sep 17 00:00:00 2001 From: avogar Date: Thu, 7 Nov 2024 20:01:40 +0000 Subject: [PATCH 530/680] Remove logging --- src/Storages/MergeTree/IMergeTreeDataPart.cpp | 1 - 1 file changed, 1 deletion(-) diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp index f73b52dbafd..4e400fb1f94 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp @@ -2527,7 +2527,6 @@ ColumnPtr IMergeTreeDataPart::getColumnSample(const NameAndTypePair & column) co Columns result; result.resize(1); - LOG_DEBUG(getLogger("IMergeTreeDataPart"), "getColumnSample"); reader->readRows(0, total_mark, false, 0, result); return result[0]; } From 76b6cf96eb3f548bc442f645a8cd8999cf3c6f63 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Thu, 7 Nov 2024 21:26:23 +0100 Subject: [PATCH 531/680] Highlight multi-statements in the client --- src/Client/ClientBaseHelpers.cpp | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-) diff --git a/src/Client/ClientBaseHelpers.cpp b/src/Client/ClientBaseHelpers.cpp index 156c0c87fb6..f7ecbfeeb43 100644 --- a/src/Client/ClientBaseHelpers.cpp +++ b/src/Client/ClientBaseHelpers.cpp @@ -141,7 +141,24 @@ void highlight(const String & query, std::vector & colors try { - parse_res = parser.parse(token_iterator, ast, expected); + while (true) + { + parse_res = parser.parse(token_iterator, ast, expected); + if (!parse_res) + break; + + if (!token_iterator->isEnd() && token_iterator->type != TokenType::Semicolon) + { + parse_res = false; + break; + } + + while (token_iterator->type == TokenType::Semicolon) + ++token_iterator; + + if (token_iterator->isEnd()) + break; + } } catch (...) { @@ -175,7 +192,7 @@ void highlight(const String & query, std::vector & colors /// Highlight the last error in red. If the parser failed or the lexer found an invalid token, /// or if it didn't parse all the data (except, the data for INSERT query, which is legitimately unparsed) - if ((!parse_res || last_token.isError() || (!token_iterator->isEnd() && token_iterator->type != TokenType::Semicolon)) + if ((!parse_res || last_token.isError()) && !(insert_data && expected.max_parsed_pos >= insert_data) && expected.max_parsed_pos >= prev) { From c8104cb2ee0f366a56bfd79a07071173a8a5a815 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Thu, 7 Nov 2024 21:28:06 +0100 Subject: [PATCH 532/680] Correct and unify exit codes --- programs/client/Client.cpp | 10 ++++++---- programs/disks/DisksApp.cpp | 8 +++++--- programs/keeper-client/KeeperClient.cpp | 6 ++++-- programs/keeper/Keeper.cpp | 4 ++-- programs/library-bridge/LibraryBridge.cpp | 2 +- programs/local/LocalServer.cpp | 12 +++++++----- programs/obfuscator/Obfuscator.cpp | 2 +- programs/odbc-bridge/ODBCBridge.cpp | 2 +- programs/server/Server.cpp | 4 ++-- 9 files changed, 29 insertions(+), 21 deletions(-) diff --git a/programs/client/Client.cpp b/programs/client/Client.cpp index d7190444f0b..05e1e61be7b 100644 --- a/programs/client/Client.cpp +++ b/programs/client/Client.cpp @@ -431,7 +431,7 @@ catch (const Exception & e) bool need_print_stack_trace = config().getBool("stacktrace", false) && e.code() != ErrorCodes::NETWORK_ERROR; std::cerr << getExceptionMessage(e, need_print_stack_trace, true) << std::endl << std::endl; /// If exception code isn't zero, we should return non-zero return code anyway. - return e.code() ? e.code() : -1; + return static_cast(e.code()) ? e.code() : -1; } catch (...) { @@ -1390,7 +1390,8 @@ int mainEntryClickHouseClient(int argc, char ** argv) catch (const DB::Exception & e) { std::cerr << DB::getExceptionMessage(e, false) << std::endl; - return 1; + auto code = DB::getCurrentExceptionCode(); + return static_cast(code) ? code : 1; } catch (const boost::program_options::error & e) { @@ -1399,7 +1400,8 @@ int mainEntryClickHouseClient(int argc, char ** argv) } catch (...) { - std::cerr << DB::getCurrentExceptionMessage(true) << std::endl; - return 1; + std::cerr << DB::getCurrentExceptionMessage(true) << '\n'; + auto code = DB::getCurrentExceptionCode(); + return static_cast(code) ? code : 1; } } diff --git a/programs/disks/DisksApp.cpp b/programs/disks/DisksApp.cpp index 610d8eaa638..d6541e99288 100644 --- a/programs/disks/DisksApp.cpp +++ b/programs/disks/DisksApp.cpp @@ -546,16 +546,18 @@ int mainEntryClickHouseDisks(int argc, char ** argv) catch (const DB::Exception & e) { std::cerr << DB::getExceptionMessage(e, false) << std::endl; - return 0; + auto code = DB::getCurrentExceptionCode(); + return static_cast(code) ? code : 1; } catch (const boost::program_options::error & e) { std::cerr << "Bad arguments: " << e.what() << std::endl; - return 0; + return DB::ErrorCodes::BAD_ARGUMENTS; } catch (...) { std::cerr << DB::getCurrentExceptionMessage(true) << std::endl; - return 0; + auto code = DB::getCurrentExceptionCode(); + return static_cast(code) ? code : 1; } } diff --git a/programs/keeper-client/KeeperClient.cpp b/programs/keeper-client/KeeperClient.cpp index 2a426fad7ac..4bdddaec59c 100644 --- a/programs/keeper-client/KeeperClient.cpp +++ b/programs/keeper-client/KeeperClient.cpp @@ -448,7 +448,8 @@ int mainEntryClickHouseKeeperClient(int argc, char ** argv) catch (const DB::Exception & e) { std::cerr << DB::getExceptionMessage(e, false) << std::endl; - return 1; + auto code = DB::getCurrentExceptionCode(); + return static_cast(code) ? code : 1; } catch (const boost::program_options::error & e) { @@ -458,6 +459,7 @@ int mainEntryClickHouseKeeperClient(int argc, char ** argv) catch (...) { std::cerr << DB::getCurrentExceptionMessage(true) << std::endl; - return 1; + auto code = DB::getCurrentExceptionCode(); + return static_cast(code) ? code : 1; } } diff --git a/programs/keeper/Keeper.cpp b/programs/keeper/Keeper.cpp index 74af9950e13..936ce15f4c9 100644 --- a/programs/keeper/Keeper.cpp +++ b/programs/keeper/Keeper.cpp @@ -81,7 +81,7 @@ int mainEntryClickHouseKeeper(int argc, char ** argv) { std::cerr << DB::getCurrentExceptionMessage(true) << "\n"; auto code = DB::getCurrentExceptionCode(); - return code ? code : 1; + return static_cast(code) ? code : 1; } } @@ -672,7 +672,7 @@ catch (...) /// Poco does not provide stacktrace. tryLogCurrentException("Application"); auto code = getCurrentExceptionCode(); - return code ? code : -1; + return static_cast(code) ? code : -1; } diff --git a/programs/library-bridge/LibraryBridge.cpp b/programs/library-bridge/LibraryBridge.cpp index 261484ac744..62dbd12aaf0 100644 --- a/programs/library-bridge/LibraryBridge.cpp +++ b/programs/library-bridge/LibraryBridge.cpp @@ -13,7 +13,7 @@ int mainEntryClickHouseLibraryBridge(int argc, char ** argv) { std::cerr << DB::getCurrentExceptionMessage(true) << "\n"; auto code = DB::getCurrentExceptionCode(); - return code ? code : 1; + return static_cast(code) ? code : 1; } } diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp index 1dcef5eb25e..d6bf0353e89 100644 --- a/programs/local/LocalServer.cpp +++ b/programs/local/LocalServer.cpp @@ -615,12 +615,14 @@ catch (const DB::Exception & e) { bool need_print_stack_trace = getClientConfiguration().getBool("stacktrace", false); std::cerr << getExceptionMessage(e, need_print_stack_trace, true) << std::endl; - return e.code() ? e.code() : -1; + auto code = DB::getCurrentExceptionCode(); + return static_cast(code) ? code : 1; } catch (...) { - std::cerr << getCurrentExceptionMessage(false) << std::endl; - return getCurrentExceptionCode(); + std::cerr << DB::getCurrentExceptionMessage(true) << '\n'; + auto code = DB::getCurrentExceptionCode(); + return static_cast(code) ? code : 1; } void LocalServer::updateLoggerLevel(const String & logs_level) @@ -1029,7 +1031,7 @@ int mainEntryClickHouseLocal(int argc, char ** argv) { std::cerr << DB::getExceptionMessage(e, false) << std::endl; auto code = DB::getCurrentExceptionCode(); - return code ? code : 1; + return static_cast(code) ? code : 1; } catch (const boost::program_options::error & e) { @@ -1040,6 +1042,6 @@ int mainEntryClickHouseLocal(int argc, char ** argv) { std::cerr << DB::getCurrentExceptionMessage(true) << '\n'; auto code = DB::getCurrentExceptionCode(); - return code ? code : 1; + return static_cast(code) ? code : 1; } } diff --git a/programs/obfuscator/Obfuscator.cpp b/programs/obfuscator/Obfuscator.cpp index 324a4573b24..6bd3865b591 100644 --- a/programs/obfuscator/Obfuscator.cpp +++ b/programs/obfuscator/Obfuscator.cpp @@ -1480,5 +1480,5 @@ catch (...) { std::cerr << DB::getCurrentExceptionMessage(true) << "\n"; auto code = DB::getCurrentExceptionCode(); - return code ? code : 1; + return static_cast(code) ? code : 1; } diff --git a/programs/odbc-bridge/ODBCBridge.cpp b/programs/odbc-bridge/ODBCBridge.cpp index 096d1b2dcca..e5ae3272d40 100644 --- a/programs/odbc-bridge/ODBCBridge.cpp +++ b/programs/odbc-bridge/ODBCBridge.cpp @@ -13,7 +13,7 @@ int mainEntryClickHouseODBCBridge(int argc, char ** argv) { std::cerr << DB::getCurrentExceptionMessage(true) << "\n"; auto code = DB::getCurrentExceptionCode(); - return code ? code : 1; + return static_cast(code) ? code : 1; } } diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp index 5159f95419e..68f262079ff 100644 --- a/programs/server/Server.cpp +++ b/programs/server/Server.cpp @@ -343,7 +343,7 @@ int mainEntryClickHouseServer(int argc, char ** argv) { std::cerr << DB::getCurrentExceptionMessage(true) << "\n"; auto code = DB::getCurrentExceptionCode(); - return code ? code : 1; + return static_cast(code) ? code : 1; } } @@ -2537,7 +2537,7 @@ catch (...) /// Poco does not provide stacktrace. tryLogCurrentException("Application"); auto code = getCurrentExceptionCode(); - return code ? code : -1; + return static_cast(code) ? code : -1; } std::unique_ptr Server::buildProtocolStackFromConfig( From a027f1bf3cde1442a427610cf17967147cb0d60c Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Thu, 7 Nov 2024 15:59:11 -0500 Subject: [PATCH 533/680] Revert "Revert "Enable enable_job_stack_trace by default"" --- src/Core/Settings.cpp | 2 +- src/Core/SettingsChangesHistory.cpp | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 6f0109fa300..01339226c2d 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -2869,7 +2869,7 @@ Limit on size of multipart/form-data content. This setting cannot be parsed from DECLARE(Bool, calculate_text_stack_trace, true, R"( Calculate text stack trace in case of exceptions during query execution. This is the default. It requires symbol lookups that may slow down fuzzing tests when a huge amount of wrong queries are executed. In normal cases, you should not disable this option. )", 0) \ - DECLARE(Bool, enable_job_stack_trace, false, R"( + DECLARE(Bool, enable_job_stack_trace, true, R"( Output stack trace of a job creator when job results in exception )", 0) \ DECLARE(Bool, allow_ddl, true, R"( diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index c6223bef2b2..edf4e60706b 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -80,6 +80,7 @@ static std::initializer_list Date: Thu, 7 Nov 2024 16:01:02 -0500 Subject: [PATCH 534/680] move enable_job_stack_trace change to 24.11 --- src/Core/SettingsChangesHistory.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index edf4e60706b..0ff9d0a6833 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -64,6 +64,7 @@ static std::initializer_list Date: Thu, 7 Nov 2024 22:40:06 +0100 Subject: [PATCH 535/680] Update src/Client/ClientBaseHelpers.cpp Co-authored-by: Konstantin Bogdanov --- src/Client/ClientBaseHelpers.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Client/ClientBaseHelpers.cpp b/src/Client/ClientBaseHelpers.cpp index f7ecbfeeb43..555e95f7a25 100644 --- a/src/Client/ClientBaseHelpers.cpp +++ b/src/Client/ClientBaseHelpers.cpp @@ -141,7 +141,7 @@ void highlight(const String & query, std::vector & colors try { - while (true) + while (!token_iterator->isEnd()) { parse_res = parser.parse(token_iterator, ast, expected); if (!parse_res) From 1e87298a1ceafcf10fe0e5586604387bab0c6048 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Thu, 7 Nov 2024 22:40:21 +0100 Subject: [PATCH 536/680] Update ClientBaseHelpers.cpp --- src/Client/ClientBaseHelpers.cpp | 3 --- 1 file changed, 3 deletions(-) diff --git a/src/Client/ClientBaseHelpers.cpp b/src/Client/ClientBaseHelpers.cpp index 555e95f7a25..8bdbab99e13 100644 --- a/src/Client/ClientBaseHelpers.cpp +++ b/src/Client/ClientBaseHelpers.cpp @@ -155,9 +155,6 @@ void highlight(const String & query, std::vector & colors while (token_iterator->type == TokenType::Semicolon) ++token_iterator; - - if (token_iterator->isEnd()) - break; } } catch (...) From 16a670166c9ad6365716d0bccb8320b0f8706efe Mon Sep 17 00:00:00 2001 From: robot-clickhouse Date: Thu, 7 Nov 2024 21:48:11 +0000 Subject: [PATCH 537/680] Update version_date.tsv and changelogs after v24.3.13.40-lts --- docs/changelogs/v24.3.13.40-lts.md | 31 ++++++++++++++++++++++++++++ utils/list-versions/version_date.tsv | 1 + 2 files changed, 32 insertions(+) create mode 100644 docs/changelogs/v24.3.13.40-lts.md diff --git a/docs/changelogs/v24.3.13.40-lts.md b/docs/changelogs/v24.3.13.40-lts.md new file mode 100644 index 00000000000..cec96e16292 --- /dev/null +++ b/docs/changelogs/v24.3.13.40-lts.md @@ -0,0 +1,31 @@ +--- +sidebar_position: 1 +sidebar_label: 2024 +--- + +# 2024 Changelog + +### ClickHouse release v24.3.13.40-lts (7acabd77389) FIXME as compared to v24.3.12.75-lts (7cb5dff8019) + +#### Bug Fix (user-visible misbehavior in an official stable release) +* Backported in [#63976](https://github.com/ClickHouse/ClickHouse/issues/63976): Fix intersect parts when restart after drop range. [#63202](https://github.com/ClickHouse/ClickHouse/pull/63202) ([Han Fei](https://github.com/hanfei1991)). +* Backported in [#71482](https://github.com/ClickHouse/ClickHouse/issues/71482): Fix `Content-Encoding` not sent in some compressed responses. [#64802](https://github.com/ClickHouse/ClickHouse/issues/64802). [#68975](https://github.com/ClickHouse/ClickHouse/pull/68975) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Backported in [#70451](https://github.com/ClickHouse/ClickHouse/issues/70451): Fix vrash during insertion into FixedString column in PostgreSQL engine. [#69584](https://github.com/ClickHouse/ClickHouse/pull/69584) ([Pavel Kruglov](https://github.com/Avogar)). +* Backported in [#70619](https://github.com/ClickHouse/ClickHouse/issues/70619): Fix server segfault on creating a materialized view with two selects and an `INTERSECT`, e.g. `CREATE MATERIALIZED VIEW v0 AS (SELECT 1) INTERSECT (SELECT 1);`. [#70264](https://github.com/ClickHouse/ClickHouse/pull/70264) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Backported in [#70877](https://github.com/ClickHouse/ClickHouse/issues/70877): Fix table creation with `CREATE ... AS table_function()` with database `Replicated` and unavailable table function source on secondary replica. [#70511](https://github.com/ClickHouse/ClickHouse/pull/70511) ([Kseniia Sumarokova](https://github.com/kssenii)). +* Backported in [#70571](https://github.com/ClickHouse/ClickHouse/issues/70571): Ignore all output on async insert with `wait_for_async_insert=1`. Closes [#62644](https://github.com/ClickHouse/ClickHouse/issues/62644). [#70530](https://github.com/ClickHouse/ClickHouse/pull/70530) ([Konstantin Bogdanov](https://github.com/thevar1able)). +* Backported in [#71146](https://github.com/ClickHouse/ClickHouse/issues/71146): Ignore frozen_metadata.txt while traversing shadow directory from system.remote_data_paths. [#70590](https://github.com/ClickHouse/ClickHouse/pull/70590) ([Aleksei Filatov](https://github.com/aalexfvk)). +* Backported in [#70682](https://github.com/ClickHouse/ClickHouse/issues/70682): Fix creation of stateful window functions on misaligned memory. [#70631](https://github.com/ClickHouse/ClickHouse/pull/70631) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#71113](https://github.com/ClickHouse/ClickHouse/issues/71113): `GroupArraySortedData` uses a PODArray with non-POD elements, manually calling constructors and destructors for the elements as needed. But it wasn't careful enough: in two places it forgot to call destructor, in one place it left elements uninitialized if an exception is thrown when deserializing previous elements. Then `GroupArraySortedData`'s destructor called destructors on uninitialized elements and crashed: ``` 2024.10.17 22:58:23.523790 [ 5233 ] {} BaseDaemon: ########## Short fault info ############ 2024.10.17 22:58:23.523834 [ 5233 ] {} BaseDaemon: (version 24.6.1.4609 (official build), build id: 5423339A6571004018D55BBE05D464AFA35E6718, git hash: fa6cdfda8a94890eb19bc7f22f8b0b56292f7a26) (from thread 682) Received signal 11 2024.10.17 22:58:23.523862 [ 5233 ] {} BaseDaemon: Signal description: Segmentation fault 2024.10.17 22:58:23.523883 [ 5233 ] {} BaseDaemon: Address: 0x8f. Access: . Address not mapped to object. 2024.10.17 22:58:23.523908 [ 5233 ] {} BaseDaemon: Stack trace: 0x0000aaaac4b78308 0x0000ffffb7701850 0x0000aaaac0104855 0x0000aaaac01048a0 0x0000aaaac501e84c 0x0000aaaac7c510d0 0x0000aaaac7c4ba20 0x0000aaaac968bbfc 0x0000aaaac968fab0 0x0000aaaac969bf50 0x0000aaaac9b7520c 0x0000aaaac9b74c74 0x0000aaaac9b8a150 0x0000aaaac9b809f0 0x0000aaaac9b80574 0x0000aaaac9b8e364 0x0000aaaac9b8e4fc 0x0000aaaac94f4328 0x0000aaaac94f428c 0x0000aaaac94f7df0 0x0000aaaac98b5a3c 0x0000aaaac950b234 0x0000aaaac49ae264 0x0000aaaac49b1dd0 0x0000aaaac49b0a80 0x0000ffffb755d5c8 0x0000ffffb75c5edc 2024.10.17 22:58:23.523936 [ 5233 ] {} BaseDaemon: ######################################## 2024.10.17 22:58:23.523959 [ 5233 ] {} BaseDaemon: (version 24.6.1.4609 (official build), build id: 5423339A6571004018D55BBE05D464AFA35E6718, git hash: fa6cdfda8a94890eb19bc7f22f8b0b56292f7a26) (from thread 682) (query_id: 6c8a33a2-f45a-4a3b-bd71-ded6a1c9ccd3::202410_534066_534078_2) (query: ) Received signal Segmentation fault (11) 2024.10.17 22:58:23.523977 [ 5233 ] {} BaseDaemon: Address: 0x8f. Access: . Address not mapped to object. 2024.10.17 22:58:23.523993 [ 5233 ] {} BaseDaemon: Stack trace: 0x0000aaaac4b78308 0x0000ffffb7701850 0x0000aaaac0104855 0x0000aaaac01048a0 0x0000aaaac501e84c 0x0000aaaac7c510d0 0x0000aaaac7c4ba20 0x0000aaaac968bbfc 0x0000aaaac968fab0 0x0000aaaac969bf50 0x0000aaaac9b7520c 0x0000aaaac9b74c74 0x0000aaaac9b8a150 0x0000aaaac9b809f0 0x0000aaaac9b80574 0x0000aaaac9b8e364 0x0000aaaac9b8e4fc 0x0000aaaac94f4328 0x0000aaaac94f428c 0x0000aaaac94f7df0 0x0000aaaac98b5a3c 0x0000aaaac950b234 0x0000aaaac49ae264 0x0000aaaac49b1dd0 0x0000aaaac49b0a80 0x0000ffffb755d5c8 0x0000ffffb75c5edc 2024.10.17 22:58:23.524817 [ 5233 ] {} BaseDaemon: 0. signalHandler(int, siginfo_t*, void*) @ 0x000000000c6f8308 2024.10.17 22:58:23.524917 [ 5233 ] {} BaseDaemon: 1. ? @ 0x0000ffffb7701850 2024.10.17 22:58:23.524962 [ 5233 ] {} BaseDaemon: 2. DB::Field::~Field() @ 0x0000000007c84855 2024.10.17 22:58:23.525012 [ 5233 ] {} BaseDaemon: 3. DB::Field::~Field() @ 0x0000000007c848a0 2024.10.17 22:58:23.526626 [ 5233 ] {} BaseDaemon: 4. DB::IAggregateFunctionDataHelper, DB::(anonymous namespace)::GroupArraySorted, DB::Field>>::destroy(char*) const (.5a6a451027f732f9fd91c13f4a13200c) @ 0x000000000cb9e84c 2024.10.17 22:58:23.527322 [ 5233 ] {} BaseDaemon: 5. DB::SerializationAggregateFunction::deserializeBinaryBulk(DB::IColumn&, DB::ReadBuffer&, unsigned long, double) const @ 0x000000000f7d10d0 2024.10.17 22:58:23.528470 [ 5233 ] {} BaseDaemon: 6. DB::ISerialization::deserializeBinaryBulkWithMultipleStreams(COW::immutable_ptr&, unsigned long, DB::ISerialization::DeserializeBinaryBulkSettings&, std::shared_ptr&, std::unordered_map::immutable_ptr, std::hash, std::equal_to, std::allocator::immutable_ptr>>>*) const @ 0x000000000f7cba20 2024.10.17 22:58:23.529213 [ 5233 ] {} BaseDaemon: 7. DB::MergeTreeReaderCompact::readData(DB::NameAndTypePair const&, COW::immutable_ptr&, unsigned long, std::function const&) @ 0x000000001120bbfc 2024.10.17 22:58:23.529277 [ 5233 ] {} BaseDaemon: 8. DB::MergeTreeReaderCompactSingleBuffer::readRows(unsigned long, unsigned long, bool, unsigned long, std::vector::immutable_ptr, std::allocator::immutable_ptr>>&) @ 0x000000001120fab0 2024.10.17 22:58:23.529319 [ 5233 ] {} BaseDaemon: 9. DB::MergeTreeSequentialSource::generate() @ 0x000000001121bf50 2024.10.17 22:58:23.529346 [ 5233 ] {} BaseDaemon: 10. DB::ISource::tryGenerate() @ 0x00000000116f520c 2024.10.17 22:58:23.529653 [ 5233 ] {} BaseDaemon: 11. DB::ISource::work() @ 0x00000000116f4c74 2024.10.17 22:58:23.529679 [ 5233 ] {} BaseDaemon: 12. DB::ExecutionThreadContext::executeTask() @ 0x000000001170a150 2024.10.17 22:58:23.529733 [ 5233 ] {} BaseDaemon: 13. DB::PipelineExecutor::executeStepImpl(unsigned long, std::atomic*) @ 0x00000000117009f0 2024.10.17 22:58:23.529763 [ 5233 ] {} BaseDaemon: 14. DB::PipelineExecutor::executeStep(std::atomic*) @ 0x0000000011700574 2024.10.17 22:58:23.530089 [ 5233 ] {} BaseDaemon: 15. DB::PullingPipelineExecutor::pull(DB::Chunk&) @ 0x000000001170e364 2024.10.17 22:58:23.530277 [ 5233 ] {} BaseDaemon: 16. DB::PullingPipelineExecutor::pull(DB::Block&) @ 0x000000001170e4fc 2024.10.17 22:58:23.530295 [ 5233 ] {} BaseDaemon: 17. DB::MergeTask::ExecuteAndFinalizeHorizontalPart::executeImpl() @ 0x0000000011074328 2024.10.17 22:58:23.530318 [ 5233 ] {} BaseDaemon: 18. DB::MergeTask::ExecuteAndFinalizeHorizontalPart::execute() @ 0x000000001107428c 2024.10.17 22:58:23.530339 [ 5233 ] {} BaseDaemon: 19. DB::MergeTask::execute() @ 0x0000000011077df0 2024.10.17 22:58:23.530362 [ 5233 ] {} BaseDaemon: 20. DB::SharedMergeMutateTaskBase::executeStep() @ 0x0000000011435a3c 2024.10.17 22:58:23.530384 [ 5233 ] {} BaseDaemon: 21. DB::MergeTreeBackgroundExecutor::threadFunction() @ 0x000000001108b234 2024.10.17 22:58:23.530410 [ 5233 ] {} BaseDaemon: 22. ThreadPoolImpl>::worker(std::__list_iterator, void*>) @ 0x000000000c52e264 2024.10.17 22:58:23.530448 [ 5233 ] {} BaseDaemon: 23. void std::__function::__policy_invoker::__call_impl::ThreadFromGlobalPoolImpl>::scheduleImpl(std::function, Priority, std::optional, bool)::'lambda0'()>(void&&)::'lambda'(), void ()>>(std::__function::__policy_storage const*) @ 0x000000000c531dd0 2024.10.17 22:58:23.530476 [ 5233 ] {} BaseDaemon: 24. void* std::__thread_proxy[abi:v15000]>, void ThreadPoolImpl::scheduleImpl(std::function, Priority, std::optional, bool)::'lambda0'()>>(void*) @ 0x000000000c530a80 2024.10.17 22:58:23.530514 [ 5233 ] {} BaseDaemon: 25. ? @ 0x000000000007d5c8 2024.10.17 22:58:23.530534 [ 5233 ] {} BaseDaemon: 26. ? @ 0x00000000000e5edc 2024.10.17 22:58:23.530551 [ 5233 ] {} BaseDaemon: Integrity check of the executable skipped because the reference checksum could not be read. 2024.10.17 22:58:23.531083 [ 5233 ] {} BaseDaemon: Report this error to https://github.com/ClickHouse/ClickHouse/issues 2024.10.17 22:58:23.531294 [ 5233 ] {} BaseDaemon: Changed settings: max_insert_threads = 4, max_threads = 42, use_hedged_requests = false, distributed_foreground_insert = true, alter_sync = 0, enable_memory_bound_merging_of_aggregation_results = true, cluster_for_parallel_replicas = 'default', do_not_merge_across_partitions_select_final = false, log_queries = true, log_queries_probability = 1., max_http_get_redirects = 10, enable_deflate_qpl_codec = false, enable_zstd_qat_codec = false, query_profiler_real_time_period_ns = 0, query_profiler_cpu_time_period_ns = 0, max_bytes_before_external_group_by = 90194313216, max_bytes_before_external_sort = 90194313216, max_memory_usage = 180388626432, backup_restore_keeper_retry_max_backoff_ms = 60000, cancel_http_readonly_queries_on_client_close = true, max_table_size_to_drop = 1000000000000, max_partition_size_to_drop = 1000000000000, default_table_engine = 'ReplicatedMergeTree', mutations_sync = 0, optimize_trivial_insert_select = false, database_replicated_allow_only_replicated_engine = true, cloud_mode = true, cloud_mode_engine = 2, distributed_ddl_output_mode = 'none_only_active', distributed_ddl_entry_format_version = 6, async_insert_max_data_size = 10485760, async_insert_busy_timeout_max_ms = 1000, enable_filesystem_cache_on_write_operations = true, load_marks_asynchronously = true, allow_prefetched_read_pool_for_remote_filesystem = true, filesystem_prefetch_max_memory_usage = 18038862643, filesystem_prefetches_limit = 200, compatibility = '24.6', insert_keeper_max_retries = 20, allow_experimental_materialized_postgresql_table = false, date_time_input_format = 'best_effort' ```. [#70820](https://github.com/ClickHouse/ClickHouse/pull/70820) ([Michael Kolupaev](https://github.com/al13n321)). +* Backported in [#70990](https://github.com/ClickHouse/ClickHouse/issues/70990): Fix a logical error due to negative zeros in the two-level hash table. This closes [#70973](https://github.com/ClickHouse/ClickHouse/issues/70973). [#70979](https://github.com/ClickHouse/ClickHouse/pull/70979) ([Alexey Milovidov](https://github.com/alexey-milovidov)). +* Backported in [#71246](https://github.com/ClickHouse/ClickHouse/issues/71246): Fixed named sessions not being closed and hanging on forever under certain circumstances. [#70998](https://github.com/ClickHouse/ClickHouse/pull/70998) ([Márcio Martins](https://github.com/marcio-absmartly)). +* Backported in [#71371](https://github.com/ClickHouse/ClickHouse/issues/71371): Add try/catch to data parts destructors to avoid terminate. [#71364](https://github.com/ClickHouse/ClickHouse/pull/71364) ([alesapin](https://github.com/alesapin)). +* Backported in [#71594](https://github.com/ClickHouse/ClickHouse/issues/71594): Prevent crash in SortCursor with 0 columns (old analyzer). [#71494](https://github.com/ClickHouse/ClickHouse/pull/71494) ([Raúl Marín](https://github.com/Algunenano)). + +#### NOT FOR CHANGELOG / INSIGNIFICANT + +* Backported in [#71022](https://github.com/ClickHouse/ClickHouse/issues/71022): Fix dropping of file cache in CHECK query in case of enabled transactions. [#69256](https://github.com/ClickHouse/ClickHouse/pull/69256) ([Anton Popov](https://github.com/CurtizJ)). +* Backported in [#70384](https://github.com/ClickHouse/ClickHouse/issues/70384): CI: Enable Integration Tests for backport PRs. [#70329](https://github.com/ClickHouse/ClickHouse/pull/70329) ([Max Kainov](https://github.com/maxknv)). +* Backported in [#70538](https://github.com/ClickHouse/ClickHouse/issues/70538): Remove slow poll() logs in keeper. [#70508](https://github.com/ClickHouse/ClickHouse/pull/70508) ([Raúl Marín](https://github.com/Algunenano)). +* Backported in [#70971](https://github.com/ClickHouse/ClickHouse/issues/70971): Limiting logging some lines about configs. [#70879](https://github.com/ClickHouse/ClickHouse/pull/70879) ([Yarik Briukhovetskyi](https://github.com/yariks5s)). + diff --git a/utils/list-versions/version_date.tsv b/utils/list-versions/version_date.tsv index cf28db5d49a..fab562a8cbb 100644 --- a/utils/list-versions/version_date.tsv +++ b/utils/list-versions/version_date.tsv @@ -31,6 +31,7 @@ v24.4.4.113-stable 2024-08-02 v24.4.3.25-stable 2024-06-14 v24.4.2.141-stable 2024-06-07 v24.4.1.2088-stable 2024-05-01 +v24.3.13.40-lts 2024-11-07 v24.3.12.75-lts 2024-10-08 v24.3.11.7-lts 2024-09-06 v24.3.10.33-lts 2024-09-03 From f71b00c5136bec4fe40393a45310c1f85a50e5d0 Mon Sep 17 00:00:00 2001 From: Konstantin Bogdanov Date: Thu, 7 Nov 2024 22:52:27 +0100 Subject: [PATCH 538/680] Lint --- docs/changelogs/v24.3.13.40-lts.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/changelogs/v24.3.13.40-lts.md b/docs/changelogs/v24.3.13.40-lts.md index cec96e16292..bce45e88710 100644 --- a/docs/changelogs/v24.3.13.40-lts.md +++ b/docs/changelogs/v24.3.13.40-lts.md @@ -16,7 +16,7 @@ sidebar_label: 2024 * Backported in [#70571](https://github.com/ClickHouse/ClickHouse/issues/70571): Ignore all output on async insert with `wait_for_async_insert=1`. Closes [#62644](https://github.com/ClickHouse/ClickHouse/issues/62644). [#70530](https://github.com/ClickHouse/ClickHouse/pull/70530) ([Konstantin Bogdanov](https://github.com/thevar1able)). * Backported in [#71146](https://github.com/ClickHouse/ClickHouse/issues/71146): Ignore frozen_metadata.txt while traversing shadow directory from system.remote_data_paths. [#70590](https://github.com/ClickHouse/ClickHouse/pull/70590) ([Aleksei Filatov](https://github.com/aalexfvk)). * Backported in [#70682](https://github.com/ClickHouse/ClickHouse/issues/70682): Fix creation of stateful window functions on misaligned memory. [#70631](https://github.com/ClickHouse/ClickHouse/pull/70631) ([Raúl Marín](https://github.com/Algunenano)). -* Backported in [#71113](https://github.com/ClickHouse/ClickHouse/issues/71113): `GroupArraySortedData` uses a PODArray with non-POD elements, manually calling constructors and destructors for the elements as needed. But it wasn't careful enough: in two places it forgot to call destructor, in one place it left elements uninitialized if an exception is thrown when deserializing previous elements. Then `GroupArraySortedData`'s destructor called destructors on uninitialized elements and crashed: ``` 2024.10.17 22:58:23.523790 [ 5233 ] {} BaseDaemon: ########## Short fault info ############ 2024.10.17 22:58:23.523834 [ 5233 ] {} BaseDaemon: (version 24.6.1.4609 (official build), build id: 5423339A6571004018D55BBE05D464AFA35E6718, git hash: fa6cdfda8a94890eb19bc7f22f8b0b56292f7a26) (from thread 682) Received signal 11 2024.10.17 22:58:23.523862 [ 5233 ] {} BaseDaemon: Signal description: Segmentation fault 2024.10.17 22:58:23.523883 [ 5233 ] {} BaseDaemon: Address: 0x8f. Access: . Address not mapped to object. 2024.10.17 22:58:23.523908 [ 5233 ] {} BaseDaemon: Stack trace: 0x0000aaaac4b78308 0x0000ffffb7701850 0x0000aaaac0104855 0x0000aaaac01048a0 0x0000aaaac501e84c 0x0000aaaac7c510d0 0x0000aaaac7c4ba20 0x0000aaaac968bbfc 0x0000aaaac968fab0 0x0000aaaac969bf50 0x0000aaaac9b7520c 0x0000aaaac9b74c74 0x0000aaaac9b8a150 0x0000aaaac9b809f0 0x0000aaaac9b80574 0x0000aaaac9b8e364 0x0000aaaac9b8e4fc 0x0000aaaac94f4328 0x0000aaaac94f428c 0x0000aaaac94f7df0 0x0000aaaac98b5a3c 0x0000aaaac950b234 0x0000aaaac49ae264 0x0000aaaac49b1dd0 0x0000aaaac49b0a80 0x0000ffffb755d5c8 0x0000ffffb75c5edc 2024.10.17 22:58:23.523936 [ 5233 ] {} BaseDaemon: ######################################## 2024.10.17 22:58:23.523959 [ 5233 ] {} BaseDaemon: (version 24.6.1.4609 (official build), build id: 5423339A6571004018D55BBE05D464AFA35E6718, git hash: fa6cdfda8a94890eb19bc7f22f8b0b56292f7a26) (from thread 682) (query_id: 6c8a33a2-f45a-4a3b-bd71-ded6a1c9ccd3::202410_534066_534078_2) (query: ) Received signal Segmentation fault (11) 2024.10.17 22:58:23.523977 [ 5233 ] {} BaseDaemon: Address: 0x8f. Access: . Address not mapped to object. 2024.10.17 22:58:23.523993 [ 5233 ] {} BaseDaemon: Stack trace: 0x0000aaaac4b78308 0x0000ffffb7701850 0x0000aaaac0104855 0x0000aaaac01048a0 0x0000aaaac501e84c 0x0000aaaac7c510d0 0x0000aaaac7c4ba20 0x0000aaaac968bbfc 0x0000aaaac968fab0 0x0000aaaac969bf50 0x0000aaaac9b7520c 0x0000aaaac9b74c74 0x0000aaaac9b8a150 0x0000aaaac9b809f0 0x0000aaaac9b80574 0x0000aaaac9b8e364 0x0000aaaac9b8e4fc 0x0000aaaac94f4328 0x0000aaaac94f428c 0x0000aaaac94f7df0 0x0000aaaac98b5a3c 0x0000aaaac950b234 0x0000aaaac49ae264 0x0000aaaac49b1dd0 0x0000aaaac49b0a80 0x0000ffffb755d5c8 0x0000ffffb75c5edc 2024.10.17 22:58:23.524817 [ 5233 ] {} BaseDaemon: 0. signalHandler(int, siginfo_t*, void*) @ 0x000000000c6f8308 2024.10.17 22:58:23.524917 [ 5233 ] {} BaseDaemon: 1. ? @ 0x0000ffffb7701850 2024.10.17 22:58:23.524962 [ 5233 ] {} BaseDaemon: 2. DB::Field::~Field() @ 0x0000000007c84855 2024.10.17 22:58:23.525012 [ 5233 ] {} BaseDaemon: 3. DB::Field::~Field() @ 0x0000000007c848a0 2024.10.17 22:58:23.526626 [ 5233 ] {} BaseDaemon: 4. DB::IAggregateFunctionDataHelper, DB::(anonymous namespace)::GroupArraySorted, DB::Field>>::destroy(char*) const (.5a6a451027f732f9fd91c13f4a13200c) @ 0x000000000cb9e84c 2024.10.17 22:58:23.527322 [ 5233 ] {} BaseDaemon: 5. DB::SerializationAggregateFunction::deserializeBinaryBulk(DB::IColumn&, DB::ReadBuffer&, unsigned long, double) const @ 0x000000000f7d10d0 2024.10.17 22:58:23.528470 [ 5233 ] {} BaseDaemon: 6. DB::ISerialization::deserializeBinaryBulkWithMultipleStreams(COW::immutable_ptr&, unsigned long, DB::ISerialization::DeserializeBinaryBulkSettings&, std::shared_ptr&, std::unordered_map::immutable_ptr, std::hash, std::equal_to, std::allocator::immutable_ptr>>>*) const @ 0x000000000f7cba20 2024.10.17 22:58:23.529213 [ 5233 ] {} BaseDaemon: 7. DB::MergeTreeReaderCompact::readData(DB::NameAndTypePair const&, COW::immutable_ptr&, unsigned long, std::function const&) @ 0x000000001120bbfc 2024.10.17 22:58:23.529277 [ 5233 ] {} BaseDaemon: 8. DB::MergeTreeReaderCompactSingleBuffer::readRows(unsigned long, unsigned long, bool, unsigned long, std::vector::immutable_ptr, std::allocator::immutable_ptr>>&) @ 0x000000001120fab0 2024.10.17 22:58:23.529319 [ 5233 ] {} BaseDaemon: 9. DB::MergeTreeSequentialSource::generate() @ 0x000000001121bf50 2024.10.17 22:58:23.529346 [ 5233 ] {} BaseDaemon: 10. DB::ISource::tryGenerate() @ 0x00000000116f520c 2024.10.17 22:58:23.529653 [ 5233 ] {} BaseDaemon: 11. DB::ISource::work() @ 0x00000000116f4c74 2024.10.17 22:58:23.529679 [ 5233 ] {} BaseDaemon: 12. DB::ExecutionThreadContext::executeTask() @ 0x000000001170a150 2024.10.17 22:58:23.529733 [ 5233 ] {} BaseDaemon: 13. DB::PipelineExecutor::executeStepImpl(unsigned long, std::atomic*) @ 0x00000000117009f0 2024.10.17 22:58:23.529763 [ 5233 ] {} BaseDaemon: 14. DB::PipelineExecutor::executeStep(std::atomic*) @ 0x0000000011700574 2024.10.17 22:58:23.530089 [ 5233 ] {} BaseDaemon: 15. DB::PullingPipelineExecutor::pull(DB::Chunk&) @ 0x000000001170e364 2024.10.17 22:58:23.530277 [ 5233 ] {} BaseDaemon: 16. DB::PullingPipelineExecutor::pull(DB::Block&) @ 0x000000001170e4fc 2024.10.17 22:58:23.530295 [ 5233 ] {} BaseDaemon: 17. DB::MergeTask::ExecuteAndFinalizeHorizontalPart::executeImpl() @ 0x0000000011074328 2024.10.17 22:58:23.530318 [ 5233 ] {} BaseDaemon: 18. DB::MergeTask::ExecuteAndFinalizeHorizontalPart::execute() @ 0x000000001107428c 2024.10.17 22:58:23.530339 [ 5233 ] {} BaseDaemon: 19. DB::MergeTask::execute() @ 0x0000000011077df0 2024.10.17 22:58:23.530362 [ 5233 ] {} BaseDaemon: 20. DB::SharedMergeMutateTaskBase::executeStep() @ 0x0000000011435a3c 2024.10.17 22:58:23.530384 [ 5233 ] {} BaseDaemon: 21. DB::MergeTreeBackgroundExecutor::threadFunction() @ 0x000000001108b234 2024.10.17 22:58:23.530410 [ 5233 ] {} BaseDaemon: 22. ThreadPoolImpl>::worker(std::__list_iterator, void*>) @ 0x000000000c52e264 2024.10.17 22:58:23.530448 [ 5233 ] {} BaseDaemon: 23. void std::__function::__policy_invoker::__call_impl::ThreadFromGlobalPoolImpl>::scheduleImpl(std::function, Priority, std::optional, bool)::'lambda0'()>(void&&)::'lambda'(), void ()>>(std::__function::__policy_storage const*) @ 0x000000000c531dd0 2024.10.17 22:58:23.530476 [ 5233 ] {} BaseDaemon: 24. void* std::__thread_proxy[abi:v15000]>, void ThreadPoolImpl::scheduleImpl(std::function, Priority, std::optional, bool)::'lambda0'()>>(void*) @ 0x000000000c530a80 2024.10.17 22:58:23.530514 [ 5233 ] {} BaseDaemon: 25. ? @ 0x000000000007d5c8 2024.10.17 22:58:23.530534 [ 5233 ] {} BaseDaemon: 26. ? @ 0x00000000000e5edc 2024.10.17 22:58:23.530551 [ 5233 ] {} BaseDaemon: Integrity check of the executable skipped because the reference checksum could not be read. 2024.10.17 22:58:23.531083 [ 5233 ] {} BaseDaemon: Report this error to https://github.com/ClickHouse/ClickHouse/issues 2024.10.17 22:58:23.531294 [ 5233 ] {} BaseDaemon: Changed settings: max_insert_threads = 4, max_threads = 42, use_hedged_requests = false, distributed_foreground_insert = true, alter_sync = 0, enable_memory_bound_merging_of_aggregation_results = true, cluster_for_parallel_replicas = 'default', do_not_merge_across_partitions_select_final = false, log_queries = true, log_queries_probability = 1., max_http_get_redirects = 10, enable_deflate_qpl_codec = false, enable_zstd_qat_codec = false, query_profiler_real_time_period_ns = 0, query_profiler_cpu_time_period_ns = 0, max_bytes_before_external_group_by = 90194313216, max_bytes_before_external_sort = 90194313216, max_memory_usage = 180388626432, backup_restore_keeper_retry_max_backoff_ms = 60000, cancel_http_readonly_queries_on_client_close = true, max_table_size_to_drop = 1000000000000, max_partition_size_to_drop = 1000000000000, default_table_engine = 'ReplicatedMergeTree', mutations_sync = 0, optimize_trivial_insert_select = false, database_replicated_allow_only_replicated_engine = true, cloud_mode = true, cloud_mode_engine = 2, distributed_ddl_output_mode = 'none_only_active', distributed_ddl_entry_format_version = 6, async_insert_max_data_size = 10485760, async_insert_busy_timeout_max_ms = 1000, enable_filesystem_cache_on_write_operations = true, load_marks_asynchronously = true, allow_prefetched_read_pool_for_remote_filesystem = true, filesystem_prefetch_max_memory_usage = 18038862643, filesystem_prefetches_limit = 200, compatibility = '24.6', insert_keeper_max_retries = 20, allow_experimental_materialized_postgresql_table = false, date_time_input_format = 'best_effort' ```. [#70820](https://github.com/ClickHouse/ClickHouse/pull/70820) ([Michael Kolupaev](https://github.com/al13n321)). +* Backported in [#71113](https://github.com/ClickHouse/ClickHouse/issues/71113): Fix a crash and a leak in AggregateFunctionGroupArraySorted. [#70820](https://github.com/ClickHouse/ClickHouse/pull/70820) ([Michael Kolupaev](https://github.com/al13n321)). * Backported in [#70990](https://github.com/ClickHouse/ClickHouse/issues/70990): Fix a logical error due to negative zeros in the two-level hash table. This closes [#70973](https://github.com/ClickHouse/ClickHouse/issues/70973). [#70979](https://github.com/ClickHouse/ClickHouse/pull/70979) ([Alexey Milovidov](https://github.com/alexey-milovidov)). * Backported in [#71246](https://github.com/ClickHouse/ClickHouse/issues/71246): Fixed named sessions not being closed and hanging on forever under certain circumstances. [#70998](https://github.com/ClickHouse/ClickHouse/pull/70998) ([Márcio Martins](https://github.com/marcio-absmartly)). * Backported in [#71371](https://github.com/ClickHouse/ClickHouse/issues/71371): Add try/catch to data parts destructors to avoid terminate. [#71364](https://github.com/ClickHouse/ClickHouse/pull/71364) ([alesapin](https://github.com/alesapin)). From dc9e1e047b5cf27dde9dd8b0184cdcdd006202ed Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Thu, 7 Nov 2024 23:18:39 +0100 Subject: [PATCH 539/680] Fix tests --- tests/queries/0_stateless/02751_multiquery_with_argument.sh | 2 +- tests/queries/0_stateless/02771_multiple_query_arguments.sh | 2 +- .../02800_clickhouse_local_default_settings.reference | 1 + 3 files changed, 3 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/02751_multiquery_with_argument.sh b/tests/queries/0_stateless/02751_multiquery_with_argument.sh index 4021194656b..4378786c145 100755 --- a/tests/queries/0_stateless/02751_multiquery_with_argument.sh +++ b/tests/queries/0_stateless/02751_multiquery_with_argument.sh @@ -9,7 +9,7 @@ $CLICKHOUSE_LOCAL "SELECT 101;" $CLICKHOUSE_LOCAL "SELECT 102;SELECT 103;" # Invalid SQL. -$CLICKHOUSE_LOCAL "SELECT 200; S" 2>&1 | grep -o 'Syntax error' +$CLICKHOUSE_LOCAL --implicit-select 0 "SELECT 200; S" 2>&1 | grep -o 'Syntax error' $CLICKHOUSE_LOCAL "; SELECT 201;" 2>&1 | grep -o 'Empty query' $CLICKHOUSE_LOCAL "; S; SELECT 202" 2>&1 | grep -o 'Empty query' diff --git a/tests/queries/0_stateless/02771_multiple_query_arguments.sh b/tests/queries/0_stateless/02771_multiple_query_arguments.sh index ae6e23eb61a..fcc1394573a 100755 --- a/tests/queries/0_stateless/02771_multiple_query_arguments.sh +++ b/tests/queries/0_stateless/02771_multiple_query_arguments.sh @@ -18,4 +18,4 @@ $CLICKHOUSE_LOCAL --query "SELECT 202;" --query "SELECT 202;" $CLICKHOUSE_LOCAL --query "SELECT 303" --query "SELECT 303; SELECT 303" $CLICKHOUSE_LOCAL --query "" --query "" $CLICKHOUSE_LOCAL --query "SELECT 303" --query 2>&1 | grep -o 'Bad arguments' -$CLICKHOUSE_LOCAL --query "SELECT 303" --query "SELE" 2>&1 | grep -o 'Syntax error' +$CLICKHOUSE_LOCAL --implicit-select 0 --query "SELECT 303" --query "SELE" 2>&1 | grep -o 'Syntax error' diff --git a/tests/queries/0_stateless/02800_clickhouse_local_default_settings.reference b/tests/queries/0_stateless/02800_clickhouse_local_default_settings.reference index 0f18d1a3897..54c6f7ce397 100644 --- a/tests/queries/0_stateless/02800_clickhouse_local_default_settings.reference +++ b/tests/queries/0_stateless/02800_clickhouse_local_default_settings.reference @@ -1,2 +1,3 @@ allow_introspection_functions 1 storage_file_read_method mmap +implicit_select 1 From 6054f43000c645a6a470d06e8d935cf792da3011 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 8 Nov 2024 00:14:25 +0100 Subject: [PATCH 540/680] Make Vertical format prettier --- src/Formats/PrettyFormatHelpers.cpp | 102 ++ src/Formats/PrettyFormatHelpers.h | 18 + src/Interpreters/InterpreterSystemQuery.cpp | 4 +- .../Formats/Impl/PrettyBlockOutputFormat.cpp | 94 +- .../Formats/Impl/PrettyBlockOutputFormat.h | 5 +- .../Impl/PrettyCompactBlockOutputFormat.cpp | 4 +- .../Impl/PrettySpaceBlockOutputFormat.cpp | 4 +- .../Formats/Impl/VerticalRowOutputFormat.cpp | 28 +- .../Formats/Impl/VerticalRowOutputFormat.h | 3 + .../03268_vertical_pretty_numbers.reference | 1532 +++++++++++++++++ .../03268_vertical_pretty_numbers.sql | 11 + 11 files changed, 1707 insertions(+), 98 deletions(-) create mode 100644 src/Formats/PrettyFormatHelpers.cpp create mode 100644 src/Formats/PrettyFormatHelpers.h create mode 100644 tests/queries/0_stateless/03268_vertical_pretty_numbers.reference create mode 100644 tests/queries/0_stateless/03268_vertical_pretty_numbers.sql diff --git a/src/Formats/PrettyFormatHelpers.cpp b/src/Formats/PrettyFormatHelpers.cpp new file mode 100644 index 00000000000..6e2af036651 --- /dev/null +++ b/src/Formats/PrettyFormatHelpers.cpp @@ -0,0 +1,102 @@ +#include +#include +#include +#include +#include + + +namespace DB +{ + +void writeReadableNumberTipIfSingleValue(WriteBuffer & out, const Chunk & chunk, const FormatSettings & settings, bool color) +{ + if (chunk.getNumRows() == 1 && chunk.getNumColumns() == 1) + writeReadableNumberTip(out, *chunk.getColumns()[0], 0, settings, color); +} + +void writeReadableNumberTip(WriteBuffer & out, const IColumn & column, size_t row, const FormatSettings & settings, bool color) +{ + if (column.isNullAt(row)) + return; + + auto value = column.getFloat64(row); + auto threshold = settings.pretty.output_format_pretty_single_large_number_tip_threshold; + + if (threshold && isFinite(value) && abs(value) > threshold) + { + if (color) + writeCString("\033[90m", out); + writeCString(" -- ", out); + formatReadableQuantity(value, out, 2); + if (color) + writeCString("\033[0m", out); + } +} + + +String highlightDigitGroups(String source) +{ + if (source.size() <= 4) + return source; + + bool is_regular_number = true; + size_t num_digits_before_decimal = 0; + for (auto c : source) + { + if (c == '-' || c == ' ') + continue; + if (c == '.') + break; + if (c >= '0' && c <= '9') + { + ++num_digits_before_decimal; + } + else + { + is_regular_number = false; + break; + } + } + + if (!is_regular_number || num_digits_before_decimal <= 4) + return source; + + String result; + size_t size = source.size(); + result.reserve(2 * size); + + bool before_decimal = true; + size_t digit_num = 0; + for (size_t i = 0; i < size; ++i) + { + auto c = source[i]; + if (before_decimal && c >= '0' && c <= '9') + { + ++digit_num; + size_t offset = num_digits_before_decimal - digit_num; + if (offset && offset % 3 == 0) + { + result += "\033[4m"; + result += c; + result += "\033[0m"; + } + else + { + result += c; + } + } + else if (c == '.') + { + before_decimal = false; + result += c; + } + else + { + result += c; + } + } + + return result; +} + +} diff --git a/src/Formats/PrettyFormatHelpers.h b/src/Formats/PrettyFormatHelpers.h new file mode 100644 index 00000000000..72ab5e3c2a0 --- /dev/null +++ b/src/Formats/PrettyFormatHelpers.h @@ -0,0 +1,18 @@ +#include + +namespace DB +{ + +class Chunk; +class IColumn; +class WriteBuffer; +struct FormatSettings; + +/// Prints text describing the number in the form of: -- 12.34 million +void writeReadableNumberTip(WriteBuffer & out, const IColumn & column, size_t row, const FormatSettings & settings, bool color); +void writeReadableNumberTipIfSingleValue(WriteBuffer & out, const Chunk & chunk, const FormatSettings & settings, bool color); + +/// Underscores digit groups related to thousands using terminal ANSI escape sequences. +String highlightDigitGroups(String source); + +} diff --git a/src/Interpreters/InterpreterSystemQuery.cpp b/src/Interpreters/InterpreterSystemQuery.cpp index 4c875026ace..b651bfb245e 100644 --- a/src/Interpreters/InterpreterSystemQuery.cpp +++ b/src/Interpreters/InterpreterSystemQuery.cpp @@ -795,9 +795,9 @@ BlockIO InterpreterSystemQuery::execute() case Type::WAIT_FAILPOINT: { getContext()->checkAccess(AccessType::SYSTEM_FAILPOINT); - LOG_TRACE(log, "waiting for failpoint {}", query.fail_point_name); + LOG_TRACE(log, "Waiting for failpoint {}", query.fail_point_name); FailPointInjection::pauseFailPoint(query.fail_point_name); - LOG_TRACE(log, "finished failpoint {}", query.fail_point_name); + LOG_TRACE(log, "Finished waiting for failpoint {}", query.fail_point_name); break; } case Type::RESET_COVERAGE: diff --git a/src/Processors/Formats/Impl/PrettyBlockOutputFormat.cpp b/src/Processors/Formats/Impl/PrettyBlockOutputFormat.cpp index ff1a048029d..e8b55ea423b 100644 --- a/src/Processors/Formats/Impl/PrettyBlockOutputFormat.cpp +++ b/src/Processors/Formats/Impl/PrettyBlockOutputFormat.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include #include @@ -352,7 +353,8 @@ void PrettyBlockOutputFormat::writeChunk(const Chunk & chunk, PortKind port_kind } writeCString(grid_symbols.bar, out); - writeReadableNumberTip(chunk); + if (readable_number_tip) + writeReadableNumberTipIfSingleValue(out, chunk, format_settings, color); writeCString("\n", out); } @@ -392,72 +394,6 @@ void PrettyBlockOutputFormat::writeChunk(const Chunk & chunk, PortKind port_kind } -static String highlightDigitGroups(String source) -{ - if (source.size() <= 4) - return source; - - bool is_regular_number = true; - size_t num_digits_before_decimal = 0; - for (auto c : source) - { - if (c == '-' || c == ' ') - continue; - if (c == '.') - break; - if (c >= '0' && c <= '9') - { - ++num_digits_before_decimal; - } - else - { - is_regular_number = false; - break; - } - } - - if (!is_regular_number || num_digits_before_decimal <= 4) - return source; - - String result; - size_t size = source.size(); - result.reserve(2 * size); - - bool before_decimal = true; - size_t digit_num = 0; - for (size_t i = 0; i < size; ++i) - { - auto c = source[i]; - if (before_decimal && c >= '0' && c <= '9') - { - ++digit_num; - size_t offset = num_digits_before_decimal - digit_num; - if (offset && offset % 3 == 0) - { - result += "\033[4m"; - result += c; - result += "\033[0m"; - } - else - { - result += c; - } - } - else if (c == '.') - { - before_decimal = false; - result += c; - } - else - { - result += c; - } - } - - return result; -} - - void PrettyBlockOutputFormat::writeValueWithPadding( const IColumn & column, const ISerialization & serialization, size_t row_num, size_t value_width, size_t pad_to_width, size_t cut_to_width, bool align_right, bool is_number) @@ -553,30 +489,6 @@ void PrettyBlockOutputFormat::writeSuffix() } } -void PrettyBlockOutputFormat::writeReadableNumberTip(const Chunk & chunk) -{ - const auto & columns = chunk.getColumns(); - auto is_single_number = readable_number_tip && chunk.getNumRows() == 1 && chunk.getNumColumns() == 1; - if (!is_single_number) - return; - - if (columns[0]->isNullAt(0)) - return; - - auto value = columns[0]->getFloat64(0); - auto threshold = format_settings.pretty.output_format_pretty_single_large_number_tip_threshold; - - if (threshold && isFinite(value) && abs(value) > threshold) - { - if (color) - writeCString("\033[90m", out); - writeCString(" -- ", out); - formatReadableQuantity(value, out, 2); - if (color) - writeCString("\033[0m", out); - } -} - void registerOutputFormatPretty(FormatFactory & factory) { registerPrettyFormatWithNoEscapesAndMonoBlock(factory, "Pretty"); diff --git a/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h b/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h index 698efecd4b2..824a2fd2e6f 100644 --- a/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h +++ b/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h @@ -38,7 +38,6 @@ protected: virtual void writeChunk(const Chunk & chunk, PortKind port_kind); void writeMonoChunkIfNeeded(); void writeSuffix() override; - void writeReadableNumberTip(const Chunk & chunk); void onRowsReadBeforeUpdate() override { total_rows = getRowsReadBefore(); } @@ -57,8 +56,10 @@ protected: bool color; -private: +protected: bool readable_number_tip = false; + +private: bool mono_block; /// For mono_block == true only Chunk mono_chunk; diff --git a/src/Processors/Formats/Impl/PrettyCompactBlockOutputFormat.cpp b/src/Processors/Formats/Impl/PrettyCompactBlockOutputFormat.cpp index 57ec23e7e3b..1e4f784bc71 100644 --- a/src/Processors/Formats/Impl/PrettyCompactBlockOutputFormat.cpp +++ b/src/Processors/Formats/Impl/PrettyCompactBlockOutputFormat.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include @@ -189,7 +190,8 @@ void PrettyCompactBlockOutputFormat::writeRow( } writeCString(grid_symbols.bar, out); - writeReadableNumberTip(chunk); + if (readable_number_tip) + writeReadableNumberTipIfSingleValue(out, chunk, format_settings, color); writeCString("\n", out); } diff --git a/src/Processors/Formats/Impl/PrettySpaceBlockOutputFormat.cpp b/src/Processors/Formats/Impl/PrettySpaceBlockOutputFormat.cpp index 0a594b54b12..5b481099e41 100644 --- a/src/Processors/Formats/Impl/PrettySpaceBlockOutputFormat.cpp +++ b/src/Processors/Formats/Impl/PrettySpaceBlockOutputFormat.cpp @@ -1,4 +1,5 @@ #include +#include #include #include #include @@ -102,7 +103,8 @@ void PrettySpaceBlockOutputFormat::writeChunk(const Chunk & chunk, PortKind port writeValueWithPadding( *columns[column], *serializations[column], row, cur_width, max_widths[column], cut_to_width, type.shouldAlignRightInPrettyFormats(), isNumber(type)); } - writeReadableNumberTip(chunk); + if (readable_number_tip) + writeReadableNumberTipIfSingleValue(out, chunk, format_settings, color); writeChar('\n', out); } diff --git a/src/Processors/Formats/Impl/VerticalRowOutputFormat.cpp b/src/Processors/Formats/Impl/VerticalRowOutputFormat.cpp index 4852af9f0c8..7b0135b3ae4 100644 --- a/src/Processors/Formats/Impl/VerticalRowOutputFormat.cpp +++ b/src/Processors/Formats/Impl/VerticalRowOutputFormat.cpp @@ -4,7 +4,10 @@ #include #include #include +#include #include +#include +#include namespace DB @@ -14,6 +17,8 @@ VerticalRowOutputFormat::VerticalRowOutputFormat( WriteBuffer & out_, const Block & header_, const FormatSettings & format_settings_) : IRowOutputFormat(header_, out_), format_settings(format_settings_) { + color = format_settings.pretty.color == 1 || (format_settings.pretty.color == 2 && format_settings.is_writing_to_terminal); + const auto & sample = getPort(PortKind::Main).getHeader(); size_t columns = sample.columns(); @@ -31,6 +36,7 @@ VerticalRowOutputFormat::VerticalRowOutputFormat( } names_and_paddings.resize(columns); + is_number.resize(columns); for (size_t i = 0; i < columns; ++i) { WriteBufferFromString buf(names_and_paddings[i]); @@ -42,6 +48,7 @@ VerticalRowOutputFormat::VerticalRowOutputFormat( { size_t new_size = max_name_width - name_widths[i] + names_and_paddings[i].size(); names_and_paddings[i].resize(new_size, ' '); + is_number[i] = isNumber(removeNullable(recursiveRemoveLowCardinality(sample.getByPosition(i).type))); } } @@ -61,7 +68,26 @@ void VerticalRowOutputFormat::writeField(const IColumn & column, const ISerializ void VerticalRowOutputFormat::writeValue(const IColumn & column, const ISerialization & serialization, size_t row_num) const { - serialization.serializeText(column, row_num, out, format_settings); + if (color && format_settings.pretty.highlight_digit_groups && is_number[field_number]) + { + String serialized_value; + { + WriteBufferFromString buf(serialized_value); + serialization.serializeText(column, row_num, buf, format_settings); + } + + /// Highlight groups of thousands. + serialized_value = highlightDigitGroups(serialized_value); + out.write(serialized_value.data(), serialized_value.size()); + } + else + { + serialization.serializeText(column, row_num, out, format_settings); + } + + /// Write a tip. + if (is_number[field_number]) + writeReadableNumberTip(out, column, row_num, format_settings, color); } diff --git a/src/Processors/Formats/Impl/VerticalRowOutputFormat.h b/src/Processors/Formats/Impl/VerticalRowOutputFormat.h index 5870c3503fc..6fe79adc9be 100644 --- a/src/Processors/Formats/Impl/VerticalRowOutputFormat.h +++ b/src/Processors/Formats/Impl/VerticalRowOutputFormat.h @@ -56,6 +56,9 @@ private: using NamesAndPaddings = std::vector; NamesAndPaddings names_and_paddings; + + std::vector is_number; + bool color; }; } diff --git a/tests/queries/0_stateless/03268_vertical_pretty_numbers.reference b/tests/queries/0_stateless/03268_vertical_pretty_numbers.reference new file mode 100644 index 00000000000..397e9145798 --- /dev/null +++ b/tests/queries/0_stateless/03268_vertical_pretty_numbers.reference @@ -0,0 +1,1532 @@ +Row 1: +────── +exp2(number): 1 +exp10(number): 1 +concat('test', number): test0 + +Row 2: +────── +exp2(number): 2 -- 2.00 +exp10(number): 10 -- 10.00 +concat('test', number): test1 + +Row 3: +────── +exp2(number): 4 -- 4.00 +exp10(number): 100 -- 100.00 +concat('test', number): test2 + +Row 4: +────── +exp2(number): 8 -- 8.00 +exp10(number): 1000 -- 1.00 thousand +concat('test', number): test3 + +Row 5: +────── +exp2(number): 16 -- 16.00 +exp10(number): 10000 -- 10.00 thousand +concat('test', number): test4 + +Row 6: +────── +exp2(number): 32 -- 32.00 +exp10(number): 100000 -- 100.00 thousand +concat('test', number): test5 + +Row 7: +────── +exp2(number): 64 -- 64.00 +exp10(number): 1000000 -- 1.00 million +concat('test', number): test6 + +Row 8: +────── +exp2(number): 128 -- 128.00 +exp10(number): 10000000 -- 10.00 million +concat('test', number): test7 + +Row 9: +─────── +exp2(number): 256 -- 256.00 +exp10(number): 100000000 -- 100.00 million +concat('test', number): test8 + +Row 10: +─────── +exp2(number): 512 -- 512.00 +exp10(number): 1000000000 -- 1.00 billion +concat('test', number): test9 + +Row 11: +─────── +exp2(number): 1024 -- 1.02 thousand +exp10(number): 10000000000 -- 10.00 billion +concat('test', number): test10 + +Row 12: +─────── +exp2(number): 2048 -- 2.05 thousand +exp10(number): 100000000000 -- 100.00 billion +concat('test', number): test11 + +Row 13: +─────── +exp2(number): 4096 -- 4.10 thousand +exp10(number): 1000000000000 -- 1.00 trillion +concat('test', number): test12 + +Row 14: +─────── +exp2(number): 8192 -- 8.19 thousand +exp10(number): 10000000000000 -- 10.00 trillion +concat('test', number): test13 + +Row 15: +─────── +exp2(number): 16384 -- 16.38 thousand +exp10(number): 100000000000000 -- 100.00 trillion +concat('test', number): test14 + +Row 16: +─────── +exp2(number): 32768 -- 32.77 thousand +exp10(number): 1000000000000000 -- 1.00 quadrillion +concat('test', number): test15 + +Row 17: +─────── +exp2(number): 65536 -- 65.54 thousand +exp10(number): 10000000000000000 -- 10.00 quadrillion +concat('test', number): test16 + +Row 18: +─────── +exp2(number): 131072 -- 131.07 thousand +exp10(number): 100000000000000000 -- 100.00 quadrillion +concat('test', number): test17 + +Row 19: +─────── +exp2(number): 262144 -- 262.14 thousand +exp10(number): 1000000000000000000 -- 1.00 quintillion +concat('test', number): test18 + +Row 20: +─────── +exp2(number): 524288 -- 524.29 thousand +exp10(number): 10000000000000000000 -- 10.00 quintillion +concat('test', number): test19 + +Row 21: +─────── +exp2(number): 1048576 -- 1.05 million +exp10(number): 100000000000000000000 -- 100.00 quintillion +concat('test', number): test20 + +Row 22: +─────── +exp2(number): 2097152 -- 2.10 million +exp10(number): 1e21 -- 1.00 sextillion +concat('test', number): test21 + +Row 23: +─────── +exp2(number): 4194304 -- 4.19 million +exp10(number): 1e22 -- 10.00 sextillion +concat('test', number): test22 + +Row 24: +─────── +exp2(number): 8388608 -- 8.39 million +exp10(number): 1e23 -- 100.00 sextillion +concat('test', number): test23 + +Row 25: +─────── +exp2(number): 16777216 -- 16.78 million +exp10(number): 1e24 -- 1.00 septillion +concat('test', number): test24 + +Row 26: +─────── +exp2(number): 33554432 -- 33.55 million +exp10(number): 1e25 -- 10.00 septillion +concat('test', number): test25 + +Row 27: +─────── +exp2(number): 67108864 -- 67.11 million +exp10(number): 1e26 -- 100.00 septillion +concat('test', number): test26 + +Row 28: +─────── +exp2(number): 134217728 -- 134.22 million +exp10(number): 1e27 -- 1.00 octillion +concat('test', number): test27 + +Row 29: +─────── +exp2(number): 268435456 -- 268.44 million +exp10(number): 1e28 -- 10.00 octillion +concat('test', number): test28 + +Row 30: +─────── +exp2(number): 536870912 -- 536.87 million +exp10(number): 1e29 -- 100.00 octillion +concat('test', number): test29 + +Row 31: +─────── +exp2(number): 1073741824 -- 1.07 billion +exp10(number): 1e30 -- 1.00 nonillion +concat('test', number): test30 + +Row 32: +─────── +exp2(number): 2147483648 -- 2.15 billion +exp10(number): 1e31 -- 10.00 nonillion +concat('test', number): test31 + +Row 33: +─────── +exp2(number): 4294967296 -- 4.29 billion +exp10(number): 1e32 -- 100.00 nonillion +concat('test', number): test32 + +Row 34: +─────── +exp2(number): 8589934592 -- 8.59 billion +exp10(number): 1e33 -- 1000.00 nonillion +concat('test', number): test33 + +Row 35: +─────── +exp2(number): 17179869184 -- 17.18 billion +exp10(number): 1e34 -- 10.00 decillion +concat('test', number): test34 + +Row 36: +─────── +exp2(number): 34359738368 -- 34.36 billion +exp10(number): 1e35 -- 100.00 decillion +concat('test', number): test35 + +Row 37: +─────── +exp2(number): 68719476736 -- 68.72 billion +exp10(number): 1e36 -- 1.00 undecillion +concat('test', number): test36 + +Row 38: +─────── +exp2(number): 137438953472 -- 137.44 billion +exp10(number): 1e37 -- 10.00 undecillion +concat('test', number): test37 + +Row 39: +─────── +exp2(number): 274877906944 -- 274.88 billion +exp10(number): 1e38 -- 100.00 undecillion +concat('test', number): test38 + +Row 40: +─────── +exp2(number): 549755813888 -- 549.76 billion +exp10(number): 1e39 -- 1000.00 undecillion +concat('test', number): test39 + +Row 41: +─────── +exp2(number): 1099511627776 -- 1.10 trillion +exp10(number): 1e40 -- 10.00 duodecillion +concat('test', number): test40 + +Row 42: +─────── +exp2(number): 2199023255552 -- 2.20 trillion +exp10(number): 1e41 -- 100.00 duodecillion +concat('test', number): test41 + +Row 43: +─────── +exp2(number): 4398046511104 -- 4.40 trillion +exp10(number): 1e42 -- 1.00 tredecillion +concat('test', number): test42 + +Row 44: +─────── +exp2(number): 8796093022208 -- 8.80 trillion +exp10(number): 1e43 -- 10.00 tredecillion +concat('test', number): test43 + +Row 45: +─────── +exp2(number): 17592186044416 -- 17.59 trillion +exp10(number): 1e44 -- 100.00 tredecillion +concat('test', number): test44 + +Row 46: +─────── +exp2(number): 35184372088832 -- 35.18 trillion +exp10(number): 1e45 -- 1000.00 tredecillion +concat('test', number): test45 + +Row 47: +─────── +exp2(number): 70368744177664 -- 70.37 trillion +exp10(number): 1e46 -- 10.00 quattuordecillion +concat('test', number): test46 + +Row 48: +─────── +exp2(number): 140737488355328 -- 140.74 trillion +exp10(number): 1e47 -- 100.00 quattuordecillion +concat('test', number): test47 + +Row 49: +─────── +exp2(number): 281474976710656 -- 281.47 trillion +exp10(number): 1e48 -- 1.00 quindecillion +concat('test', number): test48 + +Row 50: +─────── +exp2(number): 562949953421312 -- 562.95 trillion +exp10(number): 1e49 -- 10.00 quindecillion +concat('test', number): test49 + +Row 51: +─────── +exp2(number): 1125899906842624 -- 1.13 quadrillion +exp10(number): 1e50 -- 100.00 quindecillion +concat('test', number): test50 + +Row 52: +─────── +exp2(number): 2251799813685248 -- 2.25 quadrillion +exp10(number): 1e51 -- 1.00 sexdecillion +concat('test', number): test51 + +Row 53: +─────── +exp2(number): 4503599627370496 -- 4.50 quadrillion +exp10(number): 1e52 -- 10.00 sexdecillion +concat('test', number): test52 + +Row 54: +─────── +exp2(number): 9007199254740992 -- 9.01 quadrillion +exp10(number): 1e53 -- 100.00 sexdecillion +concat('test', number): test53 + +Row 55: +─────── +exp2(number): 18014398509481984 -- 18.01 quadrillion +exp10(number): 1e54 -- 1.00 septendecillion +concat('test', number): test54 + +Row 56: +─────── +exp2(number): 36028797018963970 -- 36.03 quadrillion +exp10(number): 1e55 -- 10.00 septendecillion +concat('test', number): test55 + +Row 57: +─────── +exp2(number): 72057594037927940 -- 72.06 quadrillion +exp10(number): 1e56 -- 100.00 septendecillion +concat('test', number): test56 + +Row 58: +─────── +exp2(number): 144115188075855870 -- 144.12 quadrillion +exp10(number): 1e57 -- 1.00 octodecillion +concat('test', number): test57 + +Row 59: +─────── +exp2(number): 288230376151711740 -- 288.23 quadrillion +exp10(number): 1e58 -- 10.00 octodecillion +concat('test', number): test58 + +Row 60: +─────── +exp2(number): 576460752303423500 -- 576.46 quadrillion +exp10(number): 1e59 -- 100.00 octodecillion +concat('test', number): test59 + +Row 61: +─────── +exp2(number): 1152921504606847000 -- 1.15 quintillion +exp10(number): 1e60 -- 1000.00 octodecillion +concat('test', number): test60 + +Row 62: +─────── +exp2(number): 2305843009213694000 -- 2.31 quintillion +exp10(number): 1e61 -- 10.00 novemdecillion +concat('test', number): test61 + +Row 63: +─────── +exp2(number): 4611686018427388000 -- 4.61 quintillion +exp10(number): 1e62 -- 100.00 novemdecillion +concat('test', number): test62 + +Row 64: +─────── +exp2(number): 9223372036854776000 -- 9.22 quintillion +exp10(number): 1e63 -- 1.00 vigintillion +concat('test', number): test63 +Row 1: +────── +exp2(number): 1 +exp10(number): 1 +concat('test', number): test0 + +Row 2: +────── +exp2(number): 2 -- 2.00 +exp10(number): 10 -- 10.00 +concat('test', number): test1 + +Row 3: +────── +exp2(number): 4 -- 4.00 +exp10(number): 100 -- 100.00 +concat('test', number): test2 + +Row 4: +────── +exp2(number): 8 -- 8.00 +exp10(number): 1000 -- 1.00 thousand +concat('test', number): test3 + +Row 5: +────── +exp2(number): 16 -- 16.00 +exp10(number): 10000 -- 10.00 thousand +concat('test', number): test4 + +Row 6: +────── +exp2(number): 32 -- 32.00 +exp10(number): 100000 -- 100.00 thousand +concat('test', number): test5 + +Row 7: +────── +exp2(number): 64 -- 64.00 +exp10(number): 1000000 -- 1.00 million +concat('test', number): test6 + +Row 8: +────── +exp2(number): 128 -- 128.00 +exp10(number): 10000000 -- 10.00 million +concat('test', number): test7 + +Row 9: +─────── +exp2(number): 256 -- 256.00 +exp10(number): 100000000 -- 100.00 million +concat('test', number): test8 + +Row 10: +─────── +exp2(number): 512 -- 512.00 +exp10(number): 1000000000 -- 1.00 billion +concat('test', number): test9 + +Row 11: +─────── +exp2(number): 1024 -- 1.02 thousand +exp10(number): 10000000000 -- 10.00 billion +concat('test', number): test10 + +Row 12: +─────── +exp2(number): 2048 -- 2.05 thousand +exp10(number): 100000000000 -- 100.00 billion +concat('test', number): test11 + +Row 13: +─────── +exp2(number): 4096 -- 4.10 thousand +exp10(number): 1000000000000 -- 1.00 trillion +concat('test', number): test12 + +Row 14: +─────── +exp2(number): 8192 -- 8.19 thousand +exp10(number): 10000000000000 -- 10.00 trillion +concat('test', number): test13 + +Row 15: +─────── +exp2(number): 16384 -- 16.38 thousand +exp10(number): 100000000000000 -- 100.00 trillion +concat('test', number): test14 + +Row 16: +─────── +exp2(number): 32768 -- 32.77 thousand +exp10(number): 1000000000000000 -- 1.00 quadrillion +concat('test', number): test15 + +Row 17: +─────── +exp2(number): 65536 -- 65.54 thousand +exp10(number): 10000000000000000 -- 10.00 quadrillion +concat('test', number): test16 + +Row 18: +─────── +exp2(number): 131072 -- 131.07 thousand +exp10(number): 100000000000000000 -- 100.00 quadrillion +concat('test', number): test17 + +Row 19: +─────── +exp2(number): 262144 -- 262.14 thousand +exp10(number): 1000000000000000000 -- 1.00 quintillion +concat('test', number): test18 + +Row 20: +─────── +exp2(number): 524288 -- 524.29 thousand +exp10(number): 10000000000000000000 -- 10.00 quintillion +concat('test', number): test19 + +Row 21: +─────── +exp2(number): 1048576 -- 1.05 million +exp10(number): 100000000000000000000 -- 100.00 quintillion +concat('test', number): test20 + +Row 22: +─────── +exp2(number): 2097152 -- 2.10 million +exp10(number): 1e21 -- 1.00 sextillion +concat('test', number): test21 + +Row 23: +─────── +exp2(number): 4194304 -- 4.19 million +exp10(number): 1e22 -- 10.00 sextillion +concat('test', number): test22 + +Row 24: +─────── +exp2(number): 8388608 -- 8.39 million +exp10(number): 1e23 -- 100.00 sextillion +concat('test', number): test23 + +Row 25: +─────── +exp2(number): 16777216 -- 16.78 million +exp10(number): 1e24 -- 1.00 septillion +concat('test', number): test24 + +Row 26: +─────── +exp2(number): 33554432 -- 33.55 million +exp10(number): 1e25 -- 10.00 septillion +concat('test', number): test25 + +Row 27: +─────── +exp2(number): 67108864 -- 67.11 million +exp10(number): 1e26 -- 100.00 septillion +concat('test', number): test26 + +Row 28: +─────── +exp2(number): 134217728 -- 134.22 million +exp10(number): 1e27 -- 1.00 octillion +concat('test', number): test27 + +Row 29: +─────── +exp2(number): 268435456 -- 268.44 million +exp10(number): 1e28 -- 10.00 octillion +concat('test', number): test28 + +Row 30: +─────── +exp2(number): 536870912 -- 536.87 million +exp10(number): 1e29 -- 100.00 octillion +concat('test', number): test29 + +Row 31: +─────── +exp2(number): 1073741824 -- 1.07 billion +exp10(number): 1e30 -- 1.00 nonillion +concat('test', number): test30 + +Row 32: +─────── +exp2(number): 2147483648 -- 2.15 billion +exp10(number): 1e31 -- 10.00 nonillion +concat('test', number): test31 + +Row 33: +─────── +exp2(number): 4294967296 -- 4.29 billion +exp10(number): 1e32 -- 100.00 nonillion +concat('test', number): test32 + +Row 34: +─────── +exp2(number): 8589934592 -- 8.59 billion +exp10(number): 1e33 -- 1000.00 nonillion +concat('test', number): test33 + +Row 35: +─────── +exp2(number): 17179869184 -- 17.18 billion +exp10(number): 1e34 -- 10.00 decillion +concat('test', number): test34 + +Row 36: +─────── +exp2(number): 34359738368 -- 34.36 billion +exp10(number): 1e35 -- 100.00 decillion +concat('test', number): test35 + +Row 37: +─────── +exp2(number): 68719476736 -- 68.72 billion +exp10(number): 1e36 -- 1.00 undecillion +concat('test', number): test36 + +Row 38: +─────── +exp2(number): 137438953472 -- 137.44 billion +exp10(number): 1e37 -- 10.00 undecillion +concat('test', number): test37 + +Row 39: +─────── +exp2(number): 274877906944 -- 274.88 billion +exp10(number): 1e38 -- 100.00 undecillion +concat('test', number): test38 + +Row 40: +─────── +exp2(number): 549755813888 -- 549.76 billion +exp10(number): 1e39 -- 1000.00 undecillion +concat('test', number): test39 + +Row 41: +─────── +exp2(number): 1099511627776 -- 1.10 trillion +exp10(number): 1e40 -- 10.00 duodecillion +concat('test', number): test40 + +Row 42: +─────── +exp2(number): 2199023255552 -- 2.20 trillion +exp10(number): 1e41 -- 100.00 duodecillion +concat('test', number): test41 + +Row 43: +─────── +exp2(number): 4398046511104 -- 4.40 trillion +exp10(number): 1e42 -- 1.00 tredecillion +concat('test', number): test42 + +Row 44: +─────── +exp2(number): 8796093022208 -- 8.80 trillion +exp10(number): 1e43 -- 10.00 tredecillion +concat('test', number): test43 + +Row 45: +─────── +exp2(number): 17592186044416 -- 17.59 trillion +exp10(number): 1e44 -- 100.00 tredecillion +concat('test', number): test44 + +Row 46: +─────── +exp2(number): 35184372088832 -- 35.18 trillion +exp10(number): 1e45 -- 1000.00 tredecillion +concat('test', number): test45 + +Row 47: +─────── +exp2(number): 70368744177664 -- 70.37 trillion +exp10(number): 1e46 -- 10.00 quattuordecillion +concat('test', number): test46 + +Row 48: +─────── +exp2(number): 140737488355328 -- 140.74 trillion +exp10(number): 1e47 -- 100.00 quattuordecillion +concat('test', number): test47 + +Row 49: +─────── +exp2(number): 281474976710656 -- 281.47 trillion +exp10(number): 1e48 -- 1.00 quindecillion +concat('test', number): test48 + +Row 50: +─────── +exp2(number): 562949953421312 -- 562.95 trillion +exp10(number): 1e49 -- 10.00 quindecillion +concat('test', number): test49 + +Row 51: +─────── +exp2(number): 1125899906842624 -- 1.13 quadrillion +exp10(number): 1e50 -- 100.00 quindecillion +concat('test', number): test50 + +Row 52: +─────── +exp2(number): 2251799813685248 -- 2.25 quadrillion +exp10(number): 1e51 -- 1.00 sexdecillion +concat('test', number): test51 + +Row 53: +─────── +exp2(number): 4503599627370496 -- 4.50 quadrillion +exp10(number): 1e52 -- 10.00 sexdecillion +concat('test', number): test52 + +Row 54: +─────── +exp2(number): 9007199254740992 -- 9.01 quadrillion +exp10(number): 1e53 -- 100.00 sexdecillion +concat('test', number): test53 + +Row 55: +─────── +exp2(number): 18014398509481984 -- 18.01 quadrillion +exp10(number): 1e54 -- 1.00 septendecillion +concat('test', number): test54 + +Row 56: +─────── +exp2(number): 36028797018963970 -- 36.03 quadrillion +exp10(number): 1e55 -- 10.00 septendecillion +concat('test', number): test55 + +Row 57: +─────── +exp2(number): 72057594037927940 -- 72.06 quadrillion +exp10(number): 1e56 -- 100.00 septendecillion +concat('test', number): test56 + +Row 58: +─────── +exp2(number): 144115188075855870 -- 144.12 quadrillion +exp10(number): 1e57 -- 1.00 octodecillion +concat('test', number): test57 + +Row 59: +─────── +exp2(number): 288230376151711740 -- 288.23 quadrillion +exp10(number): 1e58 -- 10.00 octodecillion +concat('test', number): test58 + +Row 60: +─────── +exp2(number): 576460752303423500 -- 576.46 quadrillion +exp10(number): 1e59 -- 100.00 octodecillion +concat('test', number): test59 + +Row 61: +─────── +exp2(number): 1152921504606847000 -- 1.15 quintillion +exp10(number): 1e60 -- 1000.00 octodecillion +concat('test', number): test60 + +Row 62: +─────── +exp2(number): 2305843009213694000 -- 2.31 quintillion +exp10(number): 1e61 -- 10.00 novemdecillion +concat('test', number): test61 + +Row 63: +─────── +exp2(number): 4611686018427388000 -- 4.61 quintillion +exp10(number): 1e62 -- 100.00 novemdecillion +concat('test', number): test62 + +Row 64: +─────── +exp2(number): 9223372036854776000 -- 9.22 quintillion +exp10(number): 1e63 -- 1.00 vigintillion +concat('test', number): test63 +Row 1: +────── +exp2(number): 1 +exp10(number): 1 +concat('test', number): test0 + +Row 2: +────── +exp2(number): 2 -- 2.00 +exp10(number): 10 -- 10.00 +concat('test', number): test1 + +Row 3: +────── +exp2(number): 4 -- 4.00 +exp10(number): 100 -- 100.00 +concat('test', number): test2 + +Row 4: +────── +exp2(number): 8 -- 8.00 +exp10(number): 1000 -- 1.00 thousand +concat('test', number): test3 + +Row 5: +────── +exp2(number): 16 -- 16.00 +exp10(number): 10000 -- 10.00 thousand +concat('test', number): test4 + +Row 6: +────── +exp2(number): 32 -- 32.00 +exp10(number): 100000 -- 100.00 thousand +concat('test', number): test5 + +Row 7: +────── +exp2(number): 64 -- 64.00 +exp10(number): 1000000 -- 1.00 million +concat('test', number): test6 + +Row 8: +────── +exp2(number): 128 -- 128.00 +exp10(number): 10000000 -- 10.00 million +concat('test', number): test7 + +Row 9: +─────── +exp2(number): 256 -- 256.00 +exp10(number): 100000000 -- 100.00 million +concat('test', number): test8 + +Row 10: +─────── +exp2(number): 512 -- 512.00 +exp10(number): 1000000000 -- 1.00 billion +concat('test', number): test9 + +Row 11: +─────── +exp2(number): 1024 -- 1.02 thousand +exp10(number): 10000000000 -- 10.00 billion +concat('test', number): test10 + +Row 12: +─────── +exp2(number): 2048 -- 2.05 thousand +exp10(number): 100000000000 -- 100.00 billion +concat('test', number): test11 + +Row 13: +─────── +exp2(number): 4096 -- 4.10 thousand +exp10(number): 1000000000000 -- 1.00 trillion +concat('test', number): test12 + +Row 14: +─────── +exp2(number): 8192 -- 8.19 thousand +exp10(number): 10000000000000 -- 10.00 trillion +concat('test', number): test13 + +Row 15: +─────── +exp2(number): 16384 -- 16.38 thousand +exp10(number): 100000000000000 -- 100.00 trillion +concat('test', number): test14 + +Row 16: +─────── +exp2(number): 32768 -- 32.77 thousand +exp10(number): 1000000000000000 -- 1.00 quadrillion +concat('test', number): test15 + +Row 17: +─────── +exp2(number): 65536 -- 65.54 thousand +exp10(number): 10000000000000000 -- 10.00 quadrillion +concat('test', number): test16 + +Row 18: +─────── +exp2(number): 131072 -- 131.07 thousand +exp10(number): 100000000000000000 -- 100.00 quadrillion +concat('test', number): test17 + +Row 19: +─────── +exp2(number): 262144 -- 262.14 thousand +exp10(number): 1000000000000000000 -- 1.00 quintillion +concat('test', number): test18 + +Row 20: +─────── +exp2(number): 524288 -- 524.29 thousand +exp10(number): 10000000000000000000 -- 10.00 quintillion +concat('test', number): test19 + +Row 21: +─────── +exp2(number): 1048576 -- 1.05 million +exp10(number): 100000000000000000000 -- 100.00 quintillion +concat('test', number): test20 + +Row 22: +─────── +exp2(number): 2097152 -- 2.10 million +exp10(number): 1e21 -- 1.00 sextillion +concat('test', number): test21 + +Row 23: +─────── +exp2(number): 4194304 -- 4.19 million +exp10(number): 1e22 -- 10.00 sextillion +concat('test', number): test22 + +Row 24: +─────── +exp2(number): 8388608 -- 8.39 million +exp10(number): 1e23 -- 100.00 sextillion +concat('test', number): test23 + +Row 25: +─────── +exp2(number): 16777216 -- 16.78 million +exp10(number): 1e24 -- 1.00 septillion +concat('test', number): test24 + +Row 26: +─────── +exp2(number): 33554432 -- 33.55 million +exp10(number): 1e25 -- 10.00 septillion +concat('test', number): test25 + +Row 27: +─────── +exp2(number): 67108864 -- 67.11 million +exp10(number): 1e26 -- 100.00 septillion +concat('test', number): test26 + +Row 28: +─────── +exp2(number): 134217728 -- 134.22 million +exp10(number): 1e27 -- 1.00 octillion +concat('test', number): test27 + +Row 29: +─────── +exp2(number): 268435456 -- 268.44 million +exp10(number): 1e28 -- 10.00 octillion +concat('test', number): test28 + +Row 30: +─────── +exp2(number): 536870912 -- 536.87 million +exp10(number): 1e29 -- 100.00 octillion +concat('test', number): test29 + +Row 31: +─────── +exp2(number): 1073741824 -- 1.07 billion +exp10(number): 1e30 -- 1.00 nonillion +concat('test', number): test30 + +Row 32: +─────── +exp2(number): 2147483648 -- 2.15 billion +exp10(number): 1e31 -- 10.00 nonillion +concat('test', number): test31 + +Row 33: +─────── +exp2(number): 4294967296 -- 4.29 billion +exp10(number): 1e32 -- 100.00 nonillion +concat('test', number): test32 + +Row 34: +─────── +exp2(number): 8589934592 -- 8.59 billion +exp10(number): 1e33 -- 1000.00 nonillion +concat('test', number): test33 + +Row 35: +─────── +exp2(number): 17179869184 -- 17.18 billion +exp10(number): 1e34 -- 10.00 decillion +concat('test', number): test34 + +Row 36: +─────── +exp2(number): 34359738368 -- 34.36 billion +exp10(number): 1e35 -- 100.00 decillion +concat('test', number): test35 + +Row 37: +─────── +exp2(number): 68719476736 -- 68.72 billion +exp10(number): 1e36 -- 1.00 undecillion +concat('test', number): test36 + +Row 38: +─────── +exp2(number): 137438953472 -- 137.44 billion +exp10(number): 1e37 -- 10.00 undecillion +concat('test', number): test37 + +Row 39: +─────── +exp2(number): 274877906944 -- 274.88 billion +exp10(number): 1e38 -- 100.00 undecillion +concat('test', number): test38 + +Row 40: +─────── +exp2(number): 549755813888 -- 549.76 billion +exp10(number): 1e39 -- 1000.00 undecillion +concat('test', number): test39 + +Row 41: +─────── +exp2(number): 1099511627776 -- 1.10 trillion +exp10(number): 1e40 -- 10.00 duodecillion +concat('test', number): test40 + +Row 42: +─────── +exp2(number): 2199023255552 -- 2.20 trillion +exp10(number): 1e41 -- 100.00 duodecillion +concat('test', number): test41 + +Row 43: +─────── +exp2(number): 4398046511104 -- 4.40 trillion +exp10(number): 1e42 -- 1.00 tredecillion +concat('test', number): test42 + +Row 44: +─────── +exp2(number): 8796093022208 -- 8.80 trillion +exp10(number): 1e43 -- 10.00 tredecillion +concat('test', number): test43 + +Row 45: +─────── +exp2(number): 17592186044416 -- 17.59 trillion +exp10(number): 1e44 -- 100.00 tredecillion +concat('test', number): test44 + +Row 46: +─────── +exp2(number): 35184372088832 -- 35.18 trillion +exp10(number): 1e45 -- 1000.00 tredecillion +concat('test', number): test45 + +Row 47: +─────── +exp2(number): 70368744177664 -- 70.37 trillion +exp10(number): 1e46 -- 10.00 quattuordecillion +concat('test', number): test46 + +Row 48: +─────── +exp2(number): 140737488355328 -- 140.74 trillion +exp10(number): 1e47 -- 100.00 quattuordecillion +concat('test', number): test47 + +Row 49: +─────── +exp2(number): 281474976710656 -- 281.47 trillion +exp10(number): 1e48 -- 1.00 quindecillion +concat('test', number): test48 + +Row 50: +─────── +exp2(number): 562949953421312 -- 562.95 trillion +exp10(number): 1e49 -- 10.00 quindecillion +concat('test', number): test49 + +Row 51: +─────── +exp2(number): 1125899906842624 -- 1.13 quadrillion +exp10(number): 1e50 -- 100.00 quindecillion +concat('test', number): test50 + +Row 52: +─────── +exp2(number): 2251799813685248 -- 2.25 quadrillion +exp10(number): 1e51 -- 1.00 sexdecillion +concat('test', number): test51 + +Row 53: +─────── +exp2(number): 4503599627370496 -- 4.50 quadrillion +exp10(number): 1e52 -- 10.00 sexdecillion +concat('test', number): test52 + +Row 54: +─────── +exp2(number): 9007199254740992 -- 9.01 quadrillion +exp10(number): 1e53 -- 100.00 sexdecillion +concat('test', number): test53 + +Row 55: +─────── +exp2(number): 18014398509481984 -- 18.01 quadrillion +exp10(number): 1e54 -- 1.00 septendecillion +concat('test', number): test54 + +Row 56: +─────── +exp2(number): 36028797018963970 -- 36.03 quadrillion +exp10(number): 1e55 -- 10.00 septendecillion +concat('test', number): test55 + +Row 57: +─────── +exp2(number): 72057594037927940 -- 72.06 quadrillion +exp10(number): 1e56 -- 100.00 septendecillion +concat('test', number): test56 + +Row 58: +─────── +exp2(number): 144115188075855870 -- 144.12 quadrillion +exp10(number): 1e57 -- 1.00 octodecillion +concat('test', number): test57 + +Row 59: +─────── +exp2(number): 288230376151711740 -- 288.23 quadrillion +exp10(number): 1e58 -- 10.00 octodecillion +concat('test', number): test58 + +Row 60: +─────── +exp2(number): 576460752303423500 -- 576.46 quadrillion +exp10(number): 1e59 -- 100.00 octodecillion +concat('test', number): test59 + +Row 61: +─────── +exp2(number): 1152921504606847000 -- 1.15 quintillion +exp10(number): 1e60 -- 1000.00 octodecillion +concat('test', number): test60 + +Row 62: +─────── +exp2(number): 2305843009213694000 -- 2.31 quintillion +exp10(number): 1e61 -- 10.00 novemdecillion +concat('test', number): test61 + +Row 63: +─────── +exp2(number): 4611686018427388000 -- 4.61 quintillion +exp10(number): 1e62 -- 100.00 novemdecillion +concat('test', number): test62 + +Row 64: +─────── +exp2(number): 9223372036854776000 -- 9.22 quintillion +exp10(number): 1e63 -- 1.00 vigintillion +concat('test', number): test63 +Row 1: +────── +exp2(number): 1 +exp10(number): 1 +concat('test', number): test0 + +Row 2: +────── +exp2(number): 2 +exp10(number): 10 +concat('test', number): test1 + +Row 3: +────── +exp2(number): 4 +exp10(number): 100 +concat('test', number): test2 + +Row 4: +────── +exp2(number): 8 +exp10(number): 1000 +concat('test', number): test3 + +Row 5: +────── +exp2(number): 16 +exp10(number): 10000 +concat('test', number): test4 + +Row 6: +────── +exp2(number): 32 +exp10(number): 100000 +concat('test', number): test5 + +Row 7: +────── +exp2(number): 64 +exp10(number): 1000000 +concat('test', number): test6 + +Row 8: +────── +exp2(number): 128 +exp10(number): 10000000 +concat('test', number): test7 + +Row 9: +─────── +exp2(number): 256 +exp10(number): 100000000 +concat('test', number): test8 + +Row 10: +─────── +exp2(number): 512 +exp10(number): 1000000000 +concat('test', number): test9 + +Row 11: +─────── +exp2(number): 1024 +exp10(number): 10000000000 +concat('test', number): test10 + +Row 12: +─────── +exp2(number): 2048 +exp10(number): 100000000000 +concat('test', number): test11 + +Row 13: +─────── +exp2(number): 4096 +exp10(number): 1000000000000 +concat('test', number): test12 + +Row 14: +─────── +exp2(number): 8192 +exp10(number): 10000000000000 +concat('test', number): test13 + +Row 15: +─────── +exp2(number): 16384 +exp10(number): 100000000000000 +concat('test', number): test14 + +Row 16: +─────── +exp2(number): 32768 +exp10(number): 1000000000000000 +concat('test', number): test15 + +Row 17: +─────── +exp2(number): 65536 +exp10(number): 10000000000000000 +concat('test', number): test16 + +Row 18: +─────── +exp2(number): 131072 +exp10(number): 100000000000000000 +concat('test', number): test17 + +Row 19: +─────── +exp2(number): 262144 +exp10(number): 1000000000000000000 +concat('test', number): test18 + +Row 20: +─────── +exp2(number): 524288 +exp10(number): 10000000000000000000 +concat('test', number): test19 + +Row 21: +─────── +exp2(number): 1048576 +exp10(number): 100000000000000000000 +concat('test', number): test20 + +Row 22: +─────── +exp2(number): 2097152 +exp10(number): 1e21 +concat('test', number): test21 + +Row 23: +─────── +exp2(number): 4194304 +exp10(number): 1e22 +concat('test', number): test22 + +Row 24: +─────── +exp2(number): 8388608 +exp10(number): 1e23 +concat('test', number): test23 + +Row 25: +─────── +exp2(number): 16777216 +exp10(number): 1e24 +concat('test', number): test24 + +Row 26: +─────── +exp2(number): 33554432 +exp10(number): 1e25 +concat('test', number): test25 + +Row 27: +─────── +exp2(number): 67108864 +exp10(number): 1e26 +concat('test', number): test26 + +Row 28: +─────── +exp2(number): 134217728 +exp10(number): 1e27 +concat('test', number): test27 + +Row 29: +─────── +exp2(number): 268435456 +exp10(number): 1e28 +concat('test', number): test28 + +Row 30: +─────── +exp2(number): 536870912 +exp10(number): 1e29 +concat('test', number): test29 + +Row 31: +─────── +exp2(number): 1073741824 +exp10(number): 1e30 +concat('test', number): test30 + +Row 32: +─────── +exp2(number): 2147483648 +exp10(number): 1e31 +concat('test', number): test31 + +Row 33: +─────── +exp2(number): 4294967296 +exp10(number): 1e32 +concat('test', number): test32 + +Row 34: +─────── +exp2(number): 8589934592 +exp10(number): 1e33 +concat('test', number): test33 + +Row 35: +─────── +exp2(number): 17179869184 +exp10(number): 1e34 +concat('test', number): test34 + +Row 36: +─────── +exp2(number): 34359738368 +exp10(number): 1e35 +concat('test', number): test35 + +Row 37: +─────── +exp2(number): 68719476736 +exp10(number): 1e36 +concat('test', number): test36 + +Row 38: +─────── +exp2(number): 137438953472 +exp10(number): 1e37 +concat('test', number): test37 + +Row 39: +─────── +exp2(number): 274877906944 +exp10(number): 1e38 +concat('test', number): test38 + +Row 40: +─────── +exp2(number): 549755813888 +exp10(number): 1e39 +concat('test', number): test39 + +Row 41: +─────── +exp2(number): 1099511627776 +exp10(number): 1e40 +concat('test', number): test40 + +Row 42: +─────── +exp2(number): 2199023255552 +exp10(number): 1e41 +concat('test', number): test41 + +Row 43: +─────── +exp2(number): 4398046511104 +exp10(number): 1e42 +concat('test', number): test42 + +Row 44: +─────── +exp2(number): 8796093022208 +exp10(number): 1e43 +concat('test', number): test43 + +Row 45: +─────── +exp2(number): 17592186044416 +exp10(number): 1e44 +concat('test', number): test44 + +Row 46: +─────── +exp2(number): 35184372088832 +exp10(number): 1e45 +concat('test', number): test45 + +Row 47: +─────── +exp2(number): 70368744177664 +exp10(number): 1e46 +concat('test', number): test46 + +Row 48: +─────── +exp2(number): 140737488355328 +exp10(number): 1e47 +concat('test', number): test47 + +Row 49: +─────── +exp2(number): 281474976710656 +exp10(number): 1e48 +concat('test', number): test48 + +Row 50: +─────── +exp2(number): 562949953421312 +exp10(number): 1e49 +concat('test', number): test49 + +Row 51: +─────── +exp2(number): 1125899906842624 +exp10(number): 1e50 +concat('test', number): test50 + +Row 52: +─────── +exp2(number): 2251799813685248 +exp10(number): 1e51 +concat('test', number): test51 + +Row 53: +─────── +exp2(number): 4503599627370496 +exp10(number): 1e52 +concat('test', number): test52 + +Row 54: +─────── +exp2(number): 9007199254740992 +exp10(number): 1e53 +concat('test', number): test53 + +Row 55: +─────── +exp2(number): 18014398509481984 +exp10(number): 1e54 +concat('test', number): test54 + +Row 56: +─────── +exp2(number): 36028797018963970 +exp10(number): 1e55 +concat('test', number): test55 + +Row 57: +─────── +exp2(number): 72057594037927940 +exp10(number): 1e56 +concat('test', number): test56 + +Row 58: +─────── +exp2(number): 144115188075855870 +exp10(number): 1e57 +concat('test', number): test57 + +Row 59: +─────── +exp2(number): 288230376151711740 +exp10(number): 1e58 +concat('test', number): test58 + +Row 60: +─────── +exp2(number): 576460752303423500 +exp10(number): 1e59 +concat('test', number): test59 + +Row 61: +─────── +exp2(number): 1152921504606847000 +exp10(number): 1e60 +concat('test', number): test60 + +Row 62: +─────── +exp2(number): 2305843009213694000 +exp10(number): 1e61 +concat('test', number): test61 + +Row 63: +─────── +exp2(number): 4611686018427388000 +exp10(number): 1e62 +concat('test', number): test62 + +Row 64: +─────── +exp2(number): 9223372036854776000 +exp10(number): 1e63 +concat('test', number): test63 diff --git a/tests/queries/0_stateless/03268_vertical_pretty_numbers.sql b/tests/queries/0_stateless/03268_vertical_pretty_numbers.sql new file mode 100644 index 00000000000..0462134ed63 --- /dev/null +++ b/tests/queries/0_stateless/03268_vertical_pretty_numbers.sql @@ -0,0 +1,11 @@ +SET output_format_pretty_color = 1, output_format_pretty_highlight_digit_groups = 1, output_format_pretty_single_large_number_tip_threshold = 1; +SELECT exp2(number), exp10(number), 'test'||number FROM numbers(64) FORMAT Vertical; + +SET output_format_pretty_color = 0, output_format_pretty_highlight_digit_groups = 1, output_format_pretty_single_large_number_tip_threshold = 1; +SELECT exp2(number), exp10(number), 'test'||number FROM numbers(64) FORMAT Vertical; + +SET output_format_pretty_color = 1, output_format_pretty_highlight_digit_groups = 0, output_format_pretty_single_large_number_tip_threshold = 1; +SELECT exp2(number), exp10(number), 'test'||number FROM numbers(64) FORMAT Vertical; + +SET output_format_pretty_color = 0, output_format_pretty_highlight_digit_groups = 0, output_format_pretty_single_large_number_tip_threshold = 0; +SELECT exp2(number), exp10(number), 'test'||number FROM numbers(64) FORMAT Vertical; From a2220233b75a5da3fd5408f77a87df7c6c5e51d2 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 8 Nov 2024 00:25:49 +0100 Subject: [PATCH 541/680] Fix test --- .../0_stateless/02050_clickhouse_local_parsing_exception.sh | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/02050_clickhouse_local_parsing_exception.sh b/tests/queries/0_stateless/02050_clickhouse_local_parsing_exception.sh index 7a92fa6fefe..65563837f55 100755 --- a/tests/queries/0_stateless/02050_clickhouse_local_parsing_exception.sh +++ b/tests/queries/0_stateless/02050_clickhouse_local_parsing_exception.sh @@ -4,5 +4,4 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) # shellcheck source=../shell_config.sh . "$CURDIR"/../shell_config.sh -$CLICKHOUSE_LOCAL --query="SELECT number FROM system.numbers INTO OUTFILE test.native.zst FORMAT Native" 2>&1 | grep -q "Code: 62. DB::Exception: Syntax error: failed at position 48 ('test'): test.native.zst FORMAT Native. Expected string literal." && echo 'OK' || echo 'FAIL' ||: - +$CLICKHOUSE_LOCAL --query="SELECT number FROM system.numbers INTO OUTFILE test.native.zst FORMAT Native" 2>&1 | grep -q "Code: 62. DB::Exception: Syntax error: failed at position 48 ('test'): test.native.zst FORMAT Native." && echo 'OK' || echo 'FAIL' ||: From 5ceb19453d108163880e9d7fdd06ec4858606c52 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 8 Nov 2024 00:26:58 +0100 Subject: [PATCH 542/680] Fix style --- src/Formats/PrettyFormatHelpers.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/Formats/PrettyFormatHelpers.h b/src/Formats/PrettyFormatHelpers.h index 72ab5e3c2a0..b5d679c5a42 100644 --- a/src/Formats/PrettyFormatHelpers.h +++ b/src/Formats/PrettyFormatHelpers.h @@ -1,5 +1,8 @@ +#pragma once + #include + namespace DB { From dd5a573302a3e38e15b3bfbc8cafe75cdc22cc7c Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 8 Nov 2024 00:50:13 +0100 Subject: [PATCH 543/680] Reset MergeTree to master --- src/Storages/MergeTree/DataPartsExchange.cpp | 2 +- .../MergeTree/FutureMergedMutatedPart.h | 1 + src/Storages/MergeTree/IDataPartStorage.h | 3 +- src/Storages/MergeTree/IMergeTreeDataPart.cpp | 2 +- src/Storages/MergeTree/IMergeTreeDataPart.h | 3 + .../MergeTree/IMergeTreeDataPartWriter.cpp | 29 +- .../MergeTree/IMergeTreeDataPartWriter.h | 8 + src/Storages/MergeTree/IMergeTreeReader.h | 1 + .../MergeTree/IMergedBlockOutputStream.cpp | 1 - .../MergeTree/IMergedBlockOutputStream.h | 11 +- src/Storages/MergeTree/KeyCondition.cpp | 25 ++ .../MergeTree/MergeFromLogEntryTask.cpp | 9 +- .../MergeTree/MergePlainMergeTreeTask.cpp | 37 +- .../MergeTree/MergeProjectionPartsTask.cpp | 3 + .../MergeSelectors/TrivialMergeSelector.cpp | 94 +++++ .../MergeSelectors/TrivialMergeSelector.h | 32 ++ .../MergeSelectors/registerMergeSelectors.cpp | 2 + src/Storages/MergeTree/MergeTask.cpp | 42 +- src/Storages/MergeTree/MergeTask.h | 9 + src/Storages/MergeTree/MergeTreeData.cpp | 160 ++++++- src/Storages/MergeTree/MergeTreeData.h | 6 +- .../MergeTree/MergeTreeDataFormatVersion.h | 4 +- .../MergeTree/MergeTreeDataMergerMutator.cpp | 34 +- .../MergeTree/MergeTreeDataMergerMutator.h | 2 + .../MergeTree/MergeTreeDataPartBuilder.cpp | 18 +- .../MergeTree/MergeTreeDataPartBuilder.h | 12 +- .../MergeTree/MergeTreeDataPartCompact.cpp | 35 +- .../MergeTree/MergeTreeDataPartCompact.h | 2 + .../MergeTree/MergeTreeDataPartType.h | 1 + .../MergeTree/MergeTreeDataPartWide.cpp | 50 ++- .../MergeTree/MergeTreeDataPartWide.h | 2 + .../MergeTreeDataPartWriterCompact.cpp | 56 +-- .../MergeTreeDataPartWriterCompact.h | 8 +- .../MergeTreeDataPartWriterOnDisk.cpp | 206 ++++----- .../MergeTree/MergeTreeDataPartWriterOnDisk.h | 40 +- .../MergeTree/MergeTreeDataPartWriterWide.cpp | 81 ++-- .../MergeTree/MergeTreeDataPartWriterWide.h | 13 +- .../MergeTree/MergeTreeDataSelectExecutor.cpp | 43 +- .../MergeTree/MergeTreeDataSelectExecutor.h | 6 +- .../MergeTree/MergeTreeDataWriter.cpp | 15 +- .../MergeTree/MergeTreeIOSettings.cpp | 4 +- src/Storages/MergeTree/MergeTreeIOSettings.h | 5 +- .../MergeTree/MergeTreeIndexGranularity.cpp | 18 +- .../MergeTree/MergeTreeIndexGranularity.h | 2 +- .../MergeTreeIndexGranularityInfo.cpp | 8 + .../MergeTree/MergeTreeIndexGranularityInfo.h | 1 + .../MergeTreeIndexVectorSimilarity.cpp | 67 +-- .../MergeTreeIndexVectorSimilarity.h | 14 +- .../MergeTree/MergeTreeMarksLoader.cpp | 29 ++ src/Storages/MergeTree/MergeTreeMarksLoader.h | 13 +- .../MergeTree/MergeTreeMutationStatus.cpp | 4 +- src/Storages/MergeTree/MergeTreePartInfo.h | 7 + .../MergeTree/MergeTreePartsMover.cpp | 2 +- .../MergeTree/MergeTreePrefetchedReadPool.cpp | 36 +- .../MergeTree/MergeTreePrefetchedReadPool.h | 1 + src/Storages/MergeTree/MergeTreeRangeReader.h | 2 +- src/Storages/MergeTree/MergeTreeReadPool.cpp | 2 + src/Storages/MergeTree/MergeTreeReadPool.h | 1 + .../MergeTree/MergeTreeReadPoolBase.cpp | 105 +++-- .../MergeTree/MergeTreeReadPoolBase.h | 4 + .../MergeTree/MergeTreeReadPoolInOrder.cpp | 2 + .../MergeTree/MergeTreeReadPoolInOrder.h | 1 + .../MergeTreeReadPoolParallelReplicas.cpp | 2 + .../MergeTreeReadPoolParallelReplicas.h | 1 + ...rgeTreeReadPoolParallelReplicasInOrder.cpp | 2 + ...MergeTreeReadPoolParallelReplicasInOrder.h | 1 + src/Storages/MergeTree/MergeTreeReadTask.cpp | 25 +- src/Storages/MergeTree/MergeTreeReadTask.h | 10 +- .../MergeTree/MergeTreeReaderWide.cpp | 2 +- .../MergeTree/MergeTreeSelectAlgorithms.cpp | 5 +- .../MergeTree/MergeTreeSelectAlgorithms.h | 8 +- .../MergeTree/MergeTreeSelectProcessor.cpp | 4 +- .../MergeTree/MergeTreeSelectProcessor.h | 1 - src/Storages/MergeTree/MergeTreeSettings.cpp | 397 +++++++++--------- src/Storages/MergeTree/MergeTreeSink.cpp | 21 +- .../MergeTree/MergedBlockOutputStream.cpp | 2 + .../MergeTree/MergedBlockOutputStream.h | 1 + .../MergedColumnOnlyOutputStream.cpp | 5 +- .../MergeTree/MergedColumnOnlyOutputStream.h | 1 + .../MergeTree/MutateFromLogEntryTask.cpp | 4 + .../MergeTree/MutatePlainMergeTreeTask.cpp | 4 + src/Storages/MergeTree/MutateTask.cpp | 5 +- .../ReplicatedMergeTreeAttachThread.cpp | 90 +--- .../ReplicatedMergeTreeAttachThread.h | 2 - .../MergeTree/ReplicatedMergeTreeQueue.cpp | 2 +- .../MergeTree/ReplicatedMergeTreeQueue.h | 1 + .../ReplicatedMergeTreeRestartingThread.cpp | 92 ++++ .../ReplicatedMergeTreeRestartingThread.h | 4 + .../MergeTree/ReplicatedMergeTreeSink.cpp | 36 +- src/Storages/MergeTree/checkDataPart.cpp | 2 +- 90 files changed, 1398 insertions(+), 768 deletions(-) create mode 100644 src/Storages/MergeTree/MergeSelectors/TrivialMergeSelector.cpp create mode 100644 src/Storages/MergeTree/MergeSelectors/TrivialMergeSelector.h diff --git a/src/Storages/MergeTree/DataPartsExchange.cpp b/src/Storages/MergeTree/DataPartsExchange.cpp index e13ec5a7515..1d79ae5aacb 100644 --- a/src/Storages/MergeTree/DataPartsExchange.cpp +++ b/src/Storages/MergeTree/DataPartsExchange.cpp @@ -908,7 +908,7 @@ MergeTreeData::MutableDataPartPtr Fetcher::downloadPartToDisk( { part_storage_for_loading->commitTransaction(); - MergeTreeDataPartBuilder builder(data, part_name, volume, part_relative_path, part_dir); + MergeTreeDataPartBuilder builder(data, part_name, volume, part_relative_path, part_dir, getReadSettings()); new_data_part = builder.withPartFormatFromDisk().build(); new_data_part->version.setCreationTID(Tx::PrehistoricTID, nullptr); diff --git a/src/Storages/MergeTree/FutureMergedMutatedPart.h b/src/Storages/MergeTree/FutureMergedMutatedPart.h index 09fb7b01678..ca607bb4e33 100644 --- a/src/Storages/MergeTree/FutureMergedMutatedPart.h +++ b/src/Storages/MergeTree/FutureMergedMutatedPart.h @@ -22,6 +22,7 @@ struct FutureMergedMutatedPart MergeTreeDataPartFormat part_format; MergeTreePartInfo part_info; MergeTreeData::DataPartsVector parts; + std::vector blocking_parts_to_remove; MergeType merge_type = MergeType::Regular; const MergeTreePartition & getPartition() const { return parts.front()->partition; } diff --git a/src/Storages/MergeTree/IDataPartStorage.h b/src/Storages/MergeTree/IDataPartStorage.h index a09c24c63ab..49d9fbf2291 100644 --- a/src/Storages/MergeTree/IDataPartStorage.h +++ b/src/Storages/MergeTree/IDataPartStorage.h @@ -1,5 +1,4 @@ #pragma once -#include #include #include #include @@ -16,7 +15,7 @@ namespace DB { - +struct ReadSettings; class ReadBufferFromFileBase; class WriteBufferFromFileBase; diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp index 20d7528d38a..41783ffddb0 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp @@ -833,7 +833,7 @@ MergeTreeDataPartBuilder IMergeTreeDataPart::getProjectionPartBuilder(const Stri { const char * projection_extension = is_temp_projection ? ".tmp_proj" : ".proj"; auto projection_storage = getDataPartStorage().getProjection(projection_name + projection_extension, !is_temp_projection); - MergeTreeDataPartBuilder builder(storage, projection_name, projection_storage); + MergeTreeDataPartBuilder builder(storage, projection_name, projection_storage, getReadSettings()); return builder.withPartInfo(MergeListElement::FAKE_RESULT_PART_FOR_PROJECTION).withParentPart(this); } diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.h b/src/Storages/MergeTree/IMergeTreeDataPart.h index 378832d32a1..b41a1d840e1 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.h +++ b/src/Storages/MergeTree/IMergeTreeDataPart.h @@ -180,6 +180,9 @@ public: void loadRowsCountFileForUnexpectedPart(); + /// Loads marks and saves them into mark cache for specified columns. + virtual void loadMarksToCache(const Names & column_names, MarkCache * mark_cache) const = 0; + String getMarksFileExtension() const { return index_granularity_info.mark_type.getFileExtension(); } /// Generate the new name for this part according to `new_part_info` and min/max dates from the old name. diff --git a/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp b/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp index a9f188338e1..dbfdbbdea88 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPartWriter.cpp @@ -91,6 +91,13 @@ Columns IMergeTreeDataPartWriter::releaseIndexColumns() return result; } +PlainMarksByName IMergeTreeDataPartWriter::releaseCachedMarks() +{ + PlainMarksByName res; + std::swap(cached_marks, res); + return res; +} + SerializationPtr IMergeTreeDataPartWriter::getSerialization(const String & column_name) const { auto it = serializations.find(column_name); @@ -178,24 +185,9 @@ MergeTreeDataPartWriterPtr createMergeTreeDataPartWriter( const MergeTreeIndexGranularity & computed_index_granularity) { if (part_type == MergeTreeDataPartType::Compact) - return createMergeTreeDataPartCompactWriter( - data_part_name_, - logger_name_, - serializations_, - data_part_storage_, - index_granularity_info_, - storage_settings_, - columns_list, - column_positions, - metadata_snapshot, - virtual_columns, - indices_to_recalc, - stats_to_recalc_, - marks_file_extension_, - default_codec_, - writer_settings, - computed_index_granularity); - + return createMergeTreeDataPartCompactWriter(data_part_name_, logger_name_, serializations_, data_part_storage_, + index_granularity_info_, storage_settings_, columns_list, column_positions, metadata_snapshot, virtual_columns, indices_to_recalc, stats_to_recalc_, + marks_file_extension_, default_codec_, writer_settings, computed_index_granularity); if (part_type == MergeTreeDataPartType::Wide) return createMergeTreeDataPartWideWriter( data_part_name_, @@ -213,7 +205,6 @@ MergeTreeDataPartWriterPtr createMergeTreeDataPartWriter( default_codec_, writer_settings, computed_index_granularity); - throw Exception(ErrorCodes::LOGICAL_ERROR, "Unknown part type: {}", part_type.toString()); } diff --git a/src/Storages/MergeTree/IMergeTreeDataPartWriter.h b/src/Storages/MergeTree/IMergeTreeDataPartWriter.h index eb51a1b2922..d1c76505d7c 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPartWriter.h +++ b/src/Storages/MergeTree/IMergeTreeDataPartWriter.h @@ -8,6 +8,7 @@ #include #include #include +#include namespace DB @@ -45,7 +46,12 @@ public: virtual void finish(bool sync) = 0; + virtual size_t getNumberOfOpenStreams() const = 0; + Columns releaseIndexColumns(); + + PlainMarksByName releaseCachedMarks(); + const MergeTreeIndexGranularity & getIndexGranularity() const { return index_granularity; } protected: @@ -69,6 +75,8 @@ protected: MutableDataPartStoragePtr data_part_storage; MutableColumns index_columns; MergeTreeIndexGranularity index_granularity; + /// Marks that will be saved to cache on finish. + PlainMarksByName cached_marks; }; using MergeTreeDataPartWriterPtr = std::unique_ptr; diff --git a/src/Storages/MergeTree/IMergeTreeReader.h b/src/Storages/MergeTree/IMergeTreeReader.h index d799ce57b40..c68617d3995 100644 --- a/src/Storages/MergeTree/IMergeTreeReader.h +++ b/src/Storages/MergeTree/IMergeTreeReader.h @@ -18,6 +18,7 @@ public: using ValueSizeMap = std::map; using VirtualFields = std::unordered_map; using DeserializeBinaryBulkStateMap = std::map; + using FileStreams = std::map>; IMergeTreeReader( MergeTreeDataPartInfoForReaderPtr data_part_info_for_read_, diff --git a/src/Storages/MergeTree/IMergedBlockOutputStream.cpp b/src/Storages/MergeTree/IMergedBlockOutputStream.cpp index 209b274ee6a..eb904a8e2ef 100644 --- a/src/Storages/MergeTree/IMergedBlockOutputStream.cpp +++ b/src/Storages/MergeTree/IMergedBlockOutputStream.cpp @@ -4,7 +4,6 @@ #include #include - namespace DB { diff --git a/src/Storages/MergeTree/IMergedBlockOutputStream.h b/src/Storages/MergeTree/IMergedBlockOutputStream.h index f67cf66ee50..7dd6d720170 100644 --- a/src/Storages/MergeTree/IMergedBlockOutputStream.h +++ b/src/Storages/MergeTree/IMergedBlockOutputStream.h @@ -7,7 +7,6 @@ #include #include - namespace DB { @@ -35,6 +34,16 @@ public: return writer->getIndexGranularity(); } + PlainMarksByName releaseCachedMarks() + { + return writer->releaseCachedMarks(); + } + + size_t getNumberOfOpenStreams() const + { + return writer->getNumberOfOpenStreams(); + } + protected: /// Remove all columns marked expired in data_part. Also, clears checksums diff --git a/src/Storages/MergeTree/KeyCondition.cpp b/src/Storages/MergeTree/KeyCondition.cpp index 1506dc38946..17723d341fb 100644 --- a/src/Storages/MergeTree/KeyCondition.cpp +++ b/src/Storages/MergeTree/KeyCondition.cpp @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -1446,6 +1447,30 @@ public: IFunctionBase::Monotonicity getMonotonicityForRange(const IDataType & type, const Field & left, const Field & right) const override { + if (const auto * adaptor = typeid_cast(func.get())) + { + if (dynamic_cast(adaptor->getFunction().get()) && kind == Kind::RIGHT_CONST) + { + auto time_zone = extractTimeZoneNameFromColumn(const_arg.column.get(), const_arg.name); + + const IDataType * type_ptr = &type; + if (const auto * low_cardinality_type = typeid_cast(type_ptr)) + type_ptr = low_cardinality_type->getDictionaryType().get(); + + if (type_ptr->isNullable()) + type_ptr = static_cast(*type_ptr).getNestedType().get(); + + DataTypePtr type_with_time_zone; + if (typeid_cast(type_ptr)) + type_with_time_zone = std::make_shared(time_zone); + else if (const auto * dt64 = typeid_cast(type_ptr)) + type_with_time_zone = std::make_shared(dt64->getScale(), time_zone); + else + return {}; /// In case we will have other types with time zone + + return func->getMonotonicityForRange(*type_with_time_zone, left, right); + } + } return func->getMonotonicityForRange(type, left, right); } diff --git a/src/Storages/MergeTree/MergeFromLogEntryTask.cpp b/src/Storages/MergeTree/MergeFromLogEntryTask.cpp index 56d7133dfc3..d7e807c689f 100644 --- a/src/Storages/MergeTree/MergeFromLogEntryTask.cpp +++ b/src/Storages/MergeTree/MergeFromLogEntryTask.cpp @@ -335,6 +335,10 @@ ReplicatedMergeMutateTaskBase::PrepareResult MergeFromLogEntryTask::prepare() future_merged_part, task_context); + storage.writePartLog( + PartLogElement::MERGE_PARTS_START, {}, 0, + entry.new_part_name, part, parts, merge_mutate_entry.get(), {}); + transaction_ptr = std::make_unique(storage, NO_TRANSACTION_RAW); merge_task = storage.merger_mutator.mergePartsToTemporaryPart( @@ -352,7 +356,6 @@ ReplicatedMergeMutateTaskBase::PrepareResult MergeFromLogEntryTask::prepare() storage.merging_params, NO_TRANSACTION_PTR); - /// Adjust priority for (auto & item : future_merged_part->parts) priority.value += item->getBytesOnDisk(); @@ -368,6 +371,7 @@ ReplicatedMergeMutateTaskBase::PrepareResult MergeFromLogEntryTask::prepare() bool MergeFromLogEntryTask::finalize(ReplicatedMergeMutateTaskBase::PartLogWriter write_part_log) { part = merge_task->getFuture().get(); + auto cached_marks = merge_task->releaseCachedMarks(); storage.merger_mutator.renameMergedTemporaryPart(part, parts, NO_TRANSACTION_PTR, *transaction_ptr); /// Why we reset task here? Because it holds shared pointer to part and tryRemovePartImmediately will @@ -441,6 +445,9 @@ bool MergeFromLogEntryTask::finalize(ReplicatedMergeMutateTaskBase::PartLogWrite finish_callback = [storage_ptr = &storage]() { storage_ptr->merge_selecting_task->schedule(); }; ProfileEvents::increment(ProfileEvents::ReplicatedPartMerges); + if (auto * mark_cache = storage.getContext()->getMarkCache().get()) + addMarksToCache(*part, cached_marks, mark_cache); + write_part_log({}); StorageReplicatedMergeTree::incrementMergedPartsProfileEvent(part->getType()); diff --git a/src/Storages/MergeTree/MergePlainMergeTreeTask.cpp b/src/Storages/MergeTree/MergePlainMergeTreeTask.cpp index be44177847c..6aca58faf47 100644 --- a/src/Storages/MergeTree/MergePlainMergeTreeTask.cpp +++ b/src/Storages/MergeTree/MergePlainMergeTreeTask.cpp @@ -92,6 +92,10 @@ void MergePlainMergeTreeTask::prepare() future_part, task_context); + storage.writePartLog( + PartLogElement::MERGE_PARTS_START, {}, 0, + future_part->name, new_part, future_part->parts, merge_list_entry.get(), {}); + write_part_log = [this] (const ExecutionStatus & execution_status) { auto profile_counters_snapshot = std::make_shared(profile_counters.getPartiallyAtomicSnapshot()); @@ -121,19 +125,19 @@ void MergePlainMergeTreeTask::prepare() }; merge_task = storage.merger_mutator.mergePartsToTemporaryPart( - future_part, - metadata_snapshot, - merge_list_entry.get(), - {} /* projection_merge_list_element */, - table_lock_holder, - time(nullptr), - task_context, - merge_mutate_entry->tagger->reserved_space, - deduplicate, - deduplicate_by_columns, - cleanup, - storage.merging_params, - txn); + future_part, + metadata_snapshot, + merge_list_entry.get(), + {} /* projection_merge_list_element */, + table_lock_holder, + time(nullptr), + task_context, + merge_mutate_entry->tagger->reserved_space, + deduplicate, + deduplicate_by_columns, + cleanup, + storage.merging_params, + txn); } @@ -148,6 +152,12 @@ void MergePlainMergeTreeTask::finish() ThreadFuzzer::maybeInjectSleep(); ThreadFuzzer::maybeInjectMemoryLimitException(); + if (auto * mark_cache = storage.getContext()->getMarkCache().get()) + { + auto marks = merge_task->releaseCachedMarks(); + addMarksToCache(*new_part, marks, mark_cache); + } + write_part_log({}); StorageMergeTree::incrementMergedPartsProfileEvent(new_part->getType()); transfer_profile_counters_to_initial_query(); @@ -159,7 +169,6 @@ void MergePlainMergeTreeTask::finish() ThreadFuzzer::maybeInjectSleep(); ThreadFuzzer::maybeInjectMemoryLimitException(); } - } ContextMutablePtr MergePlainMergeTreeTask::createTaskContext() const diff --git a/src/Storages/MergeTree/MergeProjectionPartsTask.cpp b/src/Storages/MergeTree/MergeProjectionPartsTask.cpp index 4e1bb2f11a7..34cd925a8c6 100644 --- a/src/Storages/MergeTree/MergeProjectionPartsTask.cpp +++ b/src/Storages/MergeTree/MergeProjectionPartsTask.cpp @@ -83,6 +83,9 @@ bool MergeProjectionPartsTask::executeStep() ".tmp_proj"); next_level_parts.push_back(executeHere(tmp_part_merge_task)); + /// FIXME (alesapin) we should use some temporary storage for this, + /// not commit each subprojection part + next_level_parts.back()->getDataPartStorage().commitTransaction(); next_level_parts.back()->is_temp = true; } diff --git a/src/Storages/MergeTree/MergeSelectors/TrivialMergeSelector.cpp b/src/Storages/MergeTree/MergeSelectors/TrivialMergeSelector.cpp new file mode 100644 index 00000000000..cd1fa7b01cd --- /dev/null +++ b/src/Storages/MergeTree/MergeSelectors/TrivialMergeSelector.cpp @@ -0,0 +1,94 @@ +#include +#include + +#include +#include + +#include + + +namespace DB +{ + +void registerTrivialMergeSelector(MergeSelectorFactory & factory) +{ + factory.registerPublicSelector("Trivial", MergeSelectorAlgorithm::TRIVIAL, [](const std::any &) + { + return std::make_shared(); + }); +} + +TrivialMergeSelector::PartsRange TrivialMergeSelector::select( + const PartsRanges & parts_ranges, + size_t max_total_size_to_merge) +{ + size_t num_partitions = parts_ranges.size(); + if (num_partitions == 0) + return {}; + + /// Sort partitions from the largest to smallest in the number of parts. + std::vector sorted_partition_indices; + sorted_partition_indices.reserve(num_partitions); + for (size_t i = 0; i < num_partitions; ++i) + if (parts_ranges[i].size() >= settings.num_parts_to_merge) + sorted_partition_indices.emplace_back(i); + + if (sorted_partition_indices.empty()) + return {}; + + std::sort(sorted_partition_indices.begin(), sorted_partition_indices.end(), + [&](size_t i, size_t j){ return parts_ranges[i].size() > parts_ranges[j].size(); }); + + size_t partition_idx = 0; + size_t left = 0; + size_t right = 0; + + std::vector candidates; + while (candidates.size() < settings.num_ranges_to_choose) + { + const PartsRange & partition = parts_ranges[partition_idx]; + + if (1 + right - left == settings.num_parts_to_merge) + { + ++right; + + size_t total_size = 0; + for (size_t i = left; i < right; ++i) + total_size += partition[i].size; + + if (!max_total_size_to_merge || total_size <= max_total_size_to_merge) + { + candidates.emplace_back(partition.data() + left, partition.data() + right); + if (candidates.size() == settings.num_ranges_to_choose) + break; + } + + left = right; + } + + if (partition.size() - left < settings.num_parts_to_merge) + { + ++partition_idx; + if (partition_idx == sorted_partition_indices.size()) + break; + + left = 0; + right = 0; + } + + ++right; + + if (right < partition.size() && partition[right].level < partition[left].level) + left = right; + } + + if (candidates.empty()) + return {}; + + if (candidates.size() == 1) + return candidates[0]; + + return candidates[thread_local_rng() % candidates.size()]; +} + +} diff --git a/src/Storages/MergeTree/MergeSelectors/TrivialMergeSelector.h b/src/Storages/MergeTree/MergeSelectors/TrivialMergeSelector.h new file mode 100644 index 00000000000..6d989aea0fb --- /dev/null +++ b/src/Storages/MergeTree/MergeSelectors/TrivialMergeSelector.h @@ -0,0 +1,32 @@ +#pragma once + +#include + + +namespace DB +{ + +/** Go through partitions starting from the largest (in the number of parts). + * Go through parts from left to right. + * Find the first range of N parts where their level is not decreasing. + * Then continue finding these ranges and find up to M of these ranges. + * Choose a random one from them. + */ +class TrivialMergeSelector : public IMergeSelector +{ +public: + struct Settings + { + size_t num_parts_to_merge = 10; + size_t num_ranges_to_choose = 100; + }; + + PartsRange select( + const PartsRanges & parts_ranges, + size_t max_total_size_to_merge) override; + +private: + const Settings settings; +}; + +} diff --git a/src/Storages/MergeTree/MergeSelectors/registerMergeSelectors.cpp b/src/Storages/MergeTree/MergeSelectors/registerMergeSelectors.cpp index 61f941adc36..6a3c1ef4b2b 100644 --- a/src/Storages/MergeTree/MergeSelectors/registerMergeSelectors.cpp +++ b/src/Storages/MergeTree/MergeSelectors/registerMergeSelectors.cpp @@ -7,6 +7,7 @@ namespace DB void registerSimpleMergeSelector(MergeSelectorFactory & factory); void registerStochasticSimpleMergeSelector(MergeSelectorFactory & factory); +void registerTrivialMergeSelector(MergeSelectorFactory & factory); void registerAllMergeSelector(MergeSelectorFactory & factory); void registerTTLDeleteMergeSelector(MergeSelectorFactory & factory); void registerTTLRecompressMergeSelector(MergeSelectorFactory & factory); @@ -17,6 +18,7 @@ void registerMergeSelectors() registerSimpleMergeSelector(factory); registerStochasticSimpleMergeSelector(factory); + registerTrivialMergeSelector(factory); registerAllMergeSelector(factory); registerTTLDeleteMergeSelector(factory); registerTTLRecompressMergeSelector(factory); diff --git a/src/Storages/MergeTree/MergeTask.cpp b/src/Storages/MergeTree/MergeTask.cpp index b03fb1b12cf..08066113375 100644 --- a/src/Storages/MergeTree/MergeTask.cpp +++ b/src/Storages/MergeTree/MergeTask.cpp @@ -40,10 +40,22 @@ #include #include +#ifndef NDEBUG + #include +#endif + +#ifdef CLICKHOUSE_CLOUD + #include + #include + #include + #include +#endif + namespace ProfileEvents { extern const Event Merge; + extern const Event MergeSourceParts; extern const Event MergedColumns; extern const Event GatheredColumns; extern const Event MergeTotalMilliseconds; @@ -81,6 +93,7 @@ namespace MergeTreeSetting extern const MergeTreeSettingsUInt64 vertical_merge_algorithm_min_columns_to_activate; extern const MergeTreeSettingsUInt64 vertical_merge_algorithm_min_rows_to_activate; extern const MergeTreeSettingsBool vertical_merge_remote_filesystem_prefetch; + extern const MergeTreeSettingsBool prewarm_mark_cache; } namespace ErrorCodes @@ -295,6 +308,7 @@ void MergeTask::ExecuteAndFinalizeHorizontalPart::extractMergingAndGatheringColu bool MergeTask::ExecuteAndFinalizeHorizontalPart::prepare() const { ProfileEvents::increment(ProfileEvents::Merge); + ProfileEvents::increment(ProfileEvents::MergeSourceParts, global_ctx->future_part->parts.size()); String local_tmp_prefix; if (global_ctx->need_prefix) @@ -335,13 +349,13 @@ bool MergeTask::ExecuteAndFinalizeHorizontalPart::prepare() const if (global_ctx->parent_part) { auto data_part_storage = global_ctx->parent_part->getDataPartStorage().getProjection(local_tmp_part_basename, /* use parent transaction */ false); - builder.emplace(*global_ctx->data, global_ctx->future_part->name, data_part_storage); + builder.emplace(*global_ctx->data, global_ctx->future_part->name, data_part_storage, getReadSettings()); builder->withParentPart(global_ctx->parent_part); } else { auto local_single_disk_volume = std::make_shared("volume_" + global_ctx->future_part->name, global_ctx->disk, 0); - builder.emplace(global_ctx->data->getDataPartBuilder(global_ctx->future_part->name, local_single_disk_volume, local_tmp_part_basename)); + builder.emplace(global_ctx->data->getDataPartBuilder(global_ctx->future_part->name, local_single_disk_volume, local_tmp_part_basename, getReadSettings())); builder->withPartStorageType(global_ctx->future_part->part_format.storage_type); } @@ -533,6 +547,8 @@ bool MergeTask::ExecuteAndFinalizeHorizontalPart::prepare() const } } + bool save_marks_in_cache = (*global_ctx->data->getSettings())[MergeTreeSetting::prewarm_mark_cache] && global_ctx->context->getMarkCache(); + global_ctx->to = std::make_shared( global_ctx->new_data_part, global_ctx->metadata_snapshot, @@ -542,6 +558,7 @@ bool MergeTask::ExecuteAndFinalizeHorizontalPart::prepare() const ctx->compression_codec, global_ctx->txn ? global_ctx->txn->tid : Tx::PrehistoricTID, /*reset_columns=*/ true, + save_marks_in_cache, ctx->blocks_are_granules_size, global_ctx->context->getWriteSettings()); @@ -1072,6 +1089,8 @@ void MergeTask::VerticalMergeStage::prepareVerticalMergeForOneColumn() const ctx->executor = std::make_unique(ctx->column_parts_pipeline); NamesAndTypesList columns_list = {*ctx->it_name_and_type}; + bool save_marks_in_cache = (*global_ctx->data->getSettings())[MergeTreeSetting::prewarm_mark_cache] && global_ctx->context->getMarkCache(); + ctx->column_to = std::make_unique( global_ctx->new_data_part, global_ctx->metadata_snapshot, @@ -1080,6 +1099,7 @@ void MergeTask::VerticalMergeStage::prepareVerticalMergeForOneColumn() const column_pipepline.indexes_to_recalc, getStatisticsForColumns(columns_list, global_ctx->metadata_snapshot), &global_ctx->written_offset_columns, + save_marks_in_cache, global_ctx->to->getIndexGranularity()); ctx->column_elems_written = 0; @@ -1117,6 +1137,10 @@ void MergeTask::VerticalMergeStage::finalizeVerticalMergeForOneColumn() const auto changed_checksums = ctx->column_to->fillChecksums(global_ctx->new_data_part, global_ctx->checksums_gathered_columns); global_ctx->checksums_gathered_columns.add(std::move(changed_checksums)); + auto cached_marks = ctx->column_to->releaseCachedMarks(); + for (auto & [name, marks] : cached_marks) + global_ctx->cached_marks.emplace(name, std::move(marks)); + ctx->delayed_streams.emplace_back(std::move(ctx->column_to)); while (ctx->delayed_streams.size() > ctx->max_delayed_streams) @@ -1263,6 +1287,10 @@ bool MergeTask::MergeProjectionsStage::finalizeProjectionsAndWholeMerge() const else global_ctx->to->finalizePart(global_ctx->new_data_part, ctx->need_sync, &global_ctx->storage_columns, &global_ctx->checksums_gathered_columns); + auto cached_marks = global_ctx->to->releaseCachedMarks(); + for (auto & [name, marks] : cached_marks) + global_ctx->cached_marks.emplace(name, std::move(marks)); + global_ctx->new_data_part->getDataPartStorage().precommitTransaction(); global_ctx->promise.set_value(global_ctx->new_data_part); @@ -1385,7 +1413,7 @@ bool MergeTask::execute() } -/// Apply merge strategy (Ordinary, Colapsing, Aggregating, etc) to the stream +/// Apply merge strategy (Ordinary, Collapsing, Aggregating, etc) to the stream class MergePartsStep : public ITransformingStep { public: @@ -1421,7 +1449,7 @@ public: /// that is going in insertion order. ProcessorPtr merged_transform; - const auto &header = pipeline.getHeader(); + const auto & header = pipeline.getHeader(); const auto input_streams_count = pipeline.getNumStreams(); WriteBuffer * rows_sources_write_buf = nullptr; @@ -1690,7 +1718,7 @@ void MergeTask::ExecuteAndFinalizeHorizontalPart::createMergedStream() const sort_description, partition_key_columns, global_ctx->merging_params, - (is_vertical_merge ? RowsSourcesTemporaryFile::FILE_ID : ""), /// rows_sources temporaty file is used only for vertical merge + (is_vertical_merge ? RowsSourcesTemporaryFile::FILE_ID : ""), /// rows_sources' temporary file is used only for vertical merge (*data_settings)[MergeTreeSetting::merge_max_block_size], (*data_settings)[MergeTreeSetting::merge_max_block_size_bytes], ctx->blocks_are_granules_size, @@ -1759,6 +1787,10 @@ void MergeTask::ExecuteAndFinalizeHorizontalPart::createMergedStream() const auto optimization_settings = QueryPlanOptimizationSettings::fromContext(global_ctx->context); auto builder = merge_parts_query_plan.buildQueryPipeline(optimization_settings, pipeline_settings); + // Merges are not using concurrency control now. Queries and merges running together could lead to CPU overcommit. + // TODO(serxa): Enable concurrency control for merges. This should be done after CPU scheduler introduction. + builder->setConcurrencyControl(false); + global_ctx->merged_pipeline = QueryPipelineBuilder::getPipeline(std::move(*builder)); } diff --git a/src/Storages/MergeTree/MergeTask.h b/src/Storages/MergeTree/MergeTask.h index 5a4fb1ec0b8..53792165987 100644 --- a/src/Storages/MergeTree/MergeTask.h +++ b/src/Storages/MergeTree/MergeTask.h @@ -5,6 +5,7 @@ #include #include +#include #include #include @@ -132,6 +133,13 @@ public: return nullptr; } + PlainMarksByName releaseCachedMarks() const + { + PlainMarksByName res; + std::swap(global_ctx->cached_marks, res); + return res; + } + bool execute(); private: @@ -209,6 +217,7 @@ private: std::promise promise{}; IMergedBlockOutputStream::WrittenOffsetColumns written_offset_columns{}; + PlainMarksByName cached_marks; MergeTreeTransactionPtr txn; bool need_prefix; diff --git a/src/Storages/MergeTree/MergeTreeData.cpp b/src/Storages/MergeTree/MergeTreeData.cpp index 72a41fcf2c1..b2f35d0a309 100644 --- a/src/Storages/MergeTree/MergeTreeData.cpp +++ b/src/Storages/MergeTree/MergeTreeData.cpp @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -154,6 +155,7 @@ namespace namespace DB { + namespace Setting { extern const SettingsBool allow_drop_detached; @@ -229,6 +231,12 @@ namespace MergeTreeSetting extern const MergeTreeSettingsString storage_policy; extern const MergeTreeSettingsFloat zero_copy_concurrent_part_removal_max_postpone_ratio; extern const MergeTreeSettingsUInt64 zero_copy_concurrent_part_removal_max_split_times; + extern const MergeTreeSettingsBool prewarm_mark_cache; +} + +namespace ServerSetting +{ + extern const ServerSettingsDouble mark_cache_prewarm_ratio; } namespace ErrorCodes @@ -261,6 +269,7 @@ namespace ErrorCodes extern const int SUPPORT_IS_DISABLED; extern const int TOO_MANY_SIMULTANEOUS_QUERIES; extern const int INCORRECT_QUERY; + extern const int INVALID_SETTING_VALUE; extern const int CANNOT_RESTORE_TABLE; extern const int ZERO_COPY_REPLICATION_ERROR; extern const int NOT_INITIALIZED; @@ -759,6 +768,16 @@ void MergeTreeData::checkProperties( } } + /// If adaptive index granularity is disabled, certain vector search queries with PREWHERE run into LOGICAL_ERRORs. + /// SET allow_experimental_vector_similarity_index = 1; + /// CREATE TABLE tab (`id` Int32, `vec` Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance') GRANULARITY 100000000) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes = 0; + /// INSERT INTO tab SELECT number, [toFloat32(number), 0.] FROM numbers(10000); + /// WITH [1., 0.] AS reference_vec SELECT id, L2Distance(vec, reference_vec) FROM tab PREWHERE toLowCardinality(10) ORDER BY L2Distance(vec, reference_vec) ASC LIMIT 100; + /// As a workaround, force enabled adaptive index granularity for now (it is the default anyways). + if (new_metadata.secondary_indices.hasType("vector_similarity") && (*getSettings())[MergeTreeSetting::index_granularity_bytes] == 0) + throw Exception(ErrorCodes::INVALID_SETTING_VALUE, + "Experimental vector similarity index can only be used with MergeTree setting 'index_granularity_bytes' != 0"); + if (!new_metadata.projections.empty()) { std::unordered_set projections_names; @@ -1423,7 +1442,7 @@ void MergeTreeData::loadUnexpectedDataPart(UnexpectedPartLoadState & state) try { - state.part = getDataPartBuilder(part_name, single_disk_volume, part_name) + state.part = getDataPartBuilder(part_name, single_disk_volume, part_name, getReadSettings()) .withPartInfo(part_info) .withPartFormatFromDisk() .build(); @@ -1438,7 +1457,7 @@ void MergeTreeData::loadUnexpectedDataPart(UnexpectedPartLoadState & state) /// Build a fake part and mark it as broken in case of filesystem error. /// If the error impacts part directory instead of single files, /// an exception will be thrown during detach and silently ignored. - state.part = getDataPartBuilder(part_name, single_disk_volume, part_name) + state.part = getDataPartBuilder(part_name, single_disk_volume, part_name, getReadSettings()) .withPartStorageType(MergeTreeDataPartStorageType::Full) .withPartType(MergeTreeDataPartType::Wide) .build(); @@ -1472,7 +1491,7 @@ MergeTreeData::LoadPartResult MergeTreeData::loadDataPart( /// Build a fake part and mark it as broken in case of filesystem error. /// If the error impacts part directory instead of single files, /// an exception will be thrown during detach and silently ignored. - res.part = getDataPartBuilder(part_name, single_disk_volume, part_name) + res.part = getDataPartBuilder(part_name, single_disk_volume, part_name, getReadSettings()) .withPartStorageType(MergeTreeDataPartStorageType::Full) .withPartType(MergeTreeDataPartType::Wide) .build(); @@ -1493,7 +1512,7 @@ MergeTreeData::LoadPartResult MergeTreeData::loadDataPart( try { - res.part = getDataPartBuilder(part_name, single_disk_volume, part_name) + res.part = getDataPartBuilder(part_name, single_disk_volume, part_name, getReadSettings()) .withPartInfo(part_info) .withPartFormatFromDisk() .build(); @@ -2324,6 +2343,60 @@ void MergeTreeData::stopOutdatedAndUnexpectedDataPartsLoadingTask() } } +void MergeTreeData::prewarmMarkCacheIfNeeded(ThreadPool & pool) +{ + if (!(*getSettings())[MergeTreeSetting::prewarm_mark_cache]) + return; + + prewarmMarkCache(pool); +} + +void MergeTreeData::prewarmMarkCache(ThreadPool & pool) +{ + auto * mark_cache = getContext()->getMarkCache().get(); + if (!mark_cache) + return; + + auto metadata_snaphost = getInMemoryMetadataPtr(); + auto column_names = getColumnsToPrewarmMarks(*getSettings(), metadata_snaphost->getColumns().getAllPhysical()); + + if (column_names.empty()) + return; + + Stopwatch watch; + LOG_TRACE(log, "Prewarming mark cache"); + + auto data_parts = getDataPartsVectorForInternalUsage(); + + /// Prewarm mark cache firstly for the most fresh parts according + /// to time columns in partition key (if exists) and by modification time. + + auto to_tuple = [](const auto & part) + { + return std::make_tuple(part->getMinMaxDate().second, part->getMinMaxTime().second, part->modification_time); + }; + + std::sort(data_parts.begin(), data_parts.end(), [&to_tuple](const auto & lhs, const auto & rhs) + { + return to_tuple(lhs) > to_tuple(rhs); + }); + + ThreadPoolCallbackRunnerLocal runner(pool, "PrewarmMarks"); + double ratio_to_prewarm = getContext()->getServerSettings()[ServerSetting::mark_cache_prewarm_ratio]; + + for (const auto & part : data_parts) + { + if (mark_cache->sizeInBytes() >= mark_cache->maxSizeInBytes() * ratio_to_prewarm) + break; + + runner([&] { part->loadMarksToCache(column_names, mark_cache); }); + } + + runner.waitForAllToFinishAndRethrowFirstError(); + watch.stop(); + LOG_TRACE(log, "Prewarmed mark cache in {} seconds", watch.elapsedSeconds()); +} + /// Is the part directory old. /// True if its modification time and the modification time of all files inside it is less then threshold. /// (Only files on the first level of nesting are considered). @@ -2655,6 +2728,10 @@ void MergeTreeData::removePartsFinally(const MergeTreeData::DataPartsVector & pa for (const auto & part : parts) { part_log_elem.partition_id = part->info.partition_id; + { + WriteBufferFromString out(part_log_elem.partition); + part->partition.serializeText(part->storage, out, {}); + } part_log_elem.part_name = part->name; part_log_elem.bytes_compressed_on_disk = part->getBytesOnDisk(); part_log_elem.bytes_uncompressed = part->getBytesUncompressedOnDisk(); @@ -3310,6 +3387,16 @@ void MergeTreeData::checkAlterIsPossible(const AlterCommands & commands, Context throw Exception(ErrorCodes::SUPPORT_IS_DISABLED, "Experimental vector similarity index is disabled (turn on setting 'allow_experimental_vector_similarity_index')"); + /// If adaptive index granularity is disabled, certain vector search queries with PREWHERE run into LOGICAL_ERRORs. + /// SET allow_experimental_vector_similarity_index = 1; + /// CREATE TABLE tab (`id` Int32, `vec` Array(Float32), INDEX idx vec TYPE vector_similarity('hnsw', 'L2Distance') GRANULARITY 100000000) ENGINE = MergeTree ORDER BY id SETTINGS index_granularity_bytes = 0; + /// INSERT INTO tab SELECT number, [toFloat32(number), 0.] FROM numbers(10000); + /// WITH [1., 0.] AS reference_vec SELECT id, L2Distance(vec, reference_vec) FROM tab PREWHERE toLowCardinality(10) ORDER BY L2Distance(vec, reference_vec) ASC LIMIT 100; + /// As a workaround, force enabled adaptive index granularity for now (it is the default anyways). + if (AlterCommands::hasVectorSimilarityIndex(new_metadata) && (*getSettings())[MergeTreeSetting::index_granularity_bytes] == 0) + throw Exception(ErrorCodes::INVALID_SETTING_VALUE, + "Experimental vector similarity index can only be used with MergeTree setting 'index_granularity_bytes' != 0"); + for (const auto & disk : getDisks()) if (!disk->supportsHardLinks() && !commands.isSettingsAlter() && !commands.isCommentAlter()) throw Exception( @@ -3622,6 +3709,9 @@ void MergeTreeData::checkAlterIsPossible(const AlterCommands & commands, Context const auto & new_changes = new_metadata.settings_changes->as().changes; local_context->checkMergeTreeSettingsConstraints(*settings_from_storage, new_changes); + bool found_disk_setting = false; + bool found_storage_policy_setting = false; + for (const auto & changed_setting : new_changes) { const auto & setting_name = changed_setting.name; @@ -3645,9 +3735,22 @@ void MergeTreeData::checkAlterIsPossible(const AlterCommands & commands, Context } if (setting_name == "storage_policy") + { checkStoragePolicy(local_context->getStoragePolicy(new_value.safeGet())); + found_storage_policy_setting = true; + } + else if (setting_name == "disk") + { + checkStoragePolicy(local_context->getStoragePolicyFromDisk(new_value.safeGet())); + found_disk_setting = true; + } } + if (found_storage_policy_setting && found_disk_setting) + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "MergeTree settings `storage_policy` and `disk` cannot be specified at the same time"); + /// Check if it is safe to reset the settings for (const auto & current_setting : current_changes) { @@ -3732,9 +3835,9 @@ MergeTreeDataPartFormat MergeTreeData::choosePartFormatOnDisk(size_t bytes_uncom } MergeTreeDataPartBuilder MergeTreeData::getDataPartBuilder( - const String & name, const VolumePtr & volume, const String & part_dir) const + const String & name, const VolumePtr & volume, const String & part_dir, const ReadSettings & read_settings_) const { - return MergeTreeDataPartBuilder(*this, name, volume, relative_data_path, part_dir); + return MergeTreeDataPartBuilder(*this, name, volume, relative_data_path, part_dir, read_settings_); } void MergeTreeData::changeSettings( @@ -3746,12 +3849,16 @@ void MergeTreeData::changeSettings( bool has_storage_policy_changed = false; const auto & new_changes = new_settings->as().changes; + StoragePolicyPtr new_storage_policy = nullptr; for (const auto & change : new_changes) { - if (change.name == "storage_policy") + if (change.name == "disk" || change.name == "storage_policy") { - StoragePolicyPtr new_storage_policy = getContext()->getStoragePolicy(change.value.safeGet()); + if (change.name == "disk") + new_storage_policy = getContext()->getStoragePolicyFromDisk(change.value.safeGet()); + else + new_storage_policy = getContext()->getStoragePolicy(change.value.safeGet()); StoragePolicyPtr old_storage_policy = getStoragePolicy(); /// StoragePolicy of different version or name is guaranteed to have different pointer @@ -5812,7 +5919,7 @@ MergeTreeData::MutableDataPartPtr MergeTreeData::loadPartRestoredFromBackup(cons /// Load this part from the directory `temp_part_dir`. auto load_part = [&] { - MergeTreeDataPartBuilder builder(*this, part_name, single_disk_volume, parent_part_dir, part_dir_name); + MergeTreeDataPartBuilder builder(*this, part_name, single_disk_volume, parent_part_dir, part_dir_name, getReadSettings()); builder.withPartFormatFromDisk(); part = std::move(builder).build(); part->version.setCreationTID(Tx::PrehistoricTID, nullptr); @@ -5827,7 +5934,7 @@ MergeTreeData::MutableDataPartPtr MergeTreeData::loadPartRestoredFromBackup(cons if (!part) { /// Make a fake data part only to copy its files to /detached/. - part = MergeTreeDataPartBuilder{*this, part_name, single_disk_volume, parent_part_dir, part_dir_name} + part = MergeTreeDataPartBuilder{*this, part_name, single_disk_volume, parent_part_dir, part_dir_name, getReadSettings()} .withPartStorageType(MergeTreeDataPartStorageType::Full) .withPartType(MergeTreeDataPartType::Wide) .build(); @@ -6326,6 +6433,12 @@ DetachedPartsInfo MergeTreeData::getDetachedParts() const for (const auto & disk : getDisks()) { + /// While it is possible to have detached parts on readonly/write-once disks + /// (if they were produced on another machine, where it wasn't readonly) + /// to avoid wasting resources for slow disks, avoid trying to enumerate them. + if (disk->isReadOnly() || disk->isWriteOnce()) + continue; + String detached_path = fs::path(relative_data_path) / DETACHED_DIR_NAME; /// Note: we don't care about TOCTOU issue here. @@ -6473,7 +6586,7 @@ MergeTreeData::MutableDataPartsVector MergeTreeData::tryLoadPartsToAttach(const LOG_DEBUG(log, "Checking part {}", new_name); auto single_disk_volume = std::make_shared("volume_" + old_name, disk); - auto part = getDataPartBuilder(old_name, single_disk_volume, source_dir / new_name) + auto part = getDataPartBuilder(old_name, single_disk_volume, source_dir / new_name, getReadSettings()) .withPartFormatFromDisk() .build(); @@ -7528,7 +7641,7 @@ std::pair MergeTreeData::cloneAn std::string(fs::path(dst_part_storage->getFullRootPath()) / tmp_dst_part_name), with_copy); - auto dst_data_part = MergeTreeDataPartBuilder(*this, dst_part_name, dst_part_storage) + auto dst_data_part = MergeTreeDataPartBuilder(*this, dst_part_name, dst_part_storage, getReadSettings()) .withPartFormatFromDisk() .build(); @@ -7874,7 +7987,8 @@ try part_log_elem.event_type = type; - if (part_log_elem.event_type == PartLogElement::MERGE_PARTS) + if (part_log_elem.event_type == PartLogElement::MERGE_PARTS + || part_log_elem.event_type == PartLogElement::MERGE_PARTS_START) { if (merge_entry) { @@ -7899,6 +8013,20 @@ try part_log_elem.table_name = table_id.table_name; part_log_elem.table_uuid = table_id.uuid; part_log_elem.partition_id = MergeTreePartInfo::fromPartName(new_part_name, format_version).partition_id; + + { + const DataPart * result_or_source_data_part = nullptr; + if (result_part) + result_or_source_data_part = result_part.get(); + else if (!source_parts.empty()) + result_or_source_data_part = source_parts.at(0).get(); + if (result_or_source_data_part) + { + WriteBufferFromString out(part_log_elem.partition); + result_or_source_data_part->partition.serializeText(*this, out, {}); + } + } + part_log_elem.part_name = new_part_name; if (result_part) @@ -7928,10 +8056,6 @@ try { part_log_elem.profile_counters = profile_counters; } - else - { - LOG_WARNING(log, "Profile counters are not set"); - } part_log->add(std::move(part_log_elem)); } @@ -8786,7 +8910,7 @@ std::pair MergeTreeData::createE VolumePtr data_part_volume = createVolumeFromReservation(reservation, volume); auto tmp_dir_holder = getTemporaryPartDirectoryHolder(EMPTY_PART_TMP_PREFIX + new_part_name); - auto new_data_part = getDataPartBuilder(new_part_name, data_part_volume, EMPTY_PART_TMP_PREFIX + new_part_name) + auto new_data_part = getDataPartBuilder(new_part_name, data_part_volume, EMPTY_PART_TMP_PREFIX + new_part_name, getReadSettings()) .withBytesAndRowsOnDisk(0, 0) .withPartInfo(new_part_info) .build(); diff --git a/src/Storages/MergeTree/MergeTreeData.h b/src/Storages/MergeTree/MergeTreeData.h index 7a9730e8627..fe360907875 100644 --- a/src/Storages/MergeTree/MergeTreeData.h +++ b/src/Storages/MergeTree/MergeTreeData.h @@ -241,7 +241,7 @@ public: MergeTreeDataPartFormat choosePartFormat(size_t bytes_uncompressed, size_t rows_count) const; MergeTreeDataPartFormat choosePartFormatOnDisk(size_t bytes_uncompressed, size_t rows_count) const; - MergeTreeDataPartBuilder getDataPartBuilder(const String & name, const VolumePtr & volume, const String & part_dir) const; + MergeTreeDataPartBuilder getDataPartBuilder(const String & name, const VolumePtr & volume, const String & part_dir, const ReadSettings & read_settings_) const; /// Auxiliary object to add a set of parts into the working set in two steps: /// * First, as PreActive parts (the parts are ready, but not yet in the active set). @@ -506,6 +506,10 @@ public: /// Load the set of data parts from disk. Call once - immediately after the object is created. void loadDataParts(bool skip_sanity_checks, std::optional> expected_parts); + /// Prewarm mark cache for the most recent data parts. + void prewarmMarkCache(ThreadPool & pool); + void prewarmMarkCacheIfNeeded(ThreadPool & pool); + String getLogName() const { return log.loadName(); } Int64 getMaxBlockNumber() const; diff --git a/src/Storages/MergeTree/MergeTreeDataFormatVersion.h b/src/Storages/MergeTree/MergeTreeDataFormatVersion.h index 0a84f08ea71..a61938a993c 100644 --- a/src/Storages/MergeTree/MergeTreeDataFormatVersion.h +++ b/src/Storages/MergeTree/MergeTreeDataFormatVersion.h @@ -8,7 +8,7 @@ namespace DB STRONG_TYPEDEF(UInt32, MergeTreeDataFormatVersion) -const MergeTreeDataFormatVersion MERGE_TREE_DATA_OLD_FORMAT_VERSION {0}; -const MergeTreeDataFormatVersion MERGE_TREE_DATA_MIN_FORMAT_VERSION_WITH_CUSTOM_PARTITIONING {1}; +static constexpr MergeTreeDataFormatVersion MERGE_TREE_DATA_OLD_FORMAT_VERSION {0}; +static constexpr MergeTreeDataFormatVersion MERGE_TREE_DATA_MIN_FORMAT_VERSION_WITH_CUSTOM_PARTITIONING {1}; } diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp index 8b3c7bdf3fb..176b5c00b0a 100644 --- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp +++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp @@ -48,6 +48,16 @@ namespace CurrentMetrics { extern const Metric BackgroundMergesAndMutationsPoolTask; } +namespace ProfileEvents +{ + + extern const Event MergerMutatorsGetPartsForMergeElapsedMicroseconds; + extern const Event MergerMutatorPrepareRangesForMergeElapsedMicroseconds; + extern const Event MergerMutatorSelectPartsForMergeElapsedMicroseconds; + extern const Event MergerMutatorRangesForMergeCount; + extern const Event MergerMutatorPartsInRangesForMergeCount; + extern const Event MergerMutatorSelectRangePartsCount; +} namespace DB { @@ -70,6 +80,8 @@ namespace MergeTreeSetting extern const MergeTreeSettingsBool ttl_only_drop_parts; extern const MergeTreeSettingsUInt64 parts_to_throw_insert; extern const MergeTreeSettingsMergeSelectorAlgorithm merge_selector_algorithm; + extern const MergeTreeSettingsBool merge_selector_enable_heuristic_to_remove_small_parts_at_right; + extern const MergeTreeSettingsFloat merge_selector_base; } namespace ErrorCodes @@ -213,6 +225,7 @@ MergeTreeDataMergerMutator::PartitionIdsHint MergeTreeDataMergerMutator::getPart { PartitionIdsHint res; MergeTreeData::DataPartsVector data_parts = getDataPartsToSelectMergeFrom(txn); + if (data_parts.empty()) return res; @@ -270,6 +283,8 @@ MergeTreeDataMergerMutator::PartitionIdsHint MergeTreeDataMergerMutator::getPart MergeTreeData::DataPartsVector MergeTreeDataMergerMutator::getDataPartsToSelectMergeFrom( const MergeTreeTransactionPtr & txn, const PartitionIdsHint * partitions_hint) const { + + Stopwatch get_data_parts_for_merge_timer; auto res = getDataPartsToSelectMergeFrom(txn); if (!partitions_hint) return res; @@ -278,6 +293,8 @@ MergeTreeData::DataPartsVector MergeTreeDataMergerMutator::getDataPartsToSelectM { return !partitions_hint->contains(part->info.partition_id); }); + + ProfileEvents::increment(ProfileEvents::MergerMutatorsGetPartsForMergeElapsedMicroseconds, get_data_parts_for_merge_timer.elapsedMicroseconds()); return res; } @@ -355,6 +372,7 @@ MergeTreeDataMergerMutator::MergeSelectingInfo MergeTreeDataMergerMutator::getPo const MergeTreeTransactionPtr & txn, PreformattedMessage & out_disable_reason) const { + Stopwatch ranges_for_merge_timer; MergeSelectingInfo res; res.current_time = std::time(nullptr); @@ -455,6 +473,10 @@ MergeTreeDataMergerMutator::MergeSelectingInfo MergeTreeDataMergerMutator::getPo prev_part = ∂ } + ProfileEvents::increment(ProfileEvents::MergerMutatorPartsInRangesForMergeCount, res.parts_selected_precondition); + ProfileEvents::increment(ProfileEvents::MergerMutatorRangesForMergeCount, res.parts_ranges.size()); + ProfileEvents::increment(ProfileEvents::MergerMutatorPrepareRangesForMergeElapsedMicroseconds, ranges_for_merge_timer.elapsedMicroseconds()); + return res; } @@ -469,6 +491,7 @@ SelectPartsDecision MergeTreeDataMergerMutator::selectPartsToMergeFromRanges( PreformattedMessage & out_disable_reason, bool dry_run) { + Stopwatch select_parts_from_ranges_timer; const auto data_settings = data.getSettings(); IMergeSelector::PartsRange parts_to_merge; @@ -540,6 +563,9 @@ SelectPartsDecision MergeTreeDataMergerMutator::selectPartsToMergeFromRanges( /// Override value from table settings simple_merge_settings.window_size = (*data_settings)[MergeTreeSetting::merge_selector_window_size]; simple_merge_settings.max_parts_to_merge_at_once = (*data_settings)[MergeTreeSetting::max_parts_to_merge_at_once]; + simple_merge_settings.enable_heuristic_to_remove_small_parts_at_right = (*data_settings)[MergeTreeSetting::merge_selector_enable_heuristic_to_remove_small_parts_at_right]; + simple_merge_settings.base = (*data_settings)[MergeTreeSetting::merge_selector_base]; + if (!(*data_settings)[MergeTreeSetting::min_age_to_force_merge_on_partition_only]) simple_merge_settings.min_age_to_force_merge = (*data_settings)[MergeTreeSetting::min_age_to_force_merge_seconds]; @@ -565,7 +591,8 @@ SelectPartsDecision MergeTreeDataMergerMutator::selectPartsToMergeFromRanges( if (parts_to_merge.empty()) { - out_disable_reason = PreformattedMessage::create("Did not find any parts to merge (with usual merge selectors)"); + ProfileEvents::increment(ProfileEvents::MergerMutatorSelectPartsForMergeElapsedMicroseconds, select_parts_from_ranges_timer.elapsedMicroseconds()); + out_disable_reason = PreformattedMessage::create("Did not find any parts to merge (with usual merge selectors) in {}ms", select_parts_from_ranges_timer.elapsedMicroseconds() / 1000); return SelectPartsDecision::CANNOT_SELECT; } } @@ -578,8 +605,11 @@ SelectPartsDecision MergeTreeDataMergerMutator::selectPartsToMergeFromRanges( parts.push_back(part); } - LOG_DEBUG(log, "Selected {} parts from {} to {}", parts.size(), parts.front()->name, parts.back()->name); + LOG_DEBUG(log, "Selected {} parts from {} to {} in {}ms", parts.size(), parts.front()->name, parts.back()->name, select_parts_from_ranges_timer.elapsedMicroseconds() / 1000); + ProfileEvents::increment(ProfileEvents::MergerMutatorSelectRangePartsCount, parts.size()); + future_part->assign(std::move(parts)); + ProfileEvents::increment(ProfileEvents::MergerMutatorSelectPartsForMergeElapsedMicroseconds, select_parts_from_ranges_timer.elapsedMicroseconds()); return SelectPartsDecision::SELECTED; } diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.h b/src/Storages/MergeTree/MergeTreeDataMergerMutator.h index 71fcb93f369..6d209b9f931 100644 --- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.h +++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.h @@ -106,9 +106,11 @@ public: PreformattedMessage & out_disable_reason, bool dry_run = false); + /// Actually the most fresh partition with biggest modification_time String getBestPartitionToOptimizeEntire(const PartitionsInfo & partitions_info) const; /// Useful to quickly get a list of partitions that contain parts that we may want to merge + /// The result is limited by top_number_of_partitions_to_consider_for_merge PartitionIdsHint getPartitionsThatMayBeMerged( size_t max_total_size_to_merge, const AllowedMergingPredicate & can_merge_callback, diff --git a/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp b/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp index 37f578b0c25..6ec4bc31d90 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartBuilder.cpp @@ -14,20 +14,22 @@ namespace ErrorCodes } MergeTreeDataPartBuilder::MergeTreeDataPartBuilder( - const MergeTreeData & data_, String name_, VolumePtr volume_, String root_path_, String part_dir_) + const MergeTreeData & data_, String name_, VolumePtr volume_, String root_path_, String part_dir_, const ReadSettings & read_settings_) : data(data_) , name(std::move(name_)) , volume(std::move(volume_)) , root_path(std::move(root_path_)) , part_dir(std::move(part_dir_)) + , read_settings(read_settings_) { } MergeTreeDataPartBuilder::MergeTreeDataPartBuilder( - const MergeTreeData & data_, String name_, MutableDataPartStoragePtr part_storage_) + const MergeTreeData & data_, String name_, MutableDataPartStoragePtr part_storage_, const ReadSettings & read_settings_) : data(data_) , name(std::move(name_)) , part_storage(std::move(part_storage_)) + , read_settings(read_settings_) { } @@ -73,7 +75,8 @@ MutableDataPartStoragePtr MergeTreeDataPartBuilder::getPartStorageByType( MergeTreeDataPartStorageType storage_type_, const VolumePtr & volume_, const String & root_path_, - const String & part_dir_) + const String & part_dir_, + const ReadSettings &) /// Unused here, but used in private repo. { if (!volume_) throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot create part storage, because volume is not specified"); @@ -112,7 +115,7 @@ MergeTreeDataPartBuilder & MergeTreeDataPartBuilder::withPartType(MergeTreeDataP MergeTreeDataPartBuilder & MergeTreeDataPartBuilder::withPartStorageType(MergeTreeDataPartStorageType storage_type_) { - part_storage = getPartStorageByType(storage_type_, volume, root_path, part_dir); + part_storage = getPartStorageByType(storage_type_, volume, root_path, part_dir, read_settings); return *this; } @@ -126,7 +129,8 @@ MergeTreeDataPartBuilder::PartStorageAndMarkType MergeTreeDataPartBuilder::getPartStorageAndMarkType( const VolumePtr & volume_, const String & root_path_, - const String & part_dir_) + const String & part_dir_, + const ReadSettings & read_settings_) { auto disk = volume_->getDisk(); auto part_relative_path = fs::path(root_path_) / part_dir_; @@ -138,7 +142,7 @@ MergeTreeDataPartBuilder::getPartStorageAndMarkType( if (MarkType::isMarkFileExtension(ext)) { - auto storage = getPartStorageByType(MergeTreeDataPartStorageType::Full, volume_, root_path_, part_dir_); + auto storage = getPartStorageByType(MergeTreeDataPartStorageType::Full, volume_, root_path_, part_dir_, read_settings_); return {std::move(storage), MarkType(ext)}; } } @@ -156,7 +160,7 @@ MergeTreeDataPartBuilder & MergeTreeDataPartBuilder::withPartFormatFromDisk() MergeTreeDataPartBuilder & MergeTreeDataPartBuilder::withPartFormatFromVolume() { assert(volume); - auto [storage, mark_type] = getPartStorageAndMarkType(volume, root_path, part_dir); + auto [storage, mark_type] = getPartStorageAndMarkType(volume, root_path, part_dir, read_settings); if (!storage || !mark_type) { diff --git a/src/Storages/MergeTree/MergeTreeDataPartBuilder.h b/src/Storages/MergeTree/MergeTreeDataPartBuilder.h index 0f54ff0a631..bce881a1970 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartBuilder.h +++ b/src/Storages/MergeTree/MergeTreeDataPartBuilder.h @@ -21,8 +21,8 @@ using VolumePtr = std::shared_ptr; class MergeTreeDataPartBuilder { public: - MergeTreeDataPartBuilder(const MergeTreeData & data_, String name_, VolumePtr volume_, String root_path_, String part_dir_); - MergeTreeDataPartBuilder(const MergeTreeData & data_, String name_, MutableDataPartStoragePtr part_storage_); + MergeTreeDataPartBuilder(const MergeTreeData & data_, String name_, VolumePtr volume_, String root_path_, String part_dir_, const ReadSettings & read_settings_); + MergeTreeDataPartBuilder(const MergeTreeData & data_, String name_, MutableDataPartStoragePtr part_storage_, const ReadSettings & read_settings_); std::shared_ptr build(); @@ -42,7 +42,8 @@ public: static PartStorageAndMarkType getPartStorageAndMarkType( const VolumePtr & volume_, const String & root_path_, - const String & part_dir_); + const String & part_dir_, + const ReadSettings & read_settings); private: Self & withPartFormatFromVolume(); @@ -52,7 +53,8 @@ private: MergeTreeDataPartStorageType storage_type_, const VolumePtr & volume_, const String & root_path_, - const String & part_dir_); + const String & part_dir_, + const ReadSettings & read_settings); const MergeTreeData & data; const String name; @@ -64,6 +66,8 @@ private: std::optional part_type; MutableDataPartStoragePtr part_storage; const IMergeTreeDataPart * parent_part = nullptr; + + const ReadSettings read_settings; }; } diff --git a/src/Storages/MergeTree/MergeTreeDataPartCompact.cpp b/src/Storages/MergeTree/MergeTreeDataPartCompact.cpp index fd46b3b9540..14c2da82de1 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartCompact.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartCompact.cpp @@ -136,6 +136,32 @@ void MergeTreeDataPartCompact::loadIndexGranularity() loadIndexGranularityImpl(index_granularity, index_granularity_info, columns.size(), getDataPartStorage()); } +void MergeTreeDataPartCompact::loadMarksToCache(const Names & column_names, MarkCache * mark_cache) const +{ + if (column_names.empty() || !mark_cache) + return; + + auto context = storage.getContext(); + auto read_settings = context->getReadSettings(); + auto * load_marks_threadpool = read_settings.load_marks_asynchronously ? &context->getLoadMarksThreadpool() : nullptr; + auto info_for_read = std::make_shared(shared_from_this(), std::make_shared()); + + LOG_TEST(getLogger("MergeTreeDataPartCompact"), "Loading marks into mark cache for columns {} of part {}", toString(column_names), name); + + MergeTreeMarksLoader loader( + info_for_read, + mark_cache, + index_granularity_info.getMarksFilePath(DATA_FILE_NAME), + index_granularity.getMarksCount(), + index_granularity_info, + /*save_marks_in_cache=*/ true, + read_settings, + load_marks_threadpool, + columns.size()); + + loader.loadMarks(); +} + bool MergeTreeDataPartCompact::hasColumnFiles(const NameAndTypePair & column) const { if (!getColumnPosition(column.getNameInStorage())) @@ -230,7 +256,14 @@ bool MergeTreeDataPartCompact::isStoredOnRemoteDiskWithZeroCopySupport() const MergeTreeDataPartCompact::~MergeTreeDataPartCompact() { - removeIfNeeded(); + try + { + removeIfNeeded(); + } + catch (...) + { + tryLogCurrentException(__PRETTY_FUNCTION__); + } } } diff --git a/src/Storages/MergeTree/MergeTreeDataPartCompact.h b/src/Storages/MergeTree/MergeTreeDataPartCompact.h index 9512485c54e..8e279571578 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartCompact.h +++ b/src/Storages/MergeTree/MergeTreeDataPartCompact.h @@ -54,6 +54,8 @@ public: std::optional getFileNameForColumn(const NameAndTypePair & /* column */) const override { return DATA_FILE_NAME; } + void loadMarksToCache(const Names & column_names, MarkCache * mark_cache) const override; + ~MergeTreeDataPartCompact() override; protected: diff --git a/src/Storages/MergeTree/MergeTreeDataPartType.h b/src/Storages/MergeTree/MergeTreeDataPartType.h index 8177809d41e..a59ccc2fab1 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartType.h +++ b/src/Storages/MergeTree/MergeTreeDataPartType.h @@ -45,6 +45,7 @@ public: enum Value { Full, + Packed, Unknown, }; diff --git a/src/Storages/MergeTree/MergeTreeDataPartWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWide.cpp index 9bbf0ad9739..c515d645253 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWide.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWide.cpp @@ -182,6 +182,47 @@ void MergeTreeDataPartWide::loadIndexGranularity() loadIndexGranularityImpl(index_granularity, index_granularity_info, getDataPartStorage(), *any_column_filename); } +void MergeTreeDataPartWide::loadMarksToCache(const Names & column_names, MarkCache * mark_cache) const +{ + if (column_names.empty() || !mark_cache) + return; + + std::vector> loaders; + + auto context = storage.getContext(); + auto read_settings = context->getReadSettings(); + auto * load_marks_threadpool = read_settings.load_marks_asynchronously ? &context->getLoadMarksThreadpool() : nullptr; + auto info_for_read = std::make_shared(shared_from_this(), std::make_shared()); + + LOG_TEST(getLogger("MergeTreeDataPartWide"), "Loading marks into mark cache for columns {} of part {}", toString(column_names), name); + + for (const auto & column_name : column_names) + { + auto serialization = getSerialization(column_name); + serialization->enumerateStreams([&](const auto & subpath) + { + auto stream_name = getStreamNameForColumn(column_name, subpath, checksums); + if (!stream_name) + return; + + loaders.emplace_back(std::make_unique( + info_for_read, + mark_cache, + index_granularity_info.getMarksFilePath(*stream_name), + index_granularity.getMarksCount(), + index_granularity_info, + /*save_marks_in_cache=*/ true, + read_settings, + load_marks_threadpool, + /*num_columns_in_mark=*/ 1)); + + loaders.back()->startAsyncLoad(); + }); + } + + for (auto & loader : loaders) + loader->loadMarks(); +} bool MergeTreeDataPartWide::isStoredOnRemoteDisk() const { @@ -200,7 +241,14 @@ bool MergeTreeDataPartWide::isStoredOnRemoteDiskWithZeroCopySupport() const MergeTreeDataPartWide::~MergeTreeDataPartWide() { - removeIfNeeded(); + try + { + removeIfNeeded(); + } + catch (...) + { + tryLogCurrentException(__PRETTY_FUNCTION__); + } } void MergeTreeDataPartWide::doCheckConsistency(bool require_part_metadata) const diff --git a/src/Storages/MergeTree/MergeTreeDataPartWide.h b/src/Storages/MergeTree/MergeTreeDataPartWide.h index 42893f47573..022a5fb746c 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWide.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWide.h @@ -51,6 +51,8 @@ public: std::optional getColumnModificationTime(const String & column_name) const override; + void loadMarksToCache(const Names & column_names, MarkCache * mark_cache) const override; + protected: static void loadIndexGranularityImpl( MergeTreeIndexGranularity & index_granularity_, MergeTreeIndexGranularityInfo & index_granularity_info_, diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp index a859172023f..c8d11ced683 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.cpp @@ -1,5 +1,6 @@ #include #include +#include "Formats/MarkInCompressedFile.h" namespace DB { @@ -54,26 +55,15 @@ MergeTreeDataPartWriterCompact::MergeTreeDataPartWriterCompact( marks_source_hashing = std::make_unique(*marks_compressor); } + if (settings.save_marks_in_cache) + { + cached_marks[MergeTreeDataPartCompact::DATA_FILE_NAME] = std::make_unique(); + } + for (const auto & column : columns_list) { auto compression = getCodecDescOrDefault(column.name, default_codec); - addStreams(column, nullptr, compression); - } -} - -void MergeTreeDataPartWriterCompact::initDynamicStreamsIfNeeded(const Block & block) -{ - if (is_dynamic_streams_initialized) - return; - - is_dynamic_streams_initialized = true; - for (const auto & column : columns_list) - { - if (column.type->hasDynamicSubcolumns()) - { - auto compression = getCodecDescOrDefault(column.name, default_codec); - addStreams(column, block.getByName(column.name).column, compression); - } + MergeTreeDataPartWriterCompact::addStreams(column, nullptr, compression); } } @@ -175,20 +165,25 @@ void writeColumnSingleGranule( void MergeTreeDataPartWriterCompact::write(const Block & block, const IColumn::Permutation * permutation) { - /// On first block of data initialize streams for dynamic subcolumns. - initDynamicStreamsIfNeeded(block); + Block result_block = block; + + /// During serialization columns with dynamic subcolumns (like JSON/Dynamic) must have the same dynamic structure. + /// But it may happen that they don't (for example during ALTER MODIFY COLUMN from some type to JSON/Dynamic). + /// In this case we use dynamic structure of the column from the first written block and adjust columns from + /// the next blocks so they match this dynamic structure. + initOrAdjustDynamicStructureIfNeeded(result_block); /// Fill index granularity for this block /// if it's unknown (in case of insert data or horizontal merge, /// but not in case of vertical merge) if (compute_granularity) { - size_t index_granularity_for_block = computeIndexGranularity(block); + size_t index_granularity_for_block = computeIndexGranularity(result_block); assert(index_granularity_for_block >= 1); - fillIndexGranularity(index_granularity_for_block, block.rows()); + fillIndexGranularity(index_granularity_for_block, result_block.rows()); } - Block result_block = permuteBlockIfNeeded(block, permutation); + result_block = permuteBlockIfNeeded(result_block, permutation); if (!header) header = result_block.cloneEmpty(); @@ -255,9 +250,12 @@ void MergeTreeDataPartWriterCompact::writeDataBlock(const Block & block, const G return &result_stream->hashing_buf; }; + MarkInCompressedFile mark{plain_hashing.count(), static_cast(0)}; + writeBinaryLittleEndian(mark.offset_in_compressed_file, marks_out); + writeBinaryLittleEndian(mark.offset_in_decompressed_block, marks_out); - writeBinaryLittleEndian(plain_hashing.count(), marks_out); - writeBinaryLittleEndian(static_cast(0), marks_out); + if (!cached_marks.empty()) + cached_marks.begin()->second->push_back(mark); writeColumnSingleGranule( block.getByName(name_and_type->name), getSerialization(name_and_type->name), @@ -296,11 +294,17 @@ void MergeTreeDataPartWriterCompact::fillDataChecksums(MergeTreeDataPartChecksum if (with_final_mark && data_written) { + MarkInCompressedFile mark{plain_hashing.count(), 0}; + for (size_t i = 0; i < columns_list.size(); ++i) { - writeBinaryLittleEndian(plain_hashing.count(), marks_out); - writeBinaryLittleEndian(static_cast(0), marks_out); + writeBinaryLittleEndian(mark.offset_in_compressed_file, marks_out); + writeBinaryLittleEndian(mark.offset_in_decompressed_block, marks_out); + + if (!cached_marks.empty()) + cached_marks.begin()->second->push_back(mark); } + writeBinaryLittleEndian(static_cast(0), marks_out); } diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h index b440a37222d..b3e2e78491d 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterCompact.h @@ -32,6 +32,8 @@ public: void fillChecksums(MergeTreeDataPartChecksums & checksums, NameSet & checksums_to_remove) override; void finish(bool sync) override; + size_t getNumberOfOpenStreams() const override { return 1; } + private: /// Finish serialization of the data. Flush rows in buffer to disk, compute checksums. void fillDataChecksums(MergeTreeDataPartChecksums & checksums); @@ -48,9 +50,7 @@ private: void addToChecksums(MergeTreeDataPartChecksums & checksums); - void addStreams(const NameAndTypePair & name_and_type, const ColumnPtr & column, const ASTPtr & effective_codec_desc); - - void initDynamicStreamsIfNeeded(const Block & block); + void addStreams(const NameAndTypePair & name_and_type, const ColumnPtr & column, const ASTPtr & effective_codec_desc) override; Block header; @@ -104,8 +104,6 @@ private: /// then finally to 'marks_file'. std::unique_ptr marks_compressor; std::unique_ptr marks_source_hashing; - - bool is_dynamic_streams_initialized = false; }; } diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp index 89db8174636..c483d47fed7 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.cpp @@ -3,26 +3,16 @@ #include #include #include -#include #include - namespace ProfileEvents { - extern const Event MergeTreeDataWriterSkipIndicesCalculationMicroseconds; - extern const Event MergeTreeDataWriterStatisticsCalculationMicroseconds; -} - -namespace CurrentMetrics -{ - extern const Metric CompressionThread; - extern const Metric CompressionThreadActive; - extern const Metric CompressionThreadScheduled; +extern const Event MergeTreeDataWriterSkipIndicesCalculationMicroseconds; +extern const Event MergeTreeDataWriterStatisticsCalculationMicroseconds; } namespace DB { - namespace MergeTreeSetting { extern const MergeTreeSettingsUInt64 index_granularity; @@ -35,53 +25,57 @@ namespace ErrorCodes extern const int LOGICAL_ERROR; } -void MergeTreeDataPartWriterOnDisk::Stream::preFinalize() +template +void MergeTreeDataPartWriterOnDisk::Stream::preFinalize() { /// Here the main goal is to do preFinalize calls for plain_file and marks_file /// Before that all hashing and compression buffers have to be finalized /// Otherwise some data might stuck in the buffers above plain_file and marks_file /// Also the order is important - compressed_hashing->finalize(); - compressor->finalize(); - plain_hashing->finalize(); + compressed_hashing.finalize(); + compressor.finalize(); + plain_hashing.finalize(); - if (marks_hashing) + if constexpr (!only_plain_file) { if (compress_marks) { - marks_compressed_hashing->finalize(); - marks_compressor->finalize(); + marks_compressed_hashing.finalize(); + marks_compressor.finalize(); } - marks_hashing->finalize(); + marks_hashing.finalize(); } plain_file->preFinalize(); - if (marks_file) + if constexpr (!only_plain_file) marks_file->preFinalize(); is_prefinalized = true; } -void MergeTreeDataPartWriterOnDisk::Stream::finalize() +template +void MergeTreeDataPartWriterOnDisk::Stream::finalize() { if (!is_prefinalized) preFinalize(); plain_file->finalize(); - if (marks_file) + if constexpr (!only_plain_file) marks_file->finalize(); } -void MergeTreeDataPartWriterOnDisk::Stream::sync() const +template +void MergeTreeDataPartWriterOnDisk::Stream::sync() const { plain_file->sync(); - if (marks_file) + if constexpr (!only_plain_file) marks_file->sync(); } -MergeTreeDataPartWriterOnDisk::Stream::Stream( +template<> +MergeTreeDataPartWriterOnDisk::Stream::Stream( const String & escaped_column_name_, const MutableDataPartStoragePtr & data_part_storage, const String & data_path_, @@ -96,45 +90,20 @@ MergeTreeDataPartWriterOnDisk::Stream::Stream( escaped_column_name(escaped_column_name_), data_file_extension{data_file_extension_}, marks_file_extension{marks_file_extension_}, + plain_file(data_part_storage->writeFile(data_path_ + data_file_extension, max_compress_block_size_, query_write_settings)), + plain_hashing(*plain_file), + compressor(plain_hashing, compression_codec_, max_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size), + compressed_hashing(compressor), + marks_file(data_part_storage->writeFile(marks_path_ + marks_file_extension, 4096, query_write_settings)), + marks_hashing(*marks_file), + marks_compressor(marks_hashing, marks_compression_codec_, marks_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size), + marks_compressed_hashing(marks_compressor), compress_marks(MarkType(marks_file_extension).compressed) { - plain_file = data_part_storage->writeFile(data_path_ + data_file_extension, max_compress_block_size_, query_write_settings); - plain_hashing.emplace(*plain_file); - - if (query_write_settings.max_compression_threads > 1) - { - compression_thread_pool.emplace( - CurrentMetrics::CompressionThread, CurrentMetrics::CompressionThreadActive, CurrentMetrics::CompressionThreadScheduled, - query_write_settings.max_compression_threads); - - compressor = std::make_unique( - *plain_hashing, - compression_codec_, - max_compress_block_size_, - query_write_settings.max_compression_threads, - *compression_thread_pool); - - is_compressor_parallel = true; - } - else - { - compressor = std::make_unique( - *plain_hashing, - compression_codec_, - max_compress_block_size_, - query_write_settings.use_adaptive_write_buffer, - query_write_settings.adaptive_write_buffer_initial_size); - } - - compressed_hashing.emplace(*compressor); - - marks_file = data_part_storage->writeFile(marks_path_ + marks_file_extension, 4096, query_write_settings); - marks_hashing.emplace(*marks_file); - marks_compressor.emplace(*marks_hashing, marks_compression_codec_, marks_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size); - marks_compressed_hashing.emplace(*marks_compressor); } -MergeTreeDataPartWriterOnDisk::Stream::Stream( +template<> +MergeTreeDataPartWriterOnDisk::Stream::Stream( const String & escaped_column_name_, const MutableDataPartStoragePtr & data_part_storage, const String & data_path_, @@ -146,33 +115,34 @@ MergeTreeDataPartWriterOnDisk::Stream::Stream( data_file_extension{data_file_extension_}, plain_file(data_part_storage->writeFile(data_path_ + data_file_extension, max_compress_block_size_, query_write_settings)), plain_hashing(*plain_file), - compressor(std::make_unique(*plain_hashing, compression_codec_, max_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size)), - compressed_hashing(*compressor), + compressor(plain_hashing, compression_codec_, max_compress_block_size_, query_write_settings.use_adaptive_write_buffer, query_write_settings.adaptive_write_buffer_initial_size), + compressed_hashing(compressor), compress_marks(false) { } -void MergeTreeDataPartWriterOnDisk::Stream::addToChecksums(MergeTreeData::DataPart::Checksums & checksums) +template +void MergeTreeDataPartWriterOnDisk::Stream::addToChecksums(MergeTreeData::DataPart::Checksums & checksums) { String name = escaped_column_name; checksums.files[name + data_file_extension].is_compressed = true; - checksums.files[name + data_file_extension].uncompressed_size = compressed_hashing->count(); - checksums.files[name + data_file_extension].uncompressed_hash = compressed_hashing->getHash(); - checksums.files[name + data_file_extension].file_size = plain_hashing->count(); - checksums.files[name + data_file_extension].file_hash = plain_hashing->getHash(); + checksums.files[name + data_file_extension].uncompressed_size = compressed_hashing.count(); + checksums.files[name + data_file_extension].uncompressed_hash = compressed_hashing.getHash(); + checksums.files[name + data_file_extension].file_size = plain_hashing.count(); + checksums.files[name + data_file_extension].file_hash = plain_hashing.getHash(); - if (marks_hashing) + if constexpr (!only_plain_file) { if (compress_marks) { checksums.files[name + marks_file_extension].is_compressed = true; - checksums.files[name + marks_file_extension].uncompressed_size = marks_compressed_hashing->count(); - checksums.files[name + marks_file_extension].uncompressed_hash = marks_compressed_hashing->getHash(); + checksums.files[name + marks_file_extension].uncompressed_size = marks_compressed_hashing.count(); + checksums.files[name + marks_file_extension].uncompressed_hash = marks_compressed_hashing.getHash(); } - checksums.files[name + marks_file_extension].file_size = marks_hashing->count(); - checksums.files[name + marks_file_extension].file_hash = marks_hashing->getHash(); + checksums.files[name + marks_file_extension].file_size = marks_hashing.count(); + checksums.files[name + marks_file_extension].file_hash = marks_hashing.getHash(); } } @@ -209,8 +179,8 @@ MergeTreeDataPartWriterOnDisk::MergeTreeDataPartWriterOnDisk( throw Exception(ErrorCodes::LOGICAL_ERROR, "Can't take information about index granularity from blocks, when non empty index_granularity array specified"); - if (!getDataPartStorage().exists()) - getDataPartStorage().createDirectories(); + /// We don't need to check if it exists or not, createDirectories doesn't throw + getDataPartStorage().createDirectories(); if (settings.rewrite_primary_key) initPrimaryIndex(); @@ -306,12 +276,12 @@ void MergeTreeDataPartWriterOnDisk::initStatistics() for (const auto & stat_ptr : stats) { String stats_name = stat_ptr->getFileName(); - stats_streams.emplace_back(std::make_unique( - stats_name, - data_part_storage, - stats_name, STATS_FILE_SUFFIX, - default_codec, settings.max_compress_block_size, - settings.query_write_settings)); + stats_streams.emplace_back(std::make_unique>( + stats_name, + data_part_storage, + stats_name, STATS_FILE_SUFFIX, + default_codec, settings.max_compress_block_size, + settings.query_write_settings)); } } @@ -328,14 +298,14 @@ void MergeTreeDataPartWriterOnDisk::initSkipIndices() { String stream_name = skip_index->getFileName(); skip_indices_streams.emplace_back( - std::make_unique( - stream_name, - data_part_storage, - stream_name, skip_index->getSerializedFileExtension(), - stream_name, marks_file_extension, - default_codec, settings.max_compress_block_size, - marks_compression_codec, settings.marks_compress_block_size, - settings.query_write_settings)); + std::make_unique>( + stream_name, + data_part_storage, + stream_name, skip_index->getSerializedFileExtension(), + stream_name, marks_file_extension, + default_codec, settings.max_compress_block_size, + marks_compression_codec, settings.marks_compress_block_size, + settings.query_write_settings)); GinIndexStorePtr store = nullptr; if (typeid_cast(&*skip_index) != nullptr) @@ -411,7 +381,7 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializeSkipIndices(const Block { const auto index_helper = skip_indices[i]; auto & stream = *skip_indices_streams[i]; - WriteBuffer & marks_out = stream.compress_marks ? *stream.marks_compressed_hashing : *stream.marks_hashing; + WriteBuffer & marks_out = stream.compress_marks ? stream.marks_compressed_hashing : stream.marks_hashing; GinIndexStorePtr store; if (typeid_cast(&*index_helper) != nullptr) @@ -427,7 +397,7 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializeSkipIndices(const Block { if (skip_index_accumulated_marks[i] == index_helper->index.granularity) { - skip_indices_aggregators[i]->getGranuleAndReset()->serializeBinary(*stream.compressed_hashing); + skip_indices_aggregators[i]->getGranuleAndReset()->serializeBinary(stream.compressed_hashing); skip_index_accumulated_marks[i] = 0; } @@ -435,11 +405,11 @@ void MergeTreeDataPartWriterOnDisk::calculateAndSerializeSkipIndices(const Block { skip_indices_aggregators[i] = index_helper->createIndexAggregatorForPart(store, settings); - if (stream.compressed_hashing->offset() >= settings.min_compress_block_size) - stream.compressed_hashing->next(); + if (stream.compressed_hashing.offset() >= settings.min_compress_block_size) + stream.compressed_hashing.next(); - writeBinaryLittleEndian(stream.plain_hashing->count(), marks_out); - writeBinaryLittleEndian(stream.compressed_hashing->offset(), marks_out); + writeBinaryLittleEndian(stream.plain_hashing.count(), marks_out); + writeBinaryLittleEndian(stream.compressed_hashing.offset(), marks_out); /// Actually this numbers is redundant, but we have to store them /// to be compatible with the normal .mrk2 file format @@ -519,7 +489,7 @@ void MergeTreeDataPartWriterOnDisk::fillSkipIndicesChecksums(MergeTreeData::Data { auto & stream = *skip_indices_streams[i]; if (!skip_indices_aggregators[i]->empty()) - skip_indices_aggregators[i]->getGranuleAndReset()->serializeBinary(*stream.compressed_hashing); + skip_indices_aggregators[i]->getGranuleAndReset()->serializeBinary(stream.compressed_hashing); /// Register additional files written only by the full-text index. Required because otherwise DROP TABLE complains about unknown /// files. Note that the provided actual checksums are bogus. The problem is that at this point the file writes happened already and @@ -559,7 +529,7 @@ void MergeTreeDataPartWriterOnDisk::fillStatisticsChecksums(MergeTreeData::DataP for (size_t i = 0; i < stats.size(); i++) { auto & stream = *stats_streams[i]; - stats[i]->serialize(*stream.compressed_hashing); + stats[i]->serialize(stream.compressed_hashing); stream.preFinalize(); stream.addToChecksums(checksums); } @@ -594,4 +564,46 @@ Names MergeTreeDataPartWriterOnDisk::getSkipIndicesColumns() const return Names(skip_indexes_column_names_set.begin(), skip_indexes_column_names_set.end()); } +void MergeTreeDataPartWriterOnDisk::initOrAdjustDynamicStructureIfNeeded(Block & block) +{ + if (!is_dynamic_streams_initialized) + { + for (const auto & column : columns_list) + { + if (column.type->hasDynamicSubcolumns()) + { + /// Create all streams for dynamic subcolumns using dynamic structure from block. + auto compression = getCodecDescOrDefault(column.name, default_codec); + addStreams(column, block.getByName(column.name).column, compression); + } + } + is_dynamic_streams_initialized = true; + block_sample = block.cloneEmpty(); + } + else + { + size_t size = block.columns(); + for (size_t i = 0; i != size; ++i) + { + auto & column = block.getByPosition(i); + const auto & sample_column = block_sample.getByPosition(i); + /// Check if the dynamic structure of this column is different from the sample column. + if (column.type->hasDynamicSubcolumns() && !column.column->dynamicStructureEquals(*sample_column.column)) + { + /// We need to change the dynamic structure of the column so it matches the sample column. + /// To do it, we create empty column of this type, take dynamic structure from sample column + /// and insert data into it. Resulting column will have required dynamic structure and the content + /// of the column in current block. + auto new_column = sample_column.type->createColumn(); + new_column->takeDynamicStructureFromSourceColumns({sample_column.column}); + new_column->insertRangeFrom(*column.column, 0, column.column->size()); + column.column = std::move(new_column); + } + } + } +} + +template struct MergeTreeDataPartWriterOnDisk::Stream; +template struct MergeTreeDataPartWriterOnDisk::Stream; + } diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h index cb46785ccbd..49d654c15e1 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterOnDisk.h @@ -8,6 +8,7 @@ #include #include #include +#include namespace DB { @@ -27,7 +28,7 @@ struct Granule /// this granule can be continuation of the previous one. bool mark_on_start; /// if true: When this granule will be written to disk all rows for corresponding mark will - /// be written. It doesn't mean that rows_to_write == index_granularity.getMarkRows(mark_number), + /// be wrtten. It doesn't mean that rows_to_write == index_granularity.getMarkRows(mark_number), /// We may have a lot of small blocks between two marks and this may be the last one. bool is_complete; }; @@ -44,6 +45,7 @@ public: /// Helper class, which holds chain of buffers to write data file with marks. /// It is used to write: one column, skip index or all columns (in compact format). + template struct Stream { Stream( @@ -74,32 +76,30 @@ public: /// compressed_hashing -> compressor -> plain_hashing -> plain_file std::unique_ptr plain_file; - std::optional plain_hashing; - /// This could be either CompressedWriteBuffer or ParallelCompressedWriteBuffer - bool is_compressor_parallel = false; - std::unique_ptr compressor; - std::optional compressed_hashing; + HashingWriteBuffer plain_hashing; + CompressedWriteBuffer compressor; + HashingWriteBuffer compressed_hashing; /// marks_compressed_hashing -> marks_compressor -> marks_hashing -> marks_file std::unique_ptr marks_file; - std::optional marks_hashing; - std::optional marks_compressor; - std::optional marks_compressed_hashing; + std::conditional_t marks_hashing; + std::conditional_t marks_compressor; + std::conditional_t marks_compressed_hashing; bool compress_marks; bool is_prefinalized = false; - /// Thread pool for parallel compression. - std::optional compression_thread_pool; - void preFinalize(); + void finalize(); + void sync() const; void addToChecksums(MergeTreeDataPartChecksums & checksums); }; - using StreamPtr = std::unique_ptr; + using StreamPtr = std::unique_ptr>; + using StatisticStreamPtr = std::unique_ptr>; MergeTreeDataPartWriterOnDisk( const String & data_part_name_, @@ -154,10 +154,18 @@ protected: /// Get unique non ordered skip indices column. Names getSkipIndicesColumns() const; + virtual void addStreams(const NameAndTypePair & name_and_type, const ColumnPtr & column, const ASTPtr & effective_codec_desc) = 0; + + /// On first block create all required streams for columns with dynamic subcolumns and remember the block sample. + /// On each next block check if dynamic structure of the columns equals to the dynamic structure of the same + /// columns in the sample block. If for some column dynamic structure is different, adjust it so it matches + /// the structure from the sample. + void initOrAdjustDynamicStructureIfNeeded(Block & block); + const MergeTreeIndices skip_indices; const ColumnsStatistics stats; - std::vector stats_streams; + std::vector stats_streams; const String marks_file_extension; const CompressionCodecPtr default_codec; @@ -188,6 +196,10 @@ protected: size_t current_mark = 0; GinIndexStoreFactory::GinIndexStores gin_index_stores; + + bool is_dynamic_streams_initialized = false; + Block block_sample; + private: void initSkipIndices(); void initPrimaryIndex(); diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp index 860722ba870..7c9724b1b75 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.cpp @@ -6,10 +6,11 @@ #include #include #include +#include +#include #include #include - namespace DB { @@ -106,27 +107,16 @@ MergeTreeDataPartWriterWide::MergeTreeDataPartWriterWide( indices_to_recalc_, stats_to_recalc_, marks_file_extension_, default_codec_, settings_, index_granularity_) { + if (settings.save_marks_in_cache) + { + auto columns_vec = getColumnsToPrewarmMarks(*storage_settings, columns_list); + columns_to_load_marks = NameSet(columns_vec.begin(), columns_vec.end()); + } + for (const auto & column : columns_list) { auto compression = getCodecDescOrDefault(column.name, default_codec); - addStreams(column, nullptr, compression); - } -} - -void MergeTreeDataPartWriterWide::initDynamicStreamsIfNeeded(const DB::Block & block) -{ - if (is_dynamic_streams_initialized) - return; - - is_dynamic_streams_initialized = true; - block_sample = block.cloneEmpty(); - for (const auto & column : columns_list) - { - if (column.type->hasDynamicSubcolumns()) - { - auto compression = getCodecDescOrDefault(column.name, default_codec); - addStreams(column, block_sample.getByName(column.name).column, compression); - } + MergeTreeDataPartWriterWide::addStreams(column, nullptr, compression); } } @@ -188,7 +178,7 @@ void MergeTreeDataPartWriterWide::addStreams( query_write_settings.use_adaptive_write_buffer = settings.use_adaptive_write_buffer_for_dynamic_subcolumns && ISerialization::isDynamicSubcolumn(substream_path, substream_path.size()); query_write_settings.adaptive_write_buffer_initial_size = settings.adaptive_write_buffer_initial_size; - column_streams[stream_name] = std::make_unique( + column_streams[stream_name] = std::make_unique>( stream_name, data_part_storage, stream_name, DATA_FILE_EXTENSION, @@ -199,6 +189,9 @@ void MergeTreeDataPartWriterWide::addStreams( settings.marks_compress_block_size, query_write_settings); + if (columns_to_load_marks.contains(name_and_type.name)) + cached_marks.emplace(stream_name, std::make_unique()); + full_name_to_stream_name.emplace(full_stream_name, stream_name); stream_name_to_full_name.emplace(stream_name, full_stream_name); }; @@ -231,7 +224,7 @@ ISerialization::OutputStreamGetter MergeTreeDataPartWriterWide::createStreamGett if (is_offsets && offset_columns.contains(stream_name)) return nullptr; - return &column_streams.at(stream_name)->compressed_hashing.value(); + return &column_streams.at(stream_name)->compressed_hashing; }; } @@ -267,15 +260,20 @@ void MergeTreeDataPartWriterWide::shiftCurrentMark(const Granules & granules_wri void MergeTreeDataPartWriterWide::write(const Block & block, const IColumn::Permutation * permutation) { - /// On first block of data initialize streams for dynamic subcolumns. - initDynamicStreamsIfNeeded(block); + Block block_to_write = block; + + /// During serialization columns with dynamic subcolumns (like JSON/Dynamic) must have the same dynamic structure. + /// But it may happen that they don't (for example during ALTER MODIFY COLUMN from some type to JSON/Dynamic). + /// In this case we use dynamic structure of the column from the first written block and adjust columns from + /// the next blocks so they match this dynamic structure. + initOrAdjustDynamicStructureIfNeeded(block_to_write); /// Fill index granularity for this block /// if it's unknown (in case of insert data or horizontal merge, /// but not in case of vertical part of vertical merge) if (compute_granularity) { - size_t index_granularity_for_block = computeIndexGranularity(block); + size_t index_granularity_for_block = computeIndexGranularity(block_to_write); if (rows_written_in_last_mark > 0) { size_t rows_left_in_last_mark = index_granularity.getMarkRows(getCurrentMark()) - rows_written_in_last_mark; @@ -293,11 +291,9 @@ void MergeTreeDataPartWriterWide::write(const Block & block, const IColumn::Perm } } - fillIndexGranularity(index_granularity_for_block, block.rows()); + fillIndexGranularity(index_granularity_for_block, block_to_write.rows()); } - Block block_to_write = block; - auto granules_to_write = getGranulesToWrite(index_granularity, block_to_write.rows(), getCurrentMark(), rows_written_in_last_mark); auto offset_columns = written_offset_columns ? *written_offset_columns : WrittenOffsetColumns{}; @@ -363,12 +359,16 @@ void MergeTreeDataPartWriterWide::writeSingleMark( void MergeTreeDataPartWriterWide::flushMarkToFile(const StreamNameAndMark & stream_with_mark, size_t rows_in_mark) { auto & stream = *column_streams[stream_with_mark.stream_name]; - WriteBuffer & marks_out = stream.compress_marks ? *stream.marks_compressed_hashing : *stream.marks_hashing; + WriteBuffer & marks_out = stream.compress_marks ? stream.marks_compressed_hashing : stream.marks_hashing; writeBinaryLittleEndian(stream_with_mark.mark.offset_in_compressed_file, marks_out); writeBinaryLittleEndian(stream_with_mark.mark.offset_in_decompressed_block, marks_out); + if (settings.can_use_adaptive_granularity) writeBinaryLittleEndian(rows_in_mark, marks_out); + + if (auto it = cached_marks.find(stream_with_mark.stream_name); it != cached_marks.end()) + it->second->push_back(stream_with_mark.mark); } StreamsWithMarks MergeTreeDataPartWriterWide::getCurrentMarksForColumn( @@ -400,22 +400,15 @@ StreamsWithMarks MergeTreeDataPartWriterWide::getCurrentMarksForColumn( auto & stream = *column_streams[stream_name]; /// There could already be enough data to compress into the new block. - auto push_mark = [&] - { - StreamNameAndMark stream_with_mark; - stream_with_mark.stream_name = stream_name; - stream_with_mark.mark.offset_in_compressed_file = stream.plain_hashing->count(); - stream_with_mark.mark.offset_in_decompressed_block = stream.compressed_hashing->offset(); - result.push_back(stream_with_mark); - }; + if (stream.compressed_hashing.offset() >= min_compress_block_size) + stream.compressed_hashing.next(); - if (stream.compressed_hashing->offset() >= min_compress_block_size) - { + StreamNameAndMark stream_with_mark; + stream_with_mark.stream_name = stream_name; + stream_with_mark.mark.offset_in_compressed_file = stream.plain_hashing.count(); + stream_with_mark.mark.offset_in_decompressed_block = stream.compressed_hashing.offset(); - stream.compressed_hashing->next(); - } - - push_mark(); + result.push_back(stream_with_mark); }, name_and_type.type, column_sample); return result; @@ -446,7 +439,7 @@ void MergeTreeDataPartWriterWide::writeSingleGranule( if (is_offsets && offset_columns.contains(stream_name)) return; - column_streams.at(stream_name)->compressed_hashing->nextIfAtEnd(); + column_streams.at(stream_name)->compressed_hashing.nextIfAtEnd(); }, name_and_type.type, column.getPtr()); } @@ -750,7 +743,6 @@ void MergeTreeDataPartWriterWide::fillChecksums(MergeTreeDataPartChecksums & che fillPrimaryIndexChecksums(checksums); fillSkipIndicesChecksums(checksums); - fillStatisticsChecksums(checksums); } @@ -764,7 +756,6 @@ void MergeTreeDataPartWriterWide::finish(bool sync) finishPrimaryIndexSerialization(sync); finishSkipIndicesSerialization(sync); - finishStatisticsSerialization(sync); } diff --git a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h index ab86ed27c7e..19304b28c6c 100644 --- a/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h +++ b/src/Storages/MergeTree/MergeTreeDataPartWriterWide.h @@ -43,6 +43,8 @@ public: void finish(bool sync) final; + size_t getNumberOfOpenStreams() const override { return column_streams.size(); } + private: /// Finish serialization of data: write final mark if required and compute checksums /// Also validate written data in debug mode @@ -91,9 +93,7 @@ private: void addStreams( const NameAndTypePair & name_and_type, const ColumnPtr & column, - const ASTPtr & effective_codec_desc); - - void initDynamicStreamsIfNeeded(const Block & block); + const ASTPtr & effective_codec_desc) override; /// Method for self check (used in debug-build only). Checks that written /// data and corresponding marks are consistent. Otherwise throws logical @@ -136,13 +136,12 @@ private: using MarksForColumns = std::unordered_map; MarksForColumns last_non_written_marks; + /// Set of columns to put marks in cache during write. + NameSet columns_to_load_marks; + /// How many rows we have already written in the current mark. /// More than zero when incoming blocks are smaller then their granularity. size_t rows_written_in_last_mark = 0; - - Block block_sample; - - bool is_dynamic_streams_initialized = false; }; } diff --git a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp index 13918ae8e91..1b3c58000e7 100644 --- a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp +++ b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp @@ -71,10 +71,7 @@ namespace Setting extern const SettingsString force_data_skipping_indices; extern const SettingsBool force_index_by_date; extern const SettingsSeconds lock_acquire_timeout; - extern const SettingsUInt64 max_parser_backtracks; - extern const SettingsUInt64 max_parser_depth; extern const SettingsInt64 max_partitions_to_read; - extern const SettingsUInt64 max_query_size; extern const SettingsUInt64 max_threads_for_indexes; extern const SettingsNonZeroUInt64 max_parallel_replicas; extern const SettingsUInt64 merge_tree_coarse_index_granularity; @@ -640,20 +637,11 @@ RangesInDataParts MergeTreeDataSelectExecutor::filterPartsByPrimaryKeyAndSkipInd if (use_skip_indexes && settings[Setting::force_data_skipping_indices].changed) { - const auto & indices = settings[Setting::force_data_skipping_indices].toString(); - - Strings forced_indices; - { - Tokens tokens(indices.data(), indices.data() + indices.size(), settings[Setting::max_query_size]); - IParser::Pos pos( - tokens, static_cast(settings[Setting::max_parser_depth]), static_cast(settings[Setting::max_parser_backtracks])); - Expected expected; - if (!parseIdentifiersOrStringLiterals(pos, expected, forced_indices)) - throw Exception(ErrorCodes::CANNOT_PARSE_TEXT, "Cannot parse force_data_skipping_indices ('{}')", indices); - } + const auto & indices_str = settings[Setting::force_data_skipping_indices].toString(); + auto forced_indices = parseIdentifiersOrStringLiterals(indices_str, settings); if (forced_indices.empty()) - throw Exception(ErrorCodes::CANNOT_PARSE_TEXT, "No indices parsed from force_data_skipping_indices ('{}')", indices); + throw Exception(ErrorCodes::CANNOT_PARSE_TEXT, "No indices parsed from force_data_skipping_indices ('{}')", indices_str); std::unordered_set useful_indices_names; for (const auto & useful_index : skip_indexes.useful_indices) @@ -1022,11 +1010,7 @@ size_t MergeTreeDataSelectExecutor::roundRowsOrBytesToMarks( /// Same as roundRowsOrBytesToMarks() but do not return more then max_marks size_t MergeTreeDataSelectExecutor::minMarksForConcurrentRead( - size_t rows_setting, - size_t bytes_setting, - size_t rows_granularity, - size_t bytes_granularity, - size_t max_marks) + size_t rows_setting, size_t bytes_setting, size_t rows_granularity, size_t bytes_granularity, size_t min_marks, size_t max_marks) { size_t marks = 1; @@ -1035,18 +1019,17 @@ size_t MergeTreeDataSelectExecutor::minMarksForConcurrentRead( else if (rows_setting) marks = (rows_setting + rows_granularity - 1) / rows_granularity; - if (bytes_granularity == 0) - return marks; - - /// Overflow - if (bytes_setting + bytes_granularity <= bytes_setting) /// overflow - return max_marks; - if (bytes_setting) - return std::max(marks, (bytes_setting + bytes_granularity - 1) / bytes_granularity); - return marks; + if (bytes_granularity) + { + /// Overflow + if (bytes_setting + bytes_granularity <= bytes_setting) /// overflow + marks = max_marks; + else if (bytes_setting) + marks = std::max(marks, (bytes_setting + bytes_granularity - 1) / bytes_granularity); + } + return std::max(marks, min_marks); } - /// Calculates a set of mark ranges, that could possibly contain keys, required by condition. /// In other words, it removes subranges from whole range, that definitely could not contain required keys. /// If @exact_ranges is not null, fill it with ranges containing marks of fully matched records. diff --git a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.h b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.h index 70536b7aa54..d16d9243c14 100644 --- a/src/Storages/MergeTree/MergeTreeDataSelectExecutor.h +++ b/src/Storages/MergeTree/MergeTreeDataSelectExecutor.h @@ -153,11 +153,7 @@ public: /// The same as roundRowsOrBytesToMarks, but return no more than max_marks. static size_t minMarksForConcurrentRead( - size_t rows_setting, - size_t bytes_setting, - size_t rows_granularity, - size_t bytes_granularity, - size_t max_marks); + size_t rows_setting, size_t bytes_setting, size_t rows_granularity, size_t bytes_granularity, size_t min_marks, size_t max_marks); /// If possible, construct optional key condition from predicates containing _part_offset column. static void buildKeyConditionFromPartOffset( diff --git a/src/Storages/MergeTree/MergeTreeDataWriter.cpp b/src/Storages/MergeTree/MergeTreeDataWriter.cpp index 67fef759ed4..6d19f45e2c4 100644 --- a/src/Storages/MergeTree/MergeTreeDataWriter.cpp +++ b/src/Storages/MergeTree/MergeTreeDataWriter.cpp @@ -73,6 +73,7 @@ namespace MergeTreeSetting extern const MergeTreeSettingsFloat min_free_disk_ratio_to_perform_insert; extern const MergeTreeSettingsBool optimize_row_order; extern const MergeTreeSettingsFloat ratio_of_defaults_for_sparse_serialization; + extern const MergeTreeSettingsBool prewarm_mark_cache; } namespace ErrorCodes @@ -609,7 +610,7 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl( } } - auto new_data_part = data.getDataPartBuilder(part_name, data_part_volume, part_dir) + auto new_data_part = data.getDataPartBuilder(part_name, data_part_volume, part_dir, getReadSettings()) .withPartFormat(data.choosePartFormat(expected_size, block.rows())) .withPartInfo(new_part_info) .build(); @@ -684,6 +685,7 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl( /// This effectively chooses minimal compression method: /// either default lz4 or compression method with zero thresholds on absolute and relative part size. auto compression_codec = data.getContext()->chooseCompressionCodec(0, 0); + bool save_marks_in_cache = (*data_settings)[MergeTreeSetting::prewarm_mark_cache] && data.getContext()->getMarkCache(); auto out = std::make_unique( new_data_part, @@ -693,8 +695,9 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeTempPartImpl( statistics, compression_codec, context->getCurrentTransaction() ? context->getCurrentTransaction()->tid : Tx::PrehistoricTID, - false, - false, + /*reset_columns=*/ false, + save_marks_in_cache, + /*blocks_are_granules_size=*/ false, context->getWriteSettings()); out->writeWithPermutation(block, perm_ptr); @@ -829,6 +832,7 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeProjectionPartImpl( /// This effectively chooses minimal compression method: /// either default lz4 or compression method with zero thresholds on absolute and relative part size. auto compression_codec = data.getContext()->chooseCompressionCodec(0, 0); + bool save_marks_in_cache = (*data.getSettings())[MergeTreeSetting::prewarm_mark_cache] && data.getContext()->getMarkCache(); auto out = std::make_unique( new_data_part, @@ -839,7 +843,10 @@ MergeTreeDataWriter::TemporaryPart MergeTreeDataWriter::writeProjectionPartImpl( ColumnsStatistics{}, compression_codec, Tx::PrehistoricTID, - false, false, data.getContext()->getWriteSettings()); + /*reset_columns=*/ false, + save_marks_in_cache, + /*blocks_are_granules_size=*/ false, + data.getContext()->getWriteSettings()); out->writeWithPermutation(block, perm_ptr); auto finalizer = out->finalizePartAsync(new_data_part, false); diff --git a/src/Storages/MergeTree/MergeTreeIOSettings.cpp b/src/Storages/MergeTree/MergeTreeIOSettings.cpp index 6705d75af41..bacfbbd5720 100644 --- a/src/Storages/MergeTree/MergeTreeIOSettings.cpp +++ b/src/Storages/MergeTree/MergeTreeIOSettings.cpp @@ -26,7 +26,6 @@ namespace MergeTreeSetting extern const MergeTreeSettingsString primary_key_compression_codec; extern const MergeTreeSettingsBool use_adaptive_write_buffer_for_dynamic_subcolumns; extern const MergeTreeSettingsBool use_compact_variant_discriminators_serialization; - extern const MergeTreeSettingsUInt64 max_compression_threads; } MergeTreeWriterSettings::MergeTreeWriterSettings( @@ -35,6 +34,7 @@ MergeTreeWriterSettings::MergeTreeWriterSettings( const MergeTreeSettingsPtr & storage_settings, bool can_use_adaptive_granularity_, bool rewrite_primary_key_, + bool save_marks_in_cache_, bool blocks_are_granules_size_) : min_compress_block_size( (*storage_settings)[MergeTreeSetting::min_compress_block_size] ? (*storage_settings)[MergeTreeSetting::min_compress_block_size] : global_settings[Setting::min_compress_block_size]) @@ -47,6 +47,7 @@ MergeTreeWriterSettings::MergeTreeWriterSettings( , primary_key_compress_block_size((*storage_settings)[MergeTreeSetting::primary_key_compress_block_size]) , can_use_adaptive_granularity(can_use_adaptive_granularity_) , rewrite_primary_key(rewrite_primary_key_) + , save_marks_in_cache(save_marks_in_cache_) , blocks_are_granules_size(blocks_are_granules_size_) , query_write_settings(query_write_settings_) , low_cardinality_max_dictionary_size(global_settings[Setting::low_cardinality_max_dictionary_size]) @@ -55,7 +56,6 @@ MergeTreeWriterSettings::MergeTreeWriterSettings( , use_adaptive_write_buffer_for_dynamic_subcolumns((*storage_settings)[MergeTreeSetting::use_adaptive_write_buffer_for_dynamic_subcolumns]) , adaptive_write_buffer_initial_size((*storage_settings)[MergeTreeSetting::adaptive_write_buffer_initial_size]) { - query_write_settings.max_compression_threads = (*storage_settings)[MergeTreeSetting::max_compression_threads]; } } diff --git a/src/Storages/MergeTree/MergeTreeIOSettings.h b/src/Storages/MergeTree/MergeTreeIOSettings.h index fcc72815d8f..4d1d2533729 100644 --- a/src/Storages/MergeTree/MergeTreeIOSettings.h +++ b/src/Storages/MergeTree/MergeTreeIOSettings.h @@ -2,6 +2,7 @@ #include #include #include +#include #include @@ -60,7 +61,8 @@ struct MergeTreeWriterSettings const MergeTreeSettingsPtr & storage_settings, bool can_use_adaptive_granularity_, bool rewrite_primary_key_, - bool blocks_are_granules_size_ = false); + bool save_marks_in_cache_, + bool blocks_are_granules_size_); size_t min_compress_block_size; size_t max_compress_block_size; @@ -74,6 +76,7 @@ struct MergeTreeWriterSettings bool can_use_adaptive_granularity; bool rewrite_primary_key; + bool save_marks_in_cache; bool blocks_are_granules_size; WriteSettings query_write_settings; diff --git a/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp b/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp index 467d2567df1..d69a00643f0 100644 --- a/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp @@ -96,29 +96,13 @@ size_t MergeTreeIndexGranularity::countMarksForRows(size_t from_mark, size_t num return to_mark - from_mark; } -size_t MergeTreeIndexGranularity::countRowsForRows(size_t from_mark, size_t number_of_rows, size_t offset_in_rows, size_t min_marks_to_read) const +size_t MergeTreeIndexGranularity::countRowsForRows(size_t from_mark, size_t number_of_rows, size_t offset_in_rows) const { size_t rows_before_mark = getMarkStartingRow(from_mark); size_t last_row_pos = rows_before_mark + offset_in_rows + number_of_rows; auto it = std::upper_bound(marks_rows_partial_sums.begin(), marks_rows_partial_sums.end(), last_row_pos); size_t to_mark = it - marks_rows_partial_sums.begin(); - /// This is a heuristic to respect min_marks_to_read which is ignored by MergeTreeReadPool in case of remote disk. - /// See comment in IMergeTreeSelectAlgorithm. - if (min_marks_to_read) - { - // check overflow - size_t min_marks_to_read_2 = 0; - bool overflow = common::mulOverflow(min_marks_to_read, 2, min_marks_to_read_2); - - size_t to_mark_overwrite = 0; - if (!overflow) - overflow = common::addOverflow(from_mark, min_marks_to_read_2, to_mark_overwrite); - - if (!overflow && to_mark_overwrite < to_mark) - to_mark = to_mark_overwrite; - } - return getRowsCountInRange(from_mark, std::max(1UL, to_mark)) - offset_in_rows; } diff --git a/src/Storages/MergeTree/MergeTreeIndexGranularity.h b/src/Storages/MergeTree/MergeTreeIndexGranularity.h index 78a1423ad7e..f66e721ec1e 100644 --- a/src/Storages/MergeTree/MergeTreeIndexGranularity.h +++ b/src/Storages/MergeTree/MergeTreeIndexGranularity.h @@ -37,7 +37,7 @@ public: /// |-----|---------------------------|----|----| /// ^------------------------^-----------^ //// from_mark offset_in_rows number_of_rows - size_t countRowsForRows(size_t from_mark, size_t number_of_rows, size_t offset_in_rows, size_t min_marks_to_read) const; + size_t countRowsForRows(size_t from_mark, size_t number_of_rows, size_t offset_in_rows) const; /// Total marks size_t getMarksCount() const; diff --git a/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp b/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp index 2af7abc17f9..9211ab51ad5 100644 --- a/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.cpp @@ -108,6 +108,14 @@ std::optional MergeTreeIndexGranularityInfo::getMarksTypeFromFilesyste return {}; } +MergeTreeIndexGranularityInfo::MergeTreeIndexGranularityInfo( + MarkType mark_type_, size_t index_granularity_, size_t index_granularity_bytes_) + : mark_type(mark_type_) + , fixed_index_granularity(index_granularity_) + , index_granularity_bytes(index_granularity_bytes_) +{ +} + MergeTreeIndexGranularityInfo::MergeTreeIndexGranularityInfo(const MergeTreeData & storage, MergeTreeDataPartType type_) : MergeTreeIndexGranularityInfo(storage, {storage.canUseAdaptiveGranularity(), (*storage.getSettings())[MergeTreeSetting::compress_marks], type_.getValue()}) { diff --git a/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.h b/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.h index 87445c99ade..b302d6b1a4b 100644 --- a/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.h +++ b/src/Storages/MergeTree/MergeTreeIndexGranularityInfo.h @@ -49,6 +49,7 @@ public: MergeTreeIndexGranularityInfo(const MergeTreeData & storage, MarkType mark_type_); MergeTreeIndexGranularityInfo(MergeTreeDataPartType type_, bool is_adaptive_, size_t index_granularity_, size_t index_granularity_bytes_); + MergeTreeIndexGranularityInfo(MarkType mark_type_, size_t index_granularity_, size_t index_granularity_bytes_); void changeGranularityIfRequired(const IDataPartStorage & data_part_storage); diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp index c269a0a23ae..f95b840e223 100644 --- a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.cpp @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -41,10 +42,16 @@ namespace ErrorCodes extern const int INCORRECT_DATA; extern const int INCORRECT_NUMBER_OF_COLUMNS; extern const int INCORRECT_QUERY; + extern const int INVALID_SETTING_VALUE; extern const int LOGICAL_ERROR; extern const int NOT_IMPLEMENTED; } +namespace Setting +{ + extern const SettingsUInt64 hnsw_candidate_list_size_for_search; +} + namespace { @@ -104,7 +111,7 @@ USearchIndexWithSerialization::USearchIndexWithSerialization( { USearchIndex::metric_t metric(dimensions, metric_kind, scalar_kind); - unum::usearch::index_dense_config_t config(usearch_hnsw_params.m, usearch_hnsw_params.ef_construction, usearch_hnsw_params.ef_search); + unum::usearch::index_dense_config_t config(usearch_hnsw_params.connectivity, usearch_hnsw_params.expansion_add, default_expansion_search); config.enable_key_lookups = false; /// we don't do row-to-vector lookups auto result = USearchIndex::make(metric, config); @@ -338,10 +345,11 @@ void MergeTreeIndexAggregatorVectorSimilarity::update(const Block & block, size_ throw Exception(ErrorCodes::INCORRECT_DATA, "Index granularity is too big: more than {} rows per index granule.", std::numeric_limits::max()); if (index_sample_block.columns() > 1) - throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected block with single column"); + throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected that index is build over a single column"); - const String & index_column_name = index_sample_block.getByPosition(0).name; - const ColumnPtr & index_column = block.getByName(index_column_name).column; + const auto & index_column_name = index_sample_block.getByPosition(0).name; + + const auto & index_column = block.getByName(index_column_name).column; ColumnPtr column_cut = index_column->cut(*pos, rows_read); const auto * column_array = typeid_cast(column_cut.get()); @@ -375,8 +383,7 @@ void MergeTreeIndexAggregatorVectorSimilarity::update(const Block & block, size_ if (index->size() + rows > std::numeric_limits::max()) throw Exception(ErrorCodes::INCORRECT_DATA, "Size of vector similarity index would exceed 4 billion entries"); - DataTypePtr data_type = block.getDataTypes()[0]; - const auto * data_type_array = typeid_cast(data_type.get()); + const auto * data_type_array = typeid_cast(block.getByName(index_column_name).type.get()); if (!data_type_array) throw Exception(ErrorCodes::LOGICAL_ERROR, "Expected data type Array(Float*)"); const TypeIndex nested_type_index = data_type_array->getNestedType()->getTypeId(); @@ -399,7 +406,11 @@ MergeTreeIndexConditionVectorSimilarity::MergeTreeIndexConditionVectorSimilarity ContextPtr context) : vector_similarity_condition(query, context) , metric_kind(metric_kind_) + , expansion_search(context->getSettingsRef()[Setting::hnsw_candidate_list_size_for_search]) { + if (expansion_search == 0) + throw Exception(ErrorCodes::INVALID_SETTING_VALUE, "Setting 'hnsw_candidate_list_size_for_search' must not be 0"); + } bool MergeTreeIndexConditionVectorSimilarity::mayBeTrueOnGranule(MergeTreeIndexGranulePtr) const @@ -430,13 +441,17 @@ std::vector MergeTreeIndexConditionVectorSimilarity::calculateApproximat const USearchIndexWithSerializationPtr index = granule->index; if (vector_similarity_condition.getDimensions() != index->dimensions()) - throw Exception(ErrorCodes::INCORRECT_QUERY, "The dimension of the space in the request ({}) " - "does not match the dimension in the index ({})", + throw Exception(ErrorCodes::INCORRECT_QUERY, "The dimension of the space in the request ({}) does not match the dimension in the index ({})", vector_similarity_condition.getDimensions(), index->dimensions()); const std::vector reference_vector = vector_similarity_condition.getReferenceVector(); - auto search_result = index->search(reference_vector.data(), limit); + /// We want to run the search with the user-provided value for setting hnsw_candidate_list_size_for_search (aka. expansion_search). + /// The way to do this in USearch is to call index_dense_gt::change_expansion_search. Unfortunately, this introduces a need to + /// synchronize index access, see https://github.com/unum-cloud/usearch/issues/500. As a workaround, we extended USearch' search method + /// to accept a custom expansion_add setting. The config value is only used on the fly, i.e. not persisted in the index. + + auto search_result = index->search(reference_vector.data(), limit, USearchIndex::any_thread(), false, expansion_search); if (!search_result) throw Exception(ErrorCodes::INCORRECT_DATA, "Could not search in vector similarity index. Error: {}", String(search_result.error.release())); @@ -501,13 +516,12 @@ MergeTreeIndexPtr vectorSimilarityIndexCreator(const IndexDescription & index) UsearchHnswParams usearch_hnsw_params; /// Optional parameters: - const bool has_six_args = (index.arguments.size() == 6); - if (has_six_args) + const bool has_five_args = (index.arguments.size() == 5); + if (has_five_args) { scalar_kind = quantizationToScalarKind.at(index.arguments[2].safeGet()); - usearch_hnsw_params = {.m = index.arguments[3].safeGet(), - .ef_construction = index.arguments[4].safeGet(), - .ef_search = index.arguments[5].safeGet()}; + usearch_hnsw_params = {.connectivity = index.arguments[3].safeGet(), + .expansion_add = index.arguments[4].safeGet()}; } return std::make_shared(index, metric_kind, scalar_kind, usearch_hnsw_params); @@ -516,25 +530,23 @@ MergeTreeIndexPtr vectorSimilarityIndexCreator(const IndexDescription & index) void vectorSimilarityIndexValidator(const IndexDescription & index, bool /* attach */) { const bool has_two_args = (index.arguments.size() == 2); - const bool has_six_args = (index.arguments.size() == 6); + const bool has_five_args = (index.arguments.size() == 5); /// Check number and type of arguments - if (!has_two_args && !has_six_args) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Vector similarity index must have two or six arguments"); + if (!has_two_args && !has_five_args) + throw Exception(ErrorCodes::INCORRECT_QUERY, "Vector similarity index must have two or five arguments"); if (index.arguments[0].getType() != Field::Types::String) throw Exception(ErrorCodes::INCORRECT_QUERY, "First argument of vector similarity index (method) must be of type String"); if (index.arguments[1].getType() != Field::Types::String) throw Exception(ErrorCodes::INCORRECT_QUERY, "Second argument of vector similarity index (metric) must be of type String"); - if (has_six_args) + if (has_five_args) { if (index.arguments[2].getType() != Field::Types::String) throw Exception(ErrorCodes::INCORRECT_QUERY, "Third argument of vector similarity index (quantization) must be of type String"); if (index.arguments[3].getType() != Field::Types::UInt64) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Fourth argument of vector similarity index (M) must be of type UInt64"); + throw Exception(ErrorCodes::INCORRECT_QUERY, "Fourth argument of vector similarity index (hnsw_max_connections_per_layer) must be of type UInt64"); if (index.arguments[4].getType() != Field::Types::UInt64) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Fifth argument of vector similarity index (ef_construction) must be of type UInt64"); - if (index.arguments[5].getType() != Field::Types::UInt64) - throw Exception(ErrorCodes::INCORRECT_QUERY, "Sixth argument of vector similarity index (ef_search) must be of type UInt64"); + throw Exception(ErrorCodes::INCORRECT_QUERY, "Fifth argument of vector similarity index (hnsw_candidate_list_size_for_construction) must be of type UInt64"); } /// Check that passed arguments are supported @@ -542,18 +554,17 @@ void vectorSimilarityIndexValidator(const IndexDescription & index, bool /* atta throw Exception(ErrorCodes::INCORRECT_DATA, "First argument (method) of vector similarity index is not supported. Supported methods are: {}", joinByComma(methods)); if (!distanceFunctionToMetricKind.contains(index.arguments[1].safeGet())) throw Exception(ErrorCodes::INCORRECT_DATA, "Second argument (distance function) of vector similarity index is not supported. Supported distance function are: {}", joinByComma(distanceFunctionToMetricKind)); - if (has_six_args) + if (has_five_args) { if (!quantizationToScalarKind.contains(index.arguments[2].safeGet())) throw Exception(ErrorCodes::INCORRECT_DATA, "Third argument (quantization) of vector similarity index is not supported. Supported quantizations are: {}", joinByComma(quantizationToScalarKind)); /// Call Usearch's own parameter validation method for HNSW-specific parameters - UInt64 m = index.arguments[3].safeGet(); - UInt64 ef_construction = index.arguments[4].safeGet(); - UInt64 ef_search = index.arguments[5].safeGet(); - - unum::usearch::index_dense_config_t config(m, ef_construction, ef_search); + UInt64 connectivity = index.arguments[3].safeGet(); + UInt64 expansion_add = index.arguments[4].safeGet(); + UInt64 expansion_search = default_expansion_search; + unum::usearch::index_dense_config_t config(connectivity, expansion_add, expansion_search); if (auto error = config.validate(); error) throw Exception(ErrorCodes::INCORRECT_DATA, "Invalid parameters passed to vector similarity index. Error: {}", String(error.release())); } diff --git a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h index b77473e7c2b..9a81e168393 100644 --- a/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h +++ b/src/Storages/MergeTree/MergeTreeIndexVectorSimilarity.h @@ -11,11 +11,18 @@ namespace DB { +/// Defaults for HNSW parameters. Instead of using the default parameters provided by USearch (default_connectivity(), +/// default_expansion_add(), default_expansion_search()), we experimentally came up with our own default parameters. They provide better +/// trade-offs with regards to index construction time, search precision and queries-per-second (speed). +static constexpr size_t default_connectivity = 32; +static constexpr size_t default_expansion_add = 128; +static constexpr size_t default_expansion_search = 256; + +/// Parameters for HNSW index construction. struct UsearchHnswParams { - size_t m = unum::usearch::default_connectivity(); - size_t ef_construction = unum::usearch::default_expansion_add(); - size_t ef_search = unum::usearch::default_expansion_search(); + size_t connectivity = default_connectivity; + size_t expansion_add = default_expansion_add; }; using USearchIndex = unum::usearch::index_dense_t; @@ -142,6 +149,7 @@ public: private: const VectorSimilarityCondition vector_similarity_condition; const unum::usearch::metric_kind_t metric_kind; + const size_t expansion_search; }; diff --git a/src/Storages/MergeTree/MergeTreeMarksLoader.cpp b/src/Storages/MergeTree/MergeTreeMarksLoader.cpp index 168134a329f..a271af578cc 100644 --- a/src/Storages/MergeTree/MergeTreeMarksLoader.cpp +++ b/src/Storages/MergeTree/MergeTreeMarksLoader.cpp @@ -3,10 +3,12 @@ #include #include #include +#include #include #include #include #include +#include #include @@ -21,6 +23,11 @@ namespace ProfileEvents namespace DB { +namespace MergeTreeSetting +{ + extern const MergeTreeSettingsString columns_to_prewarm_mark_cache; +} + namespace ErrorCodes { extern const int CANNOT_READ_ALL_DATA; @@ -211,6 +218,7 @@ MarkCache::MappedPtr MergeTreeMarksLoader::loadMarksSync() if (mark_cache) { auto key = MarkCache::hash(fs::path(data_part_storage->getFullPath()) / mrk_path); + if (save_marks_in_cache) { auto callback = [this] { return loadMarksImpl(); }; @@ -249,4 +257,25 @@ std::future MergeTreeMarksLoader::loadMarksAsync() "LoadMarksThread"); } +void addMarksToCache(const IMergeTreeDataPart & part, const PlainMarksByName & cached_marks, MarkCache * mark_cache) +{ + MemoryTrackerBlockerInThread temporarily_disable_memory_tracker; + + for (const auto & [stream_name, marks] : cached_marks) + { + auto mark_path = part.index_granularity_info.getMarksFilePath(stream_name); + auto key = MarkCache::hash(fs::path(part.getDataPartStorage().getFullPath()) / mark_path); + mark_cache->set(key, std::make_shared(*marks)); + } +} + +Names getColumnsToPrewarmMarks(const MergeTreeSettings & settings, const NamesAndTypesList & columns_list) +{ + auto columns_str = settings[MergeTreeSetting::columns_to_prewarm_mark_cache].toString(); + if (columns_str.empty()) + return columns_list.getNames(); + + return parseIdentifiersOrStringLiterals(columns_str, Context::getGlobalContextInstance()->getSettingsRef()); +} + } diff --git a/src/Storages/MergeTree/MergeTreeMarksLoader.h b/src/Storages/MergeTree/MergeTreeMarksLoader.h index 2aa4474e1c5..e031700d6a7 100644 --- a/src/Storages/MergeTree/MergeTreeMarksLoader.h +++ b/src/Storages/MergeTree/MergeTreeMarksLoader.h @@ -1,9 +1,8 @@ #pragma once #include -#include -#include #include +#include namespace DB @@ -11,6 +10,7 @@ namespace DB struct MergeTreeIndexGranularityInfo; using MarksPtr = MarkCache::MappedPtr; +struct ReadSettings; class Threadpool; /// Class that helps to get marks by indexes. @@ -77,4 +77,13 @@ private: using MergeTreeMarksLoaderPtr = std::shared_ptr; +class IMergeTreeDataPart; +struct MergeTreeSettings; + +/// Adds computed marks for part to the marks cache. +void addMarksToCache(const IMergeTreeDataPart & part, const PlainMarksByName & cached_marks, MarkCache * mark_cache); + +/// Returns the list of columns suitable for prewarming of mark cache according to settings. +Names getColumnsToPrewarmMarks(const MergeTreeSettings & settings, const NamesAndTypesList & columns_list); + } diff --git a/src/Storages/MergeTree/MergeTreeMutationStatus.cpp b/src/Storages/MergeTree/MergeTreeMutationStatus.cpp index 6553054774e..e0214d6a79d 100644 --- a/src/Storages/MergeTree/MergeTreeMutationStatus.cpp +++ b/src/Storages/MergeTree/MergeTreeMutationStatus.cpp @@ -26,11 +26,11 @@ void checkMutationStatus(std::optional & status, const throw Exception( ErrorCodes::UNFINISHED, "Exception happened during execution of mutation{} '{}' with part '{}' reason: '{}'. This error maybe retryable or not. " - "In case of unretryable error, mutation can be killed with KILL MUTATION query", + "In case of unretryable error, mutation can be killed with KILL MUTATION query \n\n{}\n", mutation_ids.size() > 1 ? "s" : "", boost::algorithm::join(mutation_ids, ", "), status->latest_failed_part, - status->latest_fail_reason); + status->latest_fail_reason, StackTrace().toString()); } } diff --git a/src/Storages/MergeTree/MergeTreePartInfo.h b/src/Storages/MergeTree/MergeTreePartInfo.h index f128722b03b..28b043fcf20 100644 --- a/src/Storages/MergeTree/MergeTreePartInfo.h +++ b/src/Storages/MergeTree/MergeTreePartInfo.h @@ -46,6 +46,13 @@ struct MergeTreePartInfo < std::forward_as_tuple(rhs.partition_id, rhs.min_block, rhs.max_block, rhs.level, rhs.mutation); } + bool operator>(const MergeTreePartInfo & rhs) const + { + return std::forward_as_tuple(partition_id, min_block, max_block, level, mutation) + > std::forward_as_tuple(rhs.partition_id, rhs.min_block, rhs.max_block, rhs.level, rhs.mutation); + } + + bool operator==(const MergeTreePartInfo & rhs) const { return !(*this != rhs); diff --git a/src/Storages/MergeTree/MergeTreePartsMover.cpp b/src/Storages/MergeTree/MergeTreePartsMover.cpp index 48a4a37f444..e9c9f2b4b06 100644 --- a/src/Storages/MergeTree/MergeTreePartsMover.cpp +++ b/src/Storages/MergeTree/MergeTreePartsMover.cpp @@ -280,7 +280,7 @@ MergeTreePartsMover::TemporaryClonedPart MergeTreePartsMover::clonePart(const Me cloned_part_storage = part->makeCloneOnDisk(disk, MergeTreeData::MOVING_DIR_NAME, read_settings, write_settings, cancellation_hook); } - MergeTreeDataPartBuilder builder(*data, part->name, cloned_part_storage); + MergeTreeDataPartBuilder builder(*data, part->name, cloned_part_storage, getReadSettings()); cloned_part.part = std::move(builder).withPartFormatFromDisk().build(); LOG_TRACE(log, "Part {} was cloned to {}", part->name, cloned_part.part->getDataPartStorage().getFullPath()); diff --git a/src/Storages/MergeTree/MergeTreePrefetchedReadPool.cpp b/src/Storages/MergeTree/MergeTreePrefetchedReadPool.cpp index a99172c4acd..4e5389f2869 100644 --- a/src/Storages/MergeTree/MergeTreePrefetchedReadPool.cpp +++ b/src/Storages/MergeTree/MergeTreePrefetchedReadPool.cpp @@ -1,6 +1,6 @@ +#include #include #include -#include #include #include #include @@ -8,13 +8,13 @@ #include #include #include -#include #include +#include #include #include -#include #include -#include +#include +#include namespace ProfileEvents @@ -102,6 +102,7 @@ MergeTreePrefetchedReadPool::MergeTreePrefetchedReadPool( const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & settings_, + const MergeTreeReadTask::BlockSizeParams & params_, const ContextPtr & context_) : MergeTreeReadPoolBase( std::move(parts_), @@ -113,9 +114,12 @@ MergeTreePrefetchedReadPool::MergeTreePrefetchedReadPool( reader_settings_, column_names_, settings_, + params_, context_) , prefetch_threadpool(getContext()->getPrefetchThreadpool()) - , log(getLogger("MergeTreePrefetchedReadPool(" + (parts_ranges.empty() ? "" : parts_ranges.front().data_part->storage.getStorageID().getNameForLogs()) + ")")) + , log(getLogger( + "MergeTreePrefetchedReadPool(" + + (parts_ranges.empty() ? "" : parts_ranges.front().data_part->storage.getStorageID().getNameForLogs()) + ")")) { /// Tasks creation might also create a lost of readers - check they do not /// do any time consuming operations in ctor. @@ -304,25 +308,11 @@ MergeTreeReadTaskPtr MergeTreePrefetchedReadPool::stealTask(size_t thread, Merge MergeTreeReadTaskPtr MergeTreePrefetchedReadPool::createTask(ThreadTask & task, MergeTreeReadTask * previous_task) { if (task.isValidReadersFuture()) - { - auto size_predictor = task.read_info->shared_size_predictor - ? std::make_unique(*task.read_info->shared_size_predictor) - : nullptr; - - return std::make_unique(task.read_info, task.readers_future->get(), task.ranges, std::move(size_predictor)); - } + return MergeTreeReadPoolBase::createTask(task.read_info, task.readers_future->get(), task.ranges); return MergeTreeReadPoolBase::createTask(task.read_info, task.ranges, previous_task); } -size_t getApproximateSizeOfGranule(const IMergeTreeDataPart & part, const Names & columns_to_read) -{ - ColumnSize columns_size{}; - for (const auto & col_name : columns_to_read) - columns_size.add(part.getColumnSize(col_name)); - return columns_size.data_compressed / part.getMarksCount(); -} - void MergeTreePrefetchedReadPool::fillPerPartStatistics() { per_part_statistics.clear(); @@ -338,11 +328,7 @@ void MergeTreePrefetchedReadPool::fillPerPartStatistics() for (const auto & range : parts_ranges[i].ranges) part_stat.sum_marks += range.end - range.begin; - const auto & columns = settings[Setting::merge_tree_determine_task_size_by_prewhere_columns] && prewhere_info - ? prewhere_info->prewhere_actions.getRequiredColumnsNames() - : column_names; - - part_stat.approx_size_of_mark = getApproximateSizeOfGranule(*read_info.data_part, columns); + part_stat.approx_size_of_mark = read_info.approx_size_of_mark; auto update_stat_for_column = [&](const auto & column_name) { diff --git a/src/Storages/MergeTree/MergeTreePrefetchedReadPool.h b/src/Storages/MergeTree/MergeTreePrefetchedReadPool.h index 1a709250937..b94d4ea113a 100644 --- a/src/Storages/MergeTree/MergeTreePrefetchedReadPool.h +++ b/src/Storages/MergeTree/MergeTreePrefetchedReadPool.h @@ -27,6 +27,7 @@ public: const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & settings_, + const MergeTreeReadTask::BlockSizeParams & params_, const ContextPtr & context_); String getName() const override { return "PrefetchedReadPool"; } diff --git a/src/Storages/MergeTree/MergeTreeRangeReader.h b/src/Storages/MergeTree/MergeTreeRangeReader.h index 7acc8cd88b4..13ce14e02ec 100644 --- a/src/Storages/MergeTree/MergeTreeRangeReader.h +++ b/src/Storages/MergeTree/MergeTreeRangeReader.h @@ -35,7 +35,7 @@ struct PrewhereExprStep bool remove_filter_column = false; bool need_filter = false; - /// Some PREWHERE steps should be executed without conversions. + /// Some PREWHERE steps should be executed without conversions (e.g. early mutation steps) /// A step without alter conversion cannot be executed after step with alter conversions. bool perform_alter_conversions = false; }; diff --git a/src/Storages/MergeTree/MergeTreeReadPool.cpp b/src/Storages/MergeTree/MergeTreeReadPool.cpp index 1e4922757f4..d266ad55824 100644 --- a/src/Storages/MergeTree/MergeTreeReadPool.cpp +++ b/src/Storages/MergeTree/MergeTreeReadPool.cpp @@ -45,6 +45,7 @@ MergeTreeReadPool::MergeTreeReadPool( const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & settings_, + const MergeTreeReadTask::BlockSizeParams & params_, const ContextPtr & context_) : MergeTreeReadPoolBase( std::move(parts_), @@ -56,6 +57,7 @@ MergeTreeReadPool::MergeTreeReadPool( reader_settings_, column_names_, settings_, + params_, context_) , backoff_settings{context_->getSettingsRef()} , backoff_state{pool_settings.threads} diff --git a/src/Storages/MergeTree/MergeTreeReadPool.h b/src/Storages/MergeTree/MergeTreeReadPool.h index c51dca315f9..a0425f0951c 100644 --- a/src/Storages/MergeTree/MergeTreeReadPool.h +++ b/src/Storages/MergeTree/MergeTreeReadPool.h @@ -34,6 +34,7 @@ public: const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & settings_, + const MergeTreeReadTask::BlockSizeParams & params_, const ContextPtr & context_); ~MergeTreeReadPool() override = default; diff --git a/src/Storages/MergeTree/MergeTreeReadPoolBase.cpp b/src/Storages/MergeTree/MergeTreeReadPoolBase.cpp index 6ce1726398a..15a87f463b4 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolBase.cpp +++ b/src/Storages/MergeTree/MergeTreeReadPoolBase.cpp @@ -10,6 +10,7 @@ namespace Setting { extern const SettingsBool merge_tree_determine_task_size_by_prewhere_columns; extern const SettingsUInt64 merge_tree_min_bytes_per_task_for_remote_reading; + extern const SettingsUInt64 merge_tree_min_read_task_size; } namespace ErrorCodes @@ -27,6 +28,7 @@ MergeTreeReadPoolBase::MergeTreeReadPoolBase( const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & pool_settings_, + const MergeTreeReadTask::BlockSizeParams & block_size_params_, const ContextPtr & context_) : WithContext(context_) , parts_ranges(std::move(parts_)) @@ -38,6 +40,7 @@ MergeTreeReadPoolBase::MergeTreeReadPoolBase( , reader_settings(reader_settings_) , column_names(column_names_) , pool_settings(pool_settings_) + , block_size_params(block_size_params_) , owned_mark_cache(context_->getGlobalContext()->getMarkCache()) , owned_uncompressed_cache(pool_settings_.use_uncompressed_cache ? context_->getGlobalContext()->getUncompressedCache() : nullptr) , header(storage_snapshot->getSampleBlockForColumns(column_names)) @@ -46,7 +49,7 @@ MergeTreeReadPoolBase::MergeTreeReadPoolBase( fillPerPartInfos(context_->getSettingsRef()); } -static size_t getApproxSizeOfPart(const IMergeTreeDataPart & part, const Names & columns_to_read) +static size_t getSizeOfColumns(const IMergeTreeDataPart & part, const Names & columns_to_read) { ColumnSize columns_size{}; for (const auto & col_name : columns_to_read) @@ -55,44 +58,67 @@ static size_t getApproxSizeOfPart(const IMergeTreeDataPart & part, const Names & return columns_size.data_compressed ? columns_size.data_compressed : part.getBytesOnDisk(); } -static size_t calculateMinMarksPerTask( +/// Columns from different prewhere steps are read independently, so it makes sense to use the heaviest set of columns among them as an estimation. +static Names +getHeaviestSetOfColumnsAmongPrewhereSteps(const IMergeTreeDataPart & part, const std::vector & prewhere_steps_columns) +{ + const auto it = std::ranges::max_element( + prewhere_steps_columns, + [&](const auto & lhs, const auto & rhs) + { return getSizeOfColumns(part, lhs.getNames()) < getSizeOfColumns(part, rhs.getNames()); }); + return it->getNames(); +} + +static std::pair // (min_marks_per_task, avg_mark_bytes) +calculateMinMarksPerTask( const RangesInDataPart & part, const Names & columns_to_read, - PrewhereInfoPtr prewhere_info, + const std::vector & prewhere_steps_columns, const MergeTreeReadPoolBase::PoolSettings & pool_settings, const Settings & settings) { - size_t min_marks_per_task = pool_settings.min_marks_for_concurrent_read; - const size_t part_marks_count = part.getMarksCount(); - if (part_marks_count && part.data_part->isStoredOnRemoteDisk()) + size_t min_marks_per_task + = std::max(settings[Setting::merge_tree_min_read_task_size], pool_settings.min_marks_for_concurrent_read); + size_t avg_mark_bytes = 0; + /// It is important to obtain marks count from the part itself instead of calling `part.getMarksCount()`, + /// because `part` will report number of marks selected from this part by the query. + const size_t part_marks_count = part.data_part->getMarksCount(); + if (part_marks_count) { - /// We assume that most of the time prewhere does it's job good meaning that lion's share of the rows is filtered out. - /// Which means in turn that for most of the rows we will read only the columns from prewhere clause. - /// So it makes sense to use only them for the estimation. - const auto & columns = settings[Setting::merge_tree_determine_task_size_by_prewhere_columns] && prewhere_info - ? prewhere_info->prewhere_actions.getRequiredColumnsNames() - : columns_to_read; - const size_t part_compressed_bytes = getApproxSizeOfPart(*part.data_part, columns); - - const auto avg_mark_bytes = std::max(part_compressed_bytes / part_marks_count, 1); - const auto min_bytes_per_task = settings[Setting::merge_tree_min_bytes_per_task_for_remote_reading]; - /// We're taking min here because number of tasks shouldn't be too low - it will make task stealing impossible. - /// We also create at least two tasks per thread to have something to steal from a slow thread. - const auto heuristic_min_marks - = std::min(pool_settings.sum_marks / pool_settings.threads / 2, min_bytes_per_task / avg_mark_bytes); - if (heuristic_min_marks > min_marks_per_task) + if (part.data_part->isStoredOnRemoteDisk()) { - LOG_TEST( - &Poco::Logger::get("MergeTreeReadPoolBase"), - "Increasing min_marks_per_task from {} to {} based on columns size heuristic", - min_marks_per_task, - heuristic_min_marks); - min_marks_per_task = heuristic_min_marks; + /// We assume that most of the time prewhere does it's job good meaning that lion's share of the rows is filtered out. + /// Which means in turn that for most of the rows we will read only the columns from prewhere clause. + /// So it makes sense to use only them for the estimation. + const auto & columns = settings[Setting::merge_tree_determine_task_size_by_prewhere_columns] && !prewhere_steps_columns.empty() + ? getHeaviestSetOfColumnsAmongPrewhereSteps(*part.data_part, prewhere_steps_columns) + : columns_to_read; + const size_t part_compressed_bytes = getSizeOfColumns(*part.data_part, columns); + + avg_mark_bytes = std::max(part_compressed_bytes / part_marks_count, 1); + const auto min_bytes_per_task = settings[Setting::merge_tree_min_bytes_per_task_for_remote_reading]; + /// We're taking min here because number of tasks shouldn't be too low - it will make task stealing impossible. + /// We also create at least two tasks per thread to have something to steal from a slow thread. + const auto heuristic_min_marks + = std::min(pool_settings.sum_marks / pool_settings.threads / 2, min_bytes_per_task / avg_mark_bytes); + if (heuristic_min_marks > min_marks_per_task) + { + LOG_TEST( + &Poco::Logger::get("MergeTreeReadPoolBase"), + "Increasing min_marks_per_task from {} to {} based on columns size heuristic", + min_marks_per_task, + heuristic_min_marks); + min_marks_per_task = heuristic_min_marks; + } + } + else + { + avg_mark_bytes = std::max(getSizeOfColumns(*part.data_part, columns_to_read) / part_marks_count, 1); } } LOG_TEST(&Poco::Logger::get("MergeTreeReadPoolBase"), "Will use min_marks_per_task={}", min_marks_per_task); - return min_marks_per_task; + return {min_marks_per_task, avg_mark_bytes}; } void MergeTreeReadPoolBase::fillPerPartInfos(const Settings & settings) @@ -159,8 +185,8 @@ void MergeTreeReadPoolBase::fillPerPartInfos(const Settings & settings) } is_part_on_remote_disk.push_back(part_with_ranges.data_part->isStoredOnRemoteDisk()); - read_task_info.min_marks_per_task - = calculateMinMarksPerTask(part_with_ranges, column_names, prewhere_info, pool_settings, settings); + std::tie(read_task_info.min_marks_per_task, read_task_info.approx_size_of_mark) + = calculateMinMarksPerTask(part_with_ranges, column_names, read_task_info.task_columns.pre_columns, pool_settings, settings); per_part_infos.push_back(std::make_shared(std::move(read_task_info))); } } @@ -182,15 +208,20 @@ std::vector MergeTreeReadPoolBase::getPerPartSumMarks() const return per_part_sum_marks; } -MergeTreeReadTaskPtr MergeTreeReadPoolBase::createTask( - MergeTreeReadTaskInfoPtr read_info, - MarkRanges ranges, - MergeTreeReadTask * previous_task) const +MergeTreeReadTaskPtr +MergeTreeReadPoolBase::createTask(MergeTreeReadTaskInfoPtr read_info, MergeTreeReadTask::Readers task_readers, MarkRanges ranges) const { auto task_size_predictor = read_info->shared_size_predictor ? std::make_unique(*read_info->shared_size_predictor) : nullptr; /// make a copy + return std::make_unique( + read_info, std::move(task_readers), std::move(ranges), block_size_params, std::move(task_size_predictor)); +} + +MergeTreeReadTaskPtr +MergeTreeReadPoolBase::createTask(MergeTreeReadTaskInfoPtr read_info, MarkRanges ranges, MergeTreeReadTask * previous_task) const +{ auto get_part_name = [](const auto & task_info) -> String { const auto & data_part = task_info.data_part; @@ -229,11 +260,7 @@ MergeTreeReadTaskPtr MergeTreeReadPoolBase::createTask( task_readers = previous_task->releaseReaders(); } - return std::make_unique( - read_info, - std::move(task_readers), - std::move(ranges), - std::move(task_size_predictor)); + return createTask(read_info, std::move(task_readers), std::move(ranges)); } MergeTreeReadTask::Extras MergeTreeReadPoolBase::getExtras() const diff --git a/src/Storages/MergeTree/MergeTreeReadPoolBase.h b/src/Storages/MergeTree/MergeTreeReadPoolBase.h index 7f9106d476e..19b26156433 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolBase.h +++ b/src/Storages/MergeTree/MergeTreeReadPoolBase.h @@ -33,6 +33,7 @@ public: const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & settings_, + const MergeTreeReadTask::BlockSizeParams & params_, const ContextPtr & context_); Block getHeader() const override { return header; } @@ -48,6 +49,7 @@ protected: const MergeTreeReaderSettings reader_settings; const Names column_names; const PoolSettings pool_settings; + const MergeTreeReadTask::BlockSizeParams block_size_params; const MarkCachePtr owned_mark_cache; const UncompressedCachePtr owned_uncompressed_cache; const Block header; @@ -55,6 +57,8 @@ protected: void fillPerPartInfos(const Settings & settings); std::vector getPerPartSumMarks() const; + MergeTreeReadTaskPtr createTask(MergeTreeReadTaskInfoPtr read_info, MergeTreeReadTask::Readers task_readers, MarkRanges ranges) const; + MergeTreeReadTaskPtr createTask( MergeTreeReadTaskInfoPtr read_info, MarkRanges ranges, diff --git a/src/Storages/MergeTree/MergeTreeReadPoolInOrder.cpp b/src/Storages/MergeTree/MergeTreeReadPoolInOrder.cpp index 60f127acdae..c4244ecd982 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolInOrder.cpp +++ b/src/Storages/MergeTree/MergeTreeReadPoolInOrder.cpp @@ -20,6 +20,7 @@ MergeTreeReadPoolInOrder::MergeTreeReadPoolInOrder( const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & settings_, + const MergeTreeReadTask::BlockSizeParams & params_, const ContextPtr & context_) : MergeTreeReadPoolBase( std::move(parts_), @@ -31,6 +32,7 @@ MergeTreeReadPoolInOrder::MergeTreeReadPoolInOrder( reader_settings_, column_names_, settings_, + params_, context_) , has_limit_below_one_block(has_limit_below_one_block_) , read_type(read_type_) diff --git a/src/Storages/MergeTree/MergeTreeReadPoolInOrder.h b/src/Storages/MergeTree/MergeTreeReadPoolInOrder.h index a3668acb170..41f3ab1061c 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolInOrder.h +++ b/src/Storages/MergeTree/MergeTreeReadPoolInOrder.h @@ -19,6 +19,7 @@ public: const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & settings_, + const MergeTreeReadTask::BlockSizeParams & params_, const ContextPtr & context_); String getName() const override { return "ReadPoolInOrder"; } diff --git a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.cpp b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.cpp index 075c0b1042b..8f06fc312c2 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.cpp +++ b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.cpp @@ -112,6 +112,7 @@ MergeTreeReadPoolParallelReplicas::MergeTreeReadPoolParallelReplicas( const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & settings_, + const MergeTreeReadTask::BlockSizeParams & params_, const ContextPtr & context_) : MergeTreeReadPoolBase( std::move(parts_), @@ -123,6 +124,7 @@ MergeTreeReadPoolParallelReplicas::MergeTreeReadPoolParallelReplicas( reader_settings_, column_names_, settings_, + params_, context_) , extension(std::move(extension_)) , coordination_mode(CoordinationMode::Default) diff --git a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.h b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.h index b9f2e133c4a..63816340eb1 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.h +++ b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicas.h @@ -19,6 +19,7 @@ public: const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & settings_, + const MergeTreeReadTask::BlockSizeParams & params_, const ContextPtr & context_); ~MergeTreeReadPoolParallelReplicas() override = default; diff --git a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.cpp b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.cpp index 8ff2a4f31ee..f13da426c45 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.cpp +++ b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.cpp @@ -26,6 +26,7 @@ MergeTreeReadPoolParallelReplicasInOrder::MergeTreeReadPoolParallelReplicasInOrd const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & settings_, + const MergeTreeReadTask::BlockSizeParams & params_, const ContextPtr & context_) : MergeTreeReadPoolBase( std::move(parts_), @@ -37,6 +38,7 @@ MergeTreeReadPoolParallelReplicasInOrder::MergeTreeReadPoolParallelReplicasInOrd reader_settings_, column_names_, settings_, + params_, context_) , extension(std::move(extension_)) , mode(mode_) diff --git a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.h b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.h index 98a4d95768a..a05dc54b529 100644 --- a/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.h +++ b/src/Storages/MergeTree/MergeTreeReadPoolParallelReplicasInOrder.h @@ -20,6 +20,7 @@ public: const MergeTreeReaderSettings & reader_settings_, const Names & column_names_, const PoolSettings & settings_, + const MergeTreeReadTask::BlockSizeParams & params_, const ContextPtr & context_); String getName() const override { return "ReadPoolParallelReplicasInOrder"; } diff --git a/src/Storages/MergeTree/MergeTreeReadTask.cpp b/src/Storages/MergeTree/MergeTreeReadTask.cpp index dd057dc9984..72fddb93a6d 100644 --- a/src/Storages/MergeTree/MergeTreeReadTask.cpp +++ b/src/Storages/MergeTree/MergeTreeReadTask.cpp @@ -26,10 +26,12 @@ MergeTreeReadTask::MergeTreeReadTask( MergeTreeReadTaskInfoPtr info_, Readers readers_, MarkRanges mark_ranges_, + const BlockSizeParams & block_size_params_, MergeTreeBlockSizePredictorPtr size_predictor_) : info(std::move(info_)) , readers(std::move(readers_)) , mark_ranges(std::move(mark_ranges_)) + , block_size_params(block_size_params_) , size_predictor(std::move(size_predictor_)) { } @@ -112,30 +114,31 @@ void MergeTreeReadTask::initializeRangeReaders(const PrewhereExprInfo & prewhere range_readers = createRangeReaders(readers, prewhere_actions); } -UInt64 MergeTreeReadTask::estimateNumRows(const BlockSizeParams & params) const +UInt64 MergeTreeReadTask::estimateNumRows() const { if (!size_predictor) { - if (params.preferred_block_size_bytes) + if (block_size_params.preferred_block_size_bytes) throw Exception(ErrorCodes::LOGICAL_ERROR, "Size predictor is not set, it might lead to a performance degradation"); - return static_cast(params.max_block_size_rows); + return static_cast(block_size_params.max_block_size_rows); } /// Calculates number of rows will be read using preferred_block_size_bytes. /// Can't be less than avg_index_granularity. - size_t rows_to_read = size_predictor->estimateNumRows(params.preferred_block_size_bytes); + size_t rows_to_read = size_predictor->estimateNumRows(block_size_params.preferred_block_size_bytes); if (!rows_to_read) return rows_to_read; auto total_row_in_current_granule = range_readers.main.numRowsInCurrentGranule(); rows_to_read = std::max(total_row_in_current_granule, rows_to_read); - if (params.preferred_max_column_in_block_size_bytes) + if (block_size_params.preferred_max_column_in_block_size_bytes) { /// Calculates number of rows will be read using preferred_max_column_in_block_size_bytes. - auto rows_to_read_for_max_size_column = size_predictor->estimateNumRowsForMaxSizeColumn(params.preferred_max_column_in_block_size_bytes); + auto rows_to_read_for_max_size_column + = size_predictor->estimateNumRowsForMaxSizeColumn(block_size_params.preferred_max_column_in_block_size_bytes); - double filtration_ratio = std::max(params.min_filtration_ratio, 1.0 - size_predictor->filtered_rows_ratio); + double filtration_ratio = std::max(block_size_params.min_filtration_ratio, 1.0 - size_predictor->filtered_rows_ratio); auto rows_to_read_for_max_size_column_with_filtration = static_cast(rows_to_read_for_max_size_column / filtration_ratio); @@ -148,16 +151,16 @@ UInt64 MergeTreeReadTask::estimateNumRows(const BlockSizeParams & params) const return rows_to_read; const auto & index_granularity = info->data_part->index_granularity; - return index_granularity.countRowsForRows(range_readers.main.currentMark(), rows_to_read, range_readers.main.numReadRowsInCurrentGranule(), params.min_marks_to_read); + return index_granularity.countRowsForRows(range_readers.main.currentMark(), rows_to_read, range_readers.main.numReadRowsInCurrentGranule()); } -MergeTreeReadTask::BlockAndProgress MergeTreeReadTask::read(const BlockSizeParams & params) +MergeTreeReadTask::BlockAndProgress MergeTreeReadTask::read() { if (size_predictor) size_predictor->startBlock(); - UInt64 recommended_rows = estimateNumRows(params); - UInt64 rows_to_read = std::max(static_cast(1), std::min(params.max_block_size_rows, recommended_rows)); + UInt64 recommended_rows = estimateNumRows(); + UInt64 rows_to_read = std::max(static_cast(1), std::min(block_size_params.max_block_size_rows, recommended_rows)); auto read_result = range_readers.main.read(rows_to_read, mark_ranges); diff --git a/src/Storages/MergeTree/MergeTreeReadTask.h b/src/Storages/MergeTree/MergeTreeReadTask.h index 748babb5b4c..2853cc39c51 100644 --- a/src/Storages/MergeTree/MergeTreeReadTask.h +++ b/src/Storages/MergeTree/MergeTreeReadTask.h @@ -70,6 +70,7 @@ struct MergeTreeReadTaskInfo VirtualFields const_virtual_fields; /// The amount of data to read per task based on size of the queried columns. size_t min_marks_per_task = 0; + size_t approx_size_of_mark = 0; }; using MergeTreeReadTaskInfoPtr = std::shared_ptr; @@ -110,7 +111,6 @@ public: UInt64 max_block_size_rows = DEFAULT_BLOCK_SIZE; UInt64 preferred_block_size_bytes = 1000000; UInt64 preferred_max_column_in_block_size_bytes = 0; - UInt64 min_marks_to_read = 0; double min_filtration_ratio = 0.00001; }; @@ -127,12 +127,12 @@ public: MergeTreeReadTaskInfoPtr info_, Readers readers_, MarkRanges mark_ranges_, - + const BlockSizeParams & block_size_params_, MergeTreeBlockSizePredictorPtr size_predictor_); void initializeRangeReaders(const PrewhereExprInfo & prewhere_actions); - BlockAndProgress read(const BlockSizeParams & params); + BlockAndProgress read(); bool isFinished() const { return mark_ranges.empty() && range_readers.main.isCurrentRangeFinished(); } const MergeTreeReadTaskInfo & getInfo() const { return *info; } @@ -145,7 +145,7 @@ public: static RangeReaders createRangeReaders(const Readers & readers, const PrewhereExprInfo & prewhere_actions); private: - UInt64 estimateNumRows(const BlockSizeParams & params) const; + UInt64 estimateNumRows() const; /// Shared information required for reading. MergeTreeReadTaskInfoPtr info; @@ -160,6 +160,8 @@ private: /// Ranges to read from data_part MarkRanges mark_ranges; + BlockSizeParams block_size_params; + /// Used to satistfy preferred_block_size_bytes limitation MergeTreeBlockSizePredictorPtr size_predictor; }; diff --git a/src/Storages/MergeTree/MergeTreeReaderWide.cpp b/src/Storages/MergeTree/MergeTreeReaderWide.cpp index 898bf5a2933..77231d8d392 100644 --- a/src/Storages/MergeTree/MergeTreeReaderWide.cpp +++ b/src/Storages/MergeTree/MergeTreeReaderWide.cpp @@ -262,7 +262,7 @@ MergeTreeReaderWide::FileStreams::iterator MergeTreeReaderWide::addStream(const /*num_columns_in_mark=*/ 1); auto stream_settings = settings; - stream_settings.is_low_cardinality_dictionary = substream_path.size() > 1 && substream_path[substream_path.size() - 2].type == ISerialization::Substream::Type::DictionaryKeys; + stream_settings.is_low_cardinality_dictionary = ISerialization::isLowCardinalityDictionarySubcolumn(substream_path); auto create_stream = [&]() { diff --git a/src/Storages/MergeTree/MergeTreeSelectAlgorithms.cpp b/src/Storages/MergeTree/MergeTreeSelectAlgorithms.cpp index bf97d269dc6..213eab52ad8 100644 --- a/src/Storages/MergeTree/MergeTreeSelectAlgorithms.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectAlgorithms.cpp @@ -30,7 +30,8 @@ MergeTreeReadTaskPtr MergeTreeInReverseOrderSelectAlgorithm::getNewTask(IMergeTr return pool.getTask(part_idx, previous_task); } -MergeTreeReadTask::BlockAndProgress MergeTreeInReverseOrderSelectAlgorithm::readFromTask(MergeTreeReadTask & task, const BlockSizeParams & params) +MergeTreeReadTask::BlockAndProgress +MergeTreeInReverseOrderSelectAlgorithm::readFromTask(MergeTreeReadTask & task) { MergeTreeReadTask::BlockAndProgress res; @@ -42,7 +43,7 @@ MergeTreeReadTask::BlockAndProgress MergeTreeInReverseOrderSelectAlgorithm::read } while (!task.isFinished()) - chunks.push_back(task.read(params)); + chunks.push_back(task.read()); if (chunks.empty()) return {}; diff --git a/src/Storages/MergeTree/MergeTreeSelectAlgorithms.h b/src/Storages/MergeTree/MergeTreeSelectAlgorithms.h index afc8032bb99..eeaefb0dc4f 100644 --- a/src/Storages/MergeTree/MergeTreeSelectAlgorithms.h +++ b/src/Storages/MergeTree/MergeTreeSelectAlgorithms.h @@ -21,7 +21,7 @@ public: virtual bool needNewTask(const MergeTreeReadTask & task) const = 0; virtual MergeTreeReadTaskPtr getNewTask(IMergeTreeReadPool & pool, MergeTreeReadTask * previous_task) = 0; - virtual BlockAndProgress readFromTask(MergeTreeReadTask & task, const BlockSizeParams & params) = 0; + virtual BlockAndProgress readFromTask(MergeTreeReadTask & task) = 0; }; using MergeTreeSelectAlgorithmPtr = std::unique_ptr; @@ -35,7 +35,7 @@ public: bool needNewTask(const MergeTreeReadTask & task) const override { return task.isFinished(); } MergeTreeReadTaskPtr getNewTask(IMergeTreeReadPool & pool, MergeTreeReadTask * previous_task) override { return pool.getTask(thread_idx, previous_task); } - BlockAndProgress readFromTask(MergeTreeReadTask & task, const BlockSizeParams & params) override { return task.read(params); } + BlockAndProgress readFromTask(MergeTreeReadTask & task) override { return task.read(); } private: const size_t thread_idx; @@ -50,7 +50,7 @@ public: bool needNewTask(const MergeTreeReadTask & task) const override { return task.isFinished(); } MergeTreeReadTaskPtr getNewTask(IMergeTreeReadPool & pool, MergeTreeReadTask * previous_task) override; - MergeTreeReadTask::BlockAndProgress readFromTask(MergeTreeReadTask & task, const BlockSizeParams & params) override { return task.read(params); } + MergeTreeReadTask::BlockAndProgress readFromTask(MergeTreeReadTask & task) override { return task.read(); } private: const size_t part_idx; @@ -65,7 +65,7 @@ public: bool needNewTask(const MergeTreeReadTask & task) const override { return chunks.empty() && task.isFinished(); } MergeTreeReadTaskPtr getNewTask(IMergeTreeReadPool & pool, MergeTreeReadTask * previous_task) override; - BlockAndProgress readFromTask(MergeTreeReadTask & task, const BlockSizeParams & params) override; + BlockAndProgress readFromTask(MergeTreeReadTask & task) override; private: const size_t part_idx; diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index 76bcf41d6d8..5efd33ce09a 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -86,7 +86,6 @@ MergeTreeSelectProcessor::MergeTreeSelectProcessor( MergeTreeSelectAlgorithmPtr algorithm_, const PrewhereInfoPtr & prewhere_info_, const ExpressionActionsSettings & actions_settings_, - const MergeTreeReadTask::BlockSizeParams & block_size_params_, const MergeTreeReaderSettings & reader_settings_) : pool(std::move(pool_)) , algorithm(std::move(algorithm_)) @@ -94,7 +93,6 @@ MergeTreeSelectProcessor::MergeTreeSelectProcessor( , actions_settings(actions_settings_) , prewhere_actions(getPrewhereActions(prewhere_info, actions_settings, reader_settings_.enable_multiple_prewhere_read_steps)) , reader_settings(reader_settings_) - , block_size_params(block_size_params_) , result_header(transformHeader(pool->getHeader(), prewhere_info)) { if (reader_settings.apply_deleted_mask) @@ -190,7 +188,7 @@ ChunkAndProgress MergeTreeSelectProcessor::read() if (!task->getMainRangeReader().isInitialized()) initializeRangeReaders(); - auto res = algorithm->readFromTask(*task, block_size_params); + auto res = algorithm->readFromTask(*task); if (res.row_count) { diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index 8a9e3580a9f..33069a78e33 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -57,7 +57,6 @@ public: MergeTreeSelectAlgorithmPtr algorithm_, const PrewhereInfoPtr & prewhere_info_, const ExpressionActionsSettings & actions_settings_, - const MergeTreeReadTask::BlockSizeParams & block_size_params_, const MergeTreeReaderSettings & reader_settings_); String getName() const; diff --git a/src/Storages/MergeTree/MergeTreeSettings.cpp b/src/Storages/MergeTree/MergeTreeSettings.cpp index 4e7d0c0a721..33910d1048d 100644 --- a/src/Storages/MergeTree/MergeTreeSettings.cpp +++ b/src/Storages/MergeTree/MergeTreeSettings.cpp @@ -1,4 +1,5 @@ #include +#include #include #include #include @@ -29,218 +30,220 @@ namespace ErrorCodes extern const int BAD_ARGUMENTS; } +// clang-format off + /** These settings represent fine tunes for internal details of MergeTree storages * and should not be changed by the user without a reason. */ - -#define MERGE_TREE_SETTINGS(M, ALIAS) \ - M(UInt64, min_compress_block_size, 0, "When granule is written, compress the data in buffer if the size of pending uncompressed data is larger or equal than the specified threshold. If this setting is not set, the corresponding global setting is used.", 0) \ - M(UInt64, max_compress_block_size, 0, "Compress the pending uncompressed data in buffer if its size is larger or equal than the specified threshold. Block of data will be compressed even if the current granule is not finished. If this setting is not set, the corresponding global setting is used.", 0) \ - M(UInt64, index_granularity, 8192, "How many rows correspond to one primary key value.", 0) \ - M(UInt64, max_digestion_size_per_segment, 256_MiB, "Max number of bytes to digest per segment to build GIN index.", 0) \ +#define MERGE_TREE_SETTINGS(DECLARE, ALIAS) \ + DECLARE(UInt64, min_compress_block_size, 0, "When granule is written, compress the data in buffer if the size of pending uncompressed data is larger or equal than the specified threshold. If this setting is not set, the corresponding global setting is used.", 0) \ + DECLARE(UInt64, max_compress_block_size, 0, "Compress the pending uncompressed data in buffer if its size is larger or equal than the specified threshold. Block of data will be compressed even if the current granule is not finished. If this setting is not set, the corresponding global setting is used.", 0) \ + DECLARE(UInt64, index_granularity, 8192, "How many rows correspond to one primary key value.", 0) \ + DECLARE(UInt64, max_digestion_size_per_segment, 256_MiB, "Max number of bytes to digest per segment to build GIN index.", 0) \ \ /** Data storing format settings. */ \ - M(UInt64, min_bytes_for_wide_part, 10485760, "Minimal uncompressed size in bytes to create part in wide format instead of compact", 0) \ - M(UInt64, min_rows_for_wide_part, 0, "Minimal number of rows to create part in wide format instead of compact", 0) \ - M(Float, ratio_of_defaults_for_sparse_serialization, 0.9375f, "Minimal ratio of number of default values to number of all values in column to store it in sparse serializations. If >= 1, columns will be always written in full serialization.", 0) \ - M(Bool, replace_long_file_name_to_hash, true, "If the file name for column is too long (more than 'max_file_name_length' bytes) replace it to SipHash128", 0) \ - M(UInt64, max_file_name_length, 127, "The maximal length of the file name to keep it as is without hashing", 0) \ - M(UInt64, min_bytes_for_full_part_storage, 0, "Only available in ClickHouse Cloud", 0) \ - M(UInt64, min_rows_for_full_part_storage, 0, "Only available in ClickHouse Cloud", 0) \ - M(UInt64, compact_parts_max_bytes_to_buffer, 128 * 1024 * 1024, "Only available in ClickHouse Cloud", 0) \ - M(UInt64, compact_parts_max_granules_to_buffer, 128, "Only available in ClickHouse Cloud", 0) \ - M(UInt64, compact_parts_merge_max_bytes_to_prefetch_part, 16 * 1024 * 1024, "Only available in ClickHouse Cloud", 0) \ - M(Bool, load_existing_rows_count_for_old_parts, false, "Whether to load existing_rows_count for existing parts. If false, existing_rows_count will be equal to rows_count for existing parts.", 0) \ - M(Bool, use_compact_variant_discriminators_serialization, true, "Use compact version of Variant discriminators serialization.", 0) \ - \ - /** Merge and insert settings */ \ - M(UInt64, max_compression_threads, 1, "Maximum number of threads for writing compressed data. This is an expert-level setting, do not change it.", 0) \ + DECLARE(UInt64, min_bytes_for_wide_part, 10485760, "Minimal uncompressed size in bytes to create part in wide format instead of compact", 0) \ + DECLARE(UInt64, min_rows_for_wide_part, 0, "Minimal number of rows to create part in wide format instead of compact", 0) \ + DECLARE(Float, ratio_of_defaults_for_sparse_serialization, 0.9375f, "Minimal ratio of number of default values to number of all values in column to store it in sparse serializations. If >= 1, columns will be always written in full serialization.", 0) \ + DECLARE(Bool, replace_long_file_name_to_hash, true, "If the file name for column is too long (more than 'max_file_name_length' bytes) replace it to SipHash128", 0) \ + DECLARE(UInt64, max_file_name_length, 127, "The maximal length of the file name to keep it as is without hashing", 0) \ + DECLARE(UInt64, min_bytes_for_full_part_storage, 0, "Only available in ClickHouse Cloud", 0) \ + DECLARE(UInt64, min_rows_for_full_part_storage, 0, "Only available in ClickHouse Cloud", 0) \ + DECLARE(UInt64, compact_parts_max_bytes_to_buffer, 128 * 1024 * 1024, "Only available in ClickHouse Cloud", 0) \ + DECLARE(UInt64, compact_parts_max_granules_to_buffer, 128, "Only available in ClickHouse Cloud", 0) \ + DECLARE(UInt64, compact_parts_merge_max_bytes_to_prefetch_part, 16 * 1024 * 1024, "Only available in ClickHouse Cloud", 0) \ + DECLARE(Bool, load_existing_rows_count_for_old_parts, false, "Whether to load existing_rows_count for existing parts. If false, existing_rows_count will be equal to rows_count for existing parts.", 0) \ + DECLARE(Bool, use_compact_variant_discriminators_serialization, true, "Use compact version of Variant discriminators serialization.", 0) \ \ /** Merge selector settings. */ \ - M(UInt64, merge_selector_blurry_base_scale_factor, 0, "Controls when the logic kicks in relatively to the number of parts in partition. The bigger the factor the more belated reaction will be.", 0) \ - M(UInt64, merge_selector_window_size, 1000, "How many parts to look at once.", 0) \ + DECLARE(UInt64, merge_selector_blurry_base_scale_factor, 0, "Controls when the logic kicks in relatively to the number of parts in partition. The bigger the factor the more belated reaction will be.", 0) \ + DECLARE(UInt64, merge_selector_window_size, 1000, "How many parts to look at once.", 0) \ \ /** Merge settings. */ \ - M(UInt64, merge_max_block_size, 8192, "How many rows in blocks should be formed for merge operations. By default has the same value as `index_granularity`.", 0) \ - M(UInt64, merge_max_block_size_bytes, 10 * 1024 * 1024, "How many bytes in blocks should be formed for merge operations. By default has the same value as `index_granularity_bytes`.", 0) \ - M(UInt64, max_bytes_to_merge_at_max_space_in_pool, 150ULL * 1024 * 1024 * 1024, "Maximum in total size of parts to merge, when there are maximum free threads in background pool (or entries in replication queue).", 0) \ - M(UInt64, max_bytes_to_merge_at_min_space_in_pool, 1024 * 1024, "Maximum in total size of parts to merge, when there are minimum free threads in background pool (or entries in replication queue).", 0) \ - M(UInt64, max_replicated_merges_in_queue, 1000, "How many tasks of merging and mutating parts are allowed simultaneously in ReplicatedMergeTree queue.", 0) \ - M(UInt64, max_replicated_mutations_in_queue, 8, "How many tasks of mutating parts are allowed simultaneously in ReplicatedMergeTree queue.", 0) \ - M(UInt64, max_replicated_merges_with_ttl_in_queue, 1, "How many tasks of merging parts with TTL are allowed simultaneously in ReplicatedMergeTree queue.", 0) \ - M(UInt64, number_of_free_entries_in_pool_to_lower_max_size_of_merge, 8, "When there is less than specified number of free entries in pool (or replicated queue), start to lower maximum size of merge to process (or to put in queue). This is to allow small merges to process - not filling the pool with long running merges.", 0) \ - M(UInt64, number_of_free_entries_in_pool_to_execute_mutation, 20, "When there is less than specified number of free entries in pool, do not execute part mutations. This is to leave free threads for regular merges and avoid \"Too many parts\"", 0) \ - M(UInt64, max_number_of_mutations_for_replica, 0, "Limit the number of part mutations per replica to the specified amount. Zero means no limit on the number of mutations per replica (the execution can still be constrained by other settings).", 0) \ - M(UInt64, max_number_of_merges_with_ttl_in_pool, 2, "When there is more than specified number of merges with TTL entries in pool, do not assign new merge with TTL. This is to leave free threads for regular merges and avoid \"Too many parts\"", 0) \ - M(Seconds, old_parts_lifetime, 8 * 60, "How many seconds to keep obsolete parts.", 0) \ - M(Seconds, temporary_directories_lifetime, 86400, "How many seconds to keep tmp_-directories. You should not lower this value because merges and mutations may not be able to work with low value of this setting.", 0) \ - M(Seconds, lock_acquire_timeout_for_background_operations, DBMS_DEFAULT_LOCK_ACQUIRE_TIMEOUT_SEC, "For background operations like merges, mutations etc. How many seconds before failing to acquire table locks.", 0) \ - M(UInt64, min_rows_to_fsync_after_merge, 0, "Minimal number of rows to do fsync for part after merge (0 - disabled)", 0) \ - M(UInt64, min_compressed_bytes_to_fsync_after_merge, 0, "Minimal number of compressed bytes to do fsync for part after merge (0 - disabled)", 0) \ - M(UInt64, min_compressed_bytes_to_fsync_after_fetch, 0, "Minimal number of compressed bytes to do fsync for part after fetch (0 - disabled)", 0) \ - M(Bool, fsync_after_insert, false, "Do fsync for every inserted part. Significantly decreases performance of inserts, not recommended to use with wide parts.", 0) \ - M(Bool, fsync_part_directory, false, "Do fsync for part directory after all part operations (writes, renames, etc.).", 0) \ - M(UInt64, non_replicated_deduplication_window, 0, "How many last blocks of hashes should be kept on disk (0 - disabled).", 0) \ - M(UInt64, max_parts_to_merge_at_once, 100, "Max amount of parts which can be merged at once (0 - disabled). Doesn't affect OPTIMIZE FINAL query.", 0) \ - M(UInt64, merge_selecting_sleep_ms, 5000, "Minimum time to wait before trying to select parts to merge again after no parts were selected. A lower setting will trigger selecting tasks in background_schedule_pool frequently which result in large amount of requests to zookeeper in large-scale clusters", 0) \ - M(UInt64, max_merge_selecting_sleep_ms, 60000, "Maximum time to wait before trying to select parts to merge again after no parts were selected. A lower setting will trigger selecting tasks in background_schedule_pool frequently which result in large amount of requests to zookeeper in large-scale clusters", 0) \ - M(Float, merge_selecting_sleep_slowdown_factor, 1.2f, "The sleep time for merge selecting task is multiplied by this factor when there's nothing to merge and divided when a merge was assigned", 0) \ - M(UInt64, merge_tree_clear_old_temporary_directories_interval_seconds, 60, "The period of executing the clear old temporary directories operation in background.", 0) \ - M(UInt64, merge_tree_clear_old_parts_interval_seconds, 1, "The period of executing the clear old parts operation in background.", 0) \ - M(UInt64, min_age_to_force_merge_seconds, 0, "If all parts in a certain range are older than this value, range will be always eligible for merging. Set to 0 to disable.", 0) \ - M(Bool, min_age_to_force_merge_on_partition_only, false, "Whether min_age_to_force_merge_seconds should be applied only on the entire partition and not on subset.", false) \ - M(UInt64, number_of_free_entries_in_pool_to_execute_optimize_entire_partition, 25, "When there is less than specified number of free entries in pool, do not try to execute optimize entire partition with a merge (this merge is created when set min_age_to_force_merge_seconds > 0 and min_age_to_force_merge_on_partition_only = true). This is to leave free threads for regular merges and avoid \"Too many parts\"", 0) \ - M(Bool, remove_rolled_back_parts_immediately, 1, "Setting for an incomplete experimental feature.", 0) \ - M(UInt64, replicated_max_mutations_in_one_entry, 10000, "Max number of mutation commands that can be merged together and executed in one MUTATE_PART entry (0 means unlimited)", 0) \ - M(UInt64, number_of_mutations_to_delay, 500, "If table has at least that many unfinished mutations, artificially slow down mutations of table. Disabled if set to 0", 0) \ - M(UInt64, number_of_mutations_to_throw, 1000, "If table has at least that many unfinished mutations, throw 'Too many mutations' exception. Disabled if set to 0", 0) \ - M(UInt64, min_delay_to_mutate_ms, 10, "Min delay of mutating MergeTree table in milliseconds, if there are a lot of unfinished mutations", 0) \ - M(UInt64, max_delay_to_mutate_ms, 1000, "Max delay of mutating MergeTree table in milliseconds, if there are a lot of unfinished mutations", 0) \ - M(Bool, exclude_deleted_rows_for_part_size_in_merge, false, "Use an estimated source part size (excluding lightweight deleted rows) when selecting parts to merge", 0) \ - M(String, merge_workload, "", "Name of workload to be used to access resources for merges", 0) \ - M(String, mutation_workload, "", "Name of workload to be used to access resources for mutations", 0) \ - M(Milliseconds, background_task_preferred_step_execution_time_ms, 50, "Target time to execution of one step of merge or mutation. Can be exceeded if one step takes longer time", 0) \ - M(MergeSelectorAlgorithm, merge_selector_algorithm, MergeSelectorAlgorithm::SIMPLE, "The algorithm to select parts for merges assignment", 0) \ + DECLARE(UInt64, merge_max_block_size, 8192, "How many rows in blocks should be formed for merge operations. By default has the same value as `index_granularity`.", 0) \ + DECLARE(UInt64, merge_max_block_size_bytes, 10 * 1024 * 1024, "How many bytes in blocks should be formed for merge operations. By default has the same value as `index_granularity_bytes`.", 0) \ + DECLARE(UInt64, max_bytes_to_merge_at_max_space_in_pool, 150ULL * 1024 * 1024 * 1024, "Maximum in total size of parts to merge, when there are maximum free threads in background pool (or entries in replication queue).", 0) \ + DECLARE(UInt64, max_bytes_to_merge_at_min_space_in_pool, 1024 * 1024, "Maximum in total size of parts to merge, when there are minimum free threads in background pool (or entries in replication queue).", 0) \ + DECLARE(UInt64, max_replicated_merges_in_queue, 1000, "How many tasks of merging and mutating parts are allowed simultaneously in ReplicatedMergeTree queue.", 0) \ + DECLARE(UInt64, max_replicated_mutations_in_queue, 8, "How many tasks of mutating parts are allowed simultaneously in ReplicatedMergeTree queue.", 0) \ + DECLARE(UInt64, max_replicated_merges_with_ttl_in_queue, 1, "How many tasks of merging parts with TTL are allowed simultaneously in ReplicatedMergeTree queue.", 0) \ + DECLARE(UInt64, number_of_free_entries_in_pool_to_lower_max_size_of_merge, 8, "When there is less than specified number of free entries in pool (or replicated queue), start to lower maximum size of merge to process (or to put in queue). This is to allow small merges to process - not filling the pool with long running merges.", 0) \ + DECLARE(UInt64, number_of_free_entries_in_pool_to_execute_mutation, 20, "When there is less than specified number of free entries in pool, do not execute part mutations. This is to leave free threads for regular merges and avoid \"Too many parts\"", 0) \ + DECLARE(UInt64, max_number_of_mutations_for_replica, 0, "Limit the number of part mutations per replica to the specified amount. Zero means no limit on the number of mutations per replica (the execution can still be constrained by other settings).", 0) \ + DECLARE(UInt64, max_number_of_merges_with_ttl_in_pool, 2, "When there is more than specified number of merges with TTL entries in pool, do not assign new merge with TTL. This is to leave free threads for regular merges and avoid \"Too many parts\"", 0) \ + DECLARE(Seconds, old_parts_lifetime, 8 * 60, "How many seconds to keep obsolete parts.", 0) \ + DECLARE(Seconds, temporary_directories_lifetime, 86400, "How many seconds to keep tmp_-directories. You should not lower this value because merges and mutations may not be able to work with low value of this setting.", 0) \ + DECLARE(Seconds, lock_acquire_timeout_for_background_operations, DBMS_DEFAULT_LOCK_ACQUIRE_TIMEOUT_SEC, "For background operations like merges, mutations etc. How many seconds before failing to acquire table locks.", 0) \ + DECLARE(UInt64, min_rows_to_fsync_after_merge, 0, "Minimal number of rows to do fsync for part after merge (0 - disabled)", 0) \ + DECLARE(UInt64, min_compressed_bytes_to_fsync_after_merge, 0, "Minimal number of compressed bytes to do fsync for part after merge (0 - disabled)", 0) \ + DECLARE(UInt64, min_compressed_bytes_to_fsync_after_fetch, 0, "Minimal number of compressed bytes to do fsync for part after fetch (0 - disabled)", 0) \ + DECLARE(Bool, fsync_after_insert, false, "Do fsync for every inserted part. Significantly decreases performance of inserts, not recommended to use with wide parts.", 0) \ + DECLARE(Bool, fsync_part_directory, false, "Do fsync for part directory after all part operations (writes, renames, etc.).", 0) \ + DECLARE(UInt64, non_replicated_deduplication_window, 0, "How many last blocks of hashes should be kept on disk (0 - disabled).", 0) \ + DECLARE(UInt64, max_parts_to_merge_at_once, 100, "Max amount of parts which can be merged at once (0 - disabled). Doesn't affect OPTIMIZE FINAL query.", 0) \ + DECLARE(UInt64, merge_selecting_sleep_ms, 5000, "Minimum time to wait before trying to select parts to merge again after no parts were selected. A lower setting will trigger selecting tasks in background_schedule_pool frequently which result in large amount of requests to zookeeper in large-scale clusters", 0) \ + DECLARE(UInt64, max_merge_selecting_sleep_ms, 60000, "Maximum time to wait before trying to select parts to merge again after no parts were selected. A lower setting will trigger selecting tasks in background_schedule_pool frequently which result in large amount of requests to zookeeper in large-scale clusters", 0) \ + DECLARE(Float, merge_selecting_sleep_slowdown_factor, 1.2f, "The sleep time for merge selecting task is multiplied by this factor when there's nothing to merge and divided when a merge was assigned", 0) \ + DECLARE(UInt64, merge_tree_clear_old_temporary_directories_interval_seconds, 60, "The period of executing the clear old temporary directories operation in background.", 0) \ + DECLARE(UInt64, merge_tree_clear_old_parts_interval_seconds, 1, "The period of executing the clear old parts operation in background.", 0) \ + DECLARE(UInt64, min_age_to_force_merge_seconds, 0, "If all parts in a certain range are older than this value, range will be always eligible for merging. Set to 0 to disable.", 0) \ + DECLARE(Bool, min_age_to_force_merge_on_partition_only, false, "Whether min_age_to_force_merge_seconds should be applied only on the entire partition and not on subset.", false) \ + DECLARE(UInt64, number_of_free_entries_in_pool_to_execute_optimize_entire_partition, 25, "When there is less than specified number of free entries in pool, do not try to execute optimize entire partition with a merge (this merge is created when set min_age_to_force_merge_seconds > 0 and min_age_to_force_merge_on_partition_only = true). This is to leave free threads for regular merges and avoid \"Too many parts\"", 0) \ + DECLARE(Bool, remove_rolled_back_parts_immediately, 1, "Setting for an incomplete experimental feature.", EXPERIMENTAL) \ + DECLARE(UInt64, replicated_max_mutations_in_one_entry, 10000, "Max number of mutation commands that can be merged together and executed in one MUTATE_PART entry (0 means unlimited)", 0) \ + DECLARE(UInt64, number_of_mutations_to_delay, 500, "If table has at least that many unfinished mutations, artificially slow down mutations of table. Disabled if set to 0", 0) \ + DECLARE(UInt64, number_of_mutations_to_throw, 1000, "If table has at least that many unfinished mutations, throw 'Too many mutations' exception. Disabled if set to 0", 0) \ + DECLARE(UInt64, min_delay_to_mutate_ms, 10, "Min delay of mutating MergeTree table in milliseconds, if there are a lot of unfinished mutations", 0) \ + DECLARE(UInt64, max_delay_to_mutate_ms, 1000, "Max delay of mutating MergeTree table in milliseconds, if there are a lot of unfinished mutations", 0) \ + DECLARE(Bool, exclude_deleted_rows_for_part_size_in_merge, false, "Use an estimated source part size (excluding lightweight deleted rows) when selecting parts to merge", 0) \ + DECLARE(String, merge_workload, "", "Name of workload to be used to access resources for merges", 0) \ + DECLARE(String, mutation_workload, "", "Name of workload to be used to access resources for mutations", 0) \ + DECLARE(Milliseconds, background_task_preferred_step_execution_time_ms, 50, "Target time to execution of one step of merge or mutation. Can be exceeded if one step takes longer time", 0) \ + DECLARE(MergeSelectorAlgorithm, merge_selector_algorithm, MergeSelectorAlgorithm::SIMPLE, "The algorithm to select parts for merges assignment", EXPERIMENTAL) \ + DECLARE(Bool, merge_selector_enable_heuristic_to_remove_small_parts_at_right, true, "Enable heuristic for selecting parts for merge which removes parts from right side of range, if their size is less than specified ratio (0.01) of sum_size. Works for Simple and StochasticSimple merge selectors", 0) \ + DECLARE(Float, merge_selector_base, 5.0, "Affects write amplification of assigned merges (expert level setting, don't change if you don't understand what it is doing). Works for Simple and StochasticSimple merge selectors", 0) \ \ /** Inserts settings. */ \ - M(UInt64, parts_to_delay_insert, 1000, "If table contains at least that many active parts in single partition, artificially slow down insert into table. Disabled if set to 0", 0) \ - M(UInt64, inactive_parts_to_delay_insert, 0, "If table contains at least that many inactive parts in single partition, artificially slow down insert into table.", 0) \ - M(UInt64, parts_to_throw_insert, 3000, "If more than this number active parts in single partition, throw 'Too many parts ...' exception.", 0) \ - M(UInt64, inactive_parts_to_throw_insert, 0, "If more than this number inactive parts in single partition, throw 'Too many inactive parts ...' exception.", 0) \ - M(UInt64, max_avg_part_size_for_too_many_parts, 1ULL * 1024 * 1024 * 1024, "The 'too many parts' check according to 'parts_to_delay_insert' and 'parts_to_throw_insert' will be active only if the average part size (in the relevant partition) is not larger than the specified threshold. If it is larger than the specified threshold, the INSERTs will be neither delayed or rejected. This allows to have hundreds of terabytes in a single table on a single server if the parts are successfully merged to larger parts. This does not affect the thresholds on inactive parts or total parts.", 0) \ - M(UInt64, max_delay_to_insert, 1, "Max delay of inserting data into MergeTree table in seconds, if there are a lot of unmerged parts in single partition.", 0) \ - M(UInt64, min_delay_to_insert_ms, 10, "Min delay of inserting data into MergeTree table in milliseconds, if there are a lot of unmerged parts in single partition.", 0) \ - M(UInt64, max_parts_in_total, 100000, "If more than this number active parts in all partitions in total, throw 'Too many parts ...' exception.", 0) \ - M(Bool, async_insert, false, "If true, data from INSERT query is stored in queue and later flushed to table in background.", 0) \ - M(Bool, add_implicit_sign_column_constraint_for_collapsing_engine, false, "If true, add implicit constraint for sign column for CollapsingMergeTree engine.", 0) \ - M(Milliseconds, sleep_before_commit_local_part_in_replicated_table_ms, 0, "For testing. Do not change it.", 0) \ - M(Bool, optimize_row_order, false, "Allow reshuffling of rows during part inserts and merges to improve the compressibility of the new part", 0) \ - M(Bool, use_adaptive_write_buffer_for_dynamic_subcolumns, true, "Allow to use adaptive writer buffers during writing dynamic subcolumns to reduce memory usage", 0) \ - M(UInt64, adaptive_write_buffer_initial_size, 16 * 1024, "Initial size of an adaptive write buffer", 0) \ - M(UInt64, min_free_disk_bytes_to_perform_insert, 0, "Minimum free disk space bytes to perform an insert.", 0) \ - M(Float, min_free_disk_ratio_to_perform_insert, 0.0, "Minimum free disk space ratio to perform an insert.", 0) \ + DECLARE(UInt64, parts_to_delay_insert, 1000, "If table contains at least that many active parts in single partition, artificially slow down insert into table. Disabled if set to 0", 0) \ + DECLARE(UInt64, inactive_parts_to_delay_insert, 0, "If table contains at least that many inactive parts in single partition, artificially slow down insert into table.", 0) \ + DECLARE(UInt64, parts_to_throw_insert, 3000, "If more than this number active parts in single partition, throw 'Too many parts ...' exception.", 0) \ + DECLARE(UInt64, inactive_parts_to_throw_insert, 0, "If more than this number inactive parts in single partition, throw 'Too many inactive parts ...' exception.", 0) \ + DECLARE(UInt64, max_avg_part_size_for_too_many_parts, 1ULL * 1024 * 1024 * 1024, "The 'too many parts' check according to 'parts_to_delay_insert' and 'parts_to_throw_insert' will be active only if the average part size (in the relevant partition) is not larger than the specified threshold. If it is larger than the specified threshold, the INSERTs will be neither delayed or rejected. This allows to have hundreds of terabytes in a single table on a single server if the parts are successfully merged to larger parts. This does not affect the thresholds on inactive parts or total parts.", 0) \ + DECLARE(UInt64, max_delay_to_insert, 1, "Max delay of inserting data into MergeTree table in seconds, if there are a lot of unmerged parts in single partition.", 0) \ + DECLARE(UInt64, min_delay_to_insert_ms, 10, "Min delay of inserting data into MergeTree table in milliseconds, if there are a lot of unmerged parts in single partition.", 0) \ + DECLARE(UInt64, max_parts_in_total, 100000, "If more than this number active parts in all partitions in total, throw 'Too many parts ...' exception.", 0) \ + DECLARE(Bool, async_insert, false, "If true, data from INSERT query is stored in queue and later flushed to table in background.", 0) \ + DECLARE(Bool, add_implicit_sign_column_constraint_for_collapsing_engine, false, "If true, add implicit constraint for sign column for CollapsingMergeTree engine.", 0) \ + DECLARE(Milliseconds, sleep_before_commit_local_part_in_replicated_table_ms, 0, "For testing. Do not change it.", 0) \ + DECLARE(Bool, optimize_row_order, false, "Allow reshuffling of rows during part inserts and merges to improve the compressibility of the new part", 0) \ + DECLARE(Bool, use_adaptive_write_buffer_for_dynamic_subcolumns, true, "Allow to use adaptive writer buffers during writing dynamic subcolumns to reduce memory usage", 0) \ + DECLARE(UInt64, adaptive_write_buffer_initial_size, 16 * 1024, "Initial size of an adaptive write buffer", 0) \ + DECLARE(UInt64, min_free_disk_bytes_to_perform_insert, 0, "Minimum free disk space bytes to perform an insert.", 0) \ + DECLARE(Float, min_free_disk_ratio_to_perform_insert, 0.0, "Minimum free disk space ratio to perform an insert.", 0) \ \ /* Part removal settings. */ \ - M(UInt64, simultaneous_parts_removal_limit, 0, "Maximum number of parts to remove during one CleanupThread iteration (0 means unlimited).", 0) \ + DECLARE(UInt64, simultaneous_parts_removal_limit, 0, "Maximum number of parts to remove during one CleanupThread iteration (0 means unlimited).", 0) \ \ /** Replication settings. */ \ - M(UInt64, replicated_deduplication_window, 1000, "How many last blocks of hashes should be kept in ZooKeeper (old blocks will be deleted).", 0) \ - M(UInt64, replicated_deduplication_window_seconds, 7 * 24 * 60 * 60 /* one week */, "Similar to \"replicated_deduplication_window\", but determines old blocks by their lifetime. Hash of an inserted block will be deleted (and the block will not be deduplicated after) if it outside of one \"window\". You can set very big replicated_deduplication_window to avoid duplicating INSERTs during that period of time.", 0) \ - M(UInt64, replicated_deduplication_window_for_async_inserts, 10000, "How many last hash values of async_insert blocks should be kept in ZooKeeper (old blocks will be deleted).", 0) \ - M(UInt64, replicated_deduplication_window_seconds_for_async_inserts, 7 * 24 * 60 * 60 /* one week */, "Similar to \"replicated_deduplication_window_for_async_inserts\", but determines old blocks by their lifetime. Hash of an inserted block will be deleted (and the block will not be deduplicated after) if it outside of one \"window\". You can set very big replicated_deduplication_window to avoid duplicating INSERTs during that period of time.", 0) \ - M(Milliseconds, async_block_ids_cache_update_wait_ms, 100, "How long each insert iteration will wait for async_block_ids_cache update", 0) \ - M(Bool, use_async_block_ids_cache, true, "Use in-memory cache to filter duplicated async inserts based on block ids", 0) \ - M(UInt64, max_replicated_logs_to_keep, 1000, "How many records may be in log, if there is inactive replica. Inactive replica becomes lost when when this number exceed.", 0) \ - M(UInt64, min_replicated_logs_to_keep, 10, "Keep about this number of last records in ZooKeeper log, even if they are obsolete. It doesn't affect work of tables: used only to diagnose ZooKeeper log before cleaning.", 0) \ - M(Seconds, prefer_fetch_merged_part_time_threshold, 3600, "If time passed after replication log entry creation exceeds this threshold and sum size of parts is greater than \"prefer_fetch_merged_part_size_threshold\", prefer fetching merged part from replica instead of doing merge locally. To speed up very long merges.", 0) \ - M(UInt64, prefer_fetch_merged_part_size_threshold, 10ULL * 1024 * 1024 * 1024, "If sum size of parts exceeds this threshold and time passed after replication log entry creation is greater than \"prefer_fetch_merged_part_time_threshold\", prefer fetching merged part from replica instead of doing merge locally. To speed up very long merges.", 0) \ - M(Seconds, execute_merges_on_single_replica_time_threshold, 0, "When greater than zero only a single replica starts the merge immediately, others wait up to that amount of time to download the result instead of doing merges locally. If the chosen replica doesn't finish the merge during that amount of time, fallback to standard behavior happens.", 0) \ - M(Seconds, remote_fs_execute_merges_on_single_replica_time_threshold, 3 * 60 * 60, "When greater than zero only a single replica starts the merge immediately if merged part on shared storage and 'allow_remote_fs_zero_copy_replication' is enabled.", 0) \ - M(Seconds, try_fetch_recompressed_part_timeout, 7200, "Recompression works slow in most cases, so we don't start merge with recompression until this timeout and trying to fetch recompressed part from replica which assigned this merge with recompression.", 0) \ - M(Bool, always_fetch_merged_part, false, "If true, replica never merge parts and always download merged parts from other replicas.", 0) \ - M(UInt64, max_suspicious_broken_parts, 100, "Max broken parts, if more - deny automatic deletion.", 0) \ - M(UInt64, max_suspicious_broken_parts_bytes, 1ULL * 1024 * 1024 * 1024, "Max size of all broken parts, if more - deny automatic deletion.", 0) \ - M(UInt64, max_files_to_modify_in_alter_columns, 75, "Not apply ALTER if number of files for modification(deletion, addition) more than this.", 0) \ - M(UInt64, max_files_to_remove_in_alter_columns, 50, "Not apply ALTER, if number of files for deletion more than this.", 0) \ - M(Float, replicated_max_ratio_of_wrong_parts, 0.5, "If ratio of wrong parts to total number of parts is less than this - allow to start.", 0) \ - M(Bool, replicated_can_become_leader, true, "If true, Replicated tables replicas on this node will try to acquire leadership.", 0) \ - M(Seconds, zookeeper_session_expiration_check_period, 60, "ZooKeeper session expiration check period, in seconds.", 0) \ - M(Seconds, initialization_retry_period, 60, "Retry period for table initialization, in seconds.", 0) \ - M(Bool, detach_old_local_parts_when_cloning_replica, true, "Do not remove old local parts when repairing lost replica.", 0) \ - M(Bool, detach_not_byte_identical_parts, false, "Do not remove non byte-idential parts for ReplicatedMergeTree, instead detach them (maybe useful for further analysis).", 0) \ - M(UInt64, max_replicated_fetches_network_bandwidth, 0, "The maximum speed of data exchange over the network in bytes per second for replicated fetches. Zero means unlimited.", 0) \ - M(UInt64, max_replicated_sends_network_bandwidth, 0, "The maximum speed of data exchange over the network in bytes per second for replicated sends. Zero means unlimited.", 0) \ - M(Milliseconds, wait_for_unique_parts_send_before_shutdown_ms, 0, "Before shutdown table will wait for required amount time for unique parts (exist only on current replica) to be fetched by other replicas (0 means disabled).", 0) \ - M(Float, fault_probability_before_part_commit, 0, "For testing. Do not change it.", 0) \ - M(Float, fault_probability_after_part_commit, 0, "For testing. Do not change it.", 0) \ - M(Bool, shared_merge_tree_disable_merges_and_mutations_assignment, false, "Only available in ClickHouse Cloud", 0) \ - M(Float, shared_merge_tree_partitions_hint_ratio_to_reload_merge_pred_for_mutations, 0.5, "Only available in ClickHouse Cloud", 0) \ - M(UInt64, shared_merge_tree_parts_load_batch_size, 32, "Only available in ClickHouse Cloud", 0) \ + DECLARE(UInt64, replicated_deduplication_window, 1000, "How many last blocks of hashes should be kept in ZooKeeper (old blocks will be deleted).", 0) \ + DECLARE(UInt64, replicated_deduplication_window_seconds, 7 * 24 * 60 * 60 /* one week */, "Similar to \"replicated_deduplication_window\", but determines old blocks by their lifetime. Hash of an inserted block will be deleted (and the block will not be deduplicated after) if it outside of one \"window\". You can set very big replicated_deduplication_window to avoid duplicating INSERTs during that period of time.", 0) \ + DECLARE(UInt64, replicated_deduplication_window_for_async_inserts, 10000, "How many last hash values of async_insert blocks should be kept in ZooKeeper (old blocks will be deleted).", 0) \ + DECLARE(UInt64, replicated_deduplication_window_seconds_for_async_inserts, 7 * 24 * 60 * 60 /* one week */, "Similar to \"replicated_deduplication_window_for_async_inserts\", but determines old blocks by their lifetime. Hash of an inserted block will be deleted (and the block will not be deduplicated after) if it outside of one \"window\". You can set very big replicated_deduplication_window to avoid duplicating INSERTs during that period of time.", 0) \ + DECLARE(Milliseconds, async_block_ids_cache_update_wait_ms, 100, "How long each insert iteration will wait for async_block_ids_cache update", 0) \ + DECLARE(Bool, use_async_block_ids_cache, true, "Use in-memory cache to filter duplicated async inserts based on block ids", 0) \ + DECLARE(UInt64, max_replicated_logs_to_keep, 1000, "How many records may be in log, if there is inactive replica. Inactive replica becomes lost when when this number exceed.", 0) \ + DECLARE(UInt64, min_replicated_logs_to_keep, 10, "Keep about this number of last records in ZooKeeper log, even if they are obsolete. It doesn't affect work of tables: used only to diagnose ZooKeeper log before cleaning.", 0) \ + DECLARE(Seconds, prefer_fetch_merged_part_time_threshold, 3600, "If time passed after replication log entry creation exceeds this threshold and sum size of parts is greater than \"prefer_fetch_merged_part_size_threshold\", prefer fetching merged part from replica instead of doing merge locally. To speed up very long merges.", 0) \ + DECLARE(UInt64, prefer_fetch_merged_part_size_threshold, 10ULL * 1024 * 1024 * 1024, "If sum size of parts exceeds this threshold and time passed after replication log entry creation is greater than \"prefer_fetch_merged_part_time_threshold\", prefer fetching merged part from replica instead of doing merge locally. To speed up very long merges.", 0) \ + DECLARE(Seconds, execute_merges_on_single_replica_time_threshold, 0, "When greater than zero only a single replica starts the merge immediately, others wait up to that amount of time to download the result instead of doing merges locally. If the chosen replica doesn't finish the merge during that amount of time, fallback to standard behavior happens.", 0) \ + DECLARE(Seconds, remote_fs_execute_merges_on_single_replica_time_threshold, 3 * 60 * 60, "When greater than zero only a single replica starts the merge immediately if merged part on shared storage and 'allow_remote_fs_zero_copy_replication' is enabled.", 0) \ + DECLARE(Seconds, try_fetch_recompressed_part_timeout, 7200, "Recompression works slow in most cases, so we don't start merge with recompression until this timeout and trying to fetch recompressed part from replica which assigned this merge with recompression.", 0) \ + DECLARE(Bool, always_fetch_merged_part, false, "If true, replica never merge parts and always download merged parts from other replicas.", 0) \ + DECLARE(UInt64, max_suspicious_broken_parts, 100, "Max broken parts, if more - deny automatic deletion.", 0) \ + DECLARE(UInt64, max_suspicious_broken_parts_bytes, 1ULL * 1024 * 1024 * 1024, "Max size of all broken parts, if more - deny automatic deletion.", 0) \ + DECLARE(UInt64, max_files_to_modify_in_alter_columns, 75, "Not apply ALTER if number of files for modification(deletion, addition) more than this.", 0) \ + DECLARE(UInt64, max_files_to_remove_in_alter_columns, 50, "Not apply ALTER, if number of files for deletion more than this.", 0) \ + DECLARE(Float, replicated_max_ratio_of_wrong_parts, 0.5, "If ratio of wrong parts to total number of parts is less than this - allow to start.", 0) \ + DECLARE(Bool, replicated_can_become_leader, true, "If true, Replicated tables replicas on this node will try to acquire leadership.", 0) \ + DECLARE(Seconds, zookeeper_session_expiration_check_period, 60, "ZooKeeper session expiration check period, in seconds.", 0) \ + DECLARE(Seconds, initialization_retry_period, 60, "Retry period for table initialization, in seconds.", 0) \ + DECLARE(Bool, detach_old_local_parts_when_cloning_replica, true, "Do not remove old local parts when repairing lost replica.", 0) \ + DECLARE(Bool, detach_not_byte_identical_parts, false, "Do not remove non byte-idential parts for ReplicatedMergeTree, instead detach them (maybe useful for further analysis).", 0) \ + DECLARE(UInt64, max_replicated_fetches_network_bandwidth, 0, "The maximum speed of data exchange over the network in bytes per second for replicated fetches. Zero means unlimited.", 0) \ + DECLARE(UInt64, max_replicated_sends_network_bandwidth, 0, "The maximum speed of data exchange over the network in bytes per second for replicated sends. Zero means unlimited.", 0) \ + DECLARE(Milliseconds, wait_for_unique_parts_send_before_shutdown_ms, 0, "Before shutdown table will wait for required amount time for unique parts (exist only on current replica) to be fetched by other replicas (0 means disabled).", 0) \ + DECLARE(Float, fault_probability_before_part_commit, 0, "For testing. Do not change it.", 0) \ + DECLARE(Float, fault_probability_after_part_commit, 0, "For testing. Do not change it.", 0) \ + DECLARE(Bool, shared_merge_tree_disable_merges_and_mutations_assignment, false, "Only available in ClickHouse Cloud", 0) \ + DECLARE(Float, shared_merge_tree_partitions_hint_ratio_to_reload_merge_pred_for_mutations, 0.5, "Only available in ClickHouse Cloud", 0) \ + DECLARE(UInt64, shared_merge_tree_parts_load_batch_size, 32, "Only available in ClickHouse Cloud", 0) \ \ /** Check delay of replicas settings. */ \ - M(UInt64, min_relative_delay_to_measure, 120, "Calculate relative replica delay only if absolute delay is not less that this value.", 0) \ - M(UInt64, cleanup_delay_period, 30, "Minimum period to clean old queue logs, blocks hashes and parts.", 0) \ - M(UInt64, max_cleanup_delay_period, 300, "Maximum period to clean old queue logs, blocks hashes and parts.", 0) \ - M(UInt64, cleanup_delay_period_random_add, 10, "Add uniformly distributed value from 0 to x seconds to cleanup_delay_period to avoid thundering herd effect and subsequent DoS of ZooKeeper in case of very large number of tables.", 0) \ - M(UInt64, cleanup_thread_preferred_points_per_iteration, 150, "Preferred batch size for background cleanup (points are abstract but 1 point is approximately equivalent to 1 inserted block).", 0) \ - M(UInt64, cleanup_threads, 128, "Only available in ClickHouse Cloud", 0) \ - M(UInt64, kill_delay_period, 30, "Only available in ClickHouse Cloud", 0) \ - M(UInt64, kill_delay_period_random_add, 10, "Only available in ClickHouse Cloud", 0) \ - M(UInt64, kill_threads, 128, "Only available in ClickHouse Cloud", 0) \ - M(UInt64, min_relative_delay_to_close, 300, "Minimal delay from other replicas to close, stop serving requests and not return Ok during status check.", 0) \ - M(UInt64, min_absolute_delay_to_close, 0, "Minimal absolute delay to close, stop serving requests and not return Ok during status check.", 0) \ - M(UInt64, enable_vertical_merge_algorithm, 1, "Enable usage of Vertical merge algorithm.", 0) \ - M(UInt64, vertical_merge_algorithm_min_rows_to_activate, 16 * 8192, "Minimal (approximate) sum of rows in merging parts to activate Vertical merge algorithm.", 0) \ - M(UInt64, vertical_merge_algorithm_min_bytes_to_activate, 0, "Minimal (approximate) uncompressed size in bytes in merging parts to activate Vertical merge algorithm.", 0) \ - M(UInt64, vertical_merge_algorithm_min_columns_to_activate, 11, "Minimal amount of non-PK columns to activate Vertical merge algorithm.", 0) \ - M(Bool, vertical_merge_remote_filesystem_prefetch, true, "If true prefetching of data from remote filesystem is used for the next column during merge", 0) \ - M(UInt64, max_postpone_time_for_failed_mutations_ms, 5ULL * 60 * 1000, "The maximum postpone time for failed mutations.", 0) \ + DECLARE(UInt64, min_relative_delay_to_measure, 120, "Calculate relative replica delay only if absolute delay is not less that this value.", 0) \ + DECLARE(UInt64, cleanup_delay_period, 30, "Minimum period to clean old queue logs, blocks hashes and parts.", 0) \ + DECLARE(UInt64, max_cleanup_delay_period, 300, "Maximum period to clean old queue logs, blocks hashes and parts.", 0) \ + DECLARE(UInt64, cleanup_delay_period_random_add, 10, "Add uniformly distributed value from 0 to x seconds to cleanup_delay_period to avoid thundering herd effect and subsequent DoS of ZooKeeper in case of very large number of tables.", 0) \ + DECLARE(UInt64, cleanup_thread_preferred_points_per_iteration, 150, "Preferred batch size for background cleanup (points are abstract but 1 point is approximately equivalent to 1 inserted block).", 0) \ + DECLARE(UInt64, cleanup_threads, 128, "Only available in ClickHouse Cloud", 0) \ + DECLARE(UInt64, kill_delay_period, 30, "Only available in ClickHouse Cloud", 0) \ + DECLARE(UInt64, kill_delay_period_random_add, 10, "Only available in ClickHouse Cloud", 0) \ + DECLARE(UInt64, kill_threads, 128, "Only available in ClickHouse Cloud", 0) \ + DECLARE(UInt64, min_relative_delay_to_close, 300, "Minimal delay from other replicas to close, stop serving requests and not return Ok during status check.", 0) \ + DECLARE(UInt64, min_absolute_delay_to_close, 0, "Minimal absolute delay to close, stop serving requests and not return Ok during status check.", 0) \ + DECLARE(UInt64, enable_vertical_merge_algorithm, 1, "Enable usage of Vertical merge algorithm.", 0) \ + DECLARE(UInt64, vertical_merge_algorithm_min_rows_to_activate, 16 * 8192, "Minimal (approximate) sum of rows in merging parts to activate Vertical merge algorithm.", 0) \ + DECLARE(UInt64, vertical_merge_algorithm_min_bytes_to_activate, 0, "Minimal (approximate) uncompressed size in bytes in merging parts to activate Vertical merge algorithm.", 0) \ + DECLARE(UInt64, vertical_merge_algorithm_min_columns_to_activate, 11, "Minimal amount of non-PK columns to activate Vertical merge algorithm.", 0) \ + DECLARE(Bool, vertical_merge_remote_filesystem_prefetch, true, "If true prefetching of data from remote filesystem is used for the next column during merge", 0) \ + DECLARE(UInt64, max_postpone_time_for_failed_mutations_ms, 5ULL * 60 * 1000, "The maximum postpone time for failed mutations.", 0) \ \ /** Compatibility settings */ \ - M(Bool, allow_suspicious_indices, false, "Reject primary/secondary indexes and sorting keys with identical expressions", 0) \ - M(Bool, compatibility_allow_sampling_expression_not_in_primary_key, false, "Allow to create a table with sampling expression not in primary key. This is needed only to temporarily allow to run the server with wrong tables for backward compatibility.", 0) \ - M(Bool, use_minimalistic_checksums_in_zookeeper, true, "Use small format (dozens bytes) for part checksums in ZooKeeper instead of ordinary ones (dozens KB). Before enabling check that all replicas support new format.", 0) \ - M(Bool, use_minimalistic_part_header_in_zookeeper, true, "Store part header (checksums and columns) in a compact format and a single part znode instead of separate znodes (/columns and /checksums). This can dramatically reduce snapshot size in ZooKeeper. Before enabling check that all replicas support new format.", 0) \ - M(UInt64, finished_mutations_to_keep, 100, "How many records about mutations that are done to keep. If zero, then keep all of them.", 0) \ - M(UInt64, min_merge_bytes_to_use_direct_io, 10ULL * 1024 * 1024 * 1024, "Minimal amount of bytes to enable O_DIRECT in merge (0 - disabled).", 0) \ - M(UInt64, index_granularity_bytes, 10 * 1024 * 1024, "Approximate amount of bytes in single granule (0 - disabled).", 0) \ - M(UInt64, min_index_granularity_bytes, 1024, "Minimum amount of bytes in single granule.", 1024) \ - M(Int64, merge_with_ttl_timeout, 3600 * 4, "Minimal time in seconds, when merge with delete TTL can be repeated.", 0) \ - M(Int64, merge_with_recompression_ttl_timeout, 3600 * 4, "Minimal time in seconds, when merge with recompression TTL can be repeated.", 0) \ - M(Bool, ttl_only_drop_parts, false, "Only drop altogether the expired parts and not partially prune them.", 0) \ - M(Bool, materialize_ttl_recalculate_only, false, "Only recalculate ttl info when MATERIALIZE TTL", 0) \ - M(Bool, enable_mixed_granularity_parts, true, "Enable parts with adaptive and non adaptive granularity", 0) \ - M(UInt64, concurrent_part_removal_threshold, 100, "Activate concurrent part removal (see 'max_part_removal_threads') only if the number of inactive data parts is at least this.", 0) \ - M(UInt64, zero_copy_concurrent_part_removal_max_split_times, 5, "Max recursion depth for splitting independent Outdated parts ranges into smaller subranges (highly not recommended to change)", 0) \ - M(Float, zero_copy_concurrent_part_removal_max_postpone_ratio, static_cast(0.05), "Max percentage of top level parts to postpone removal in order to get smaller independent ranges (highly not recommended to change)", 0) \ - M(String, storage_policy, "default", "Name of storage disk policy", 0) \ - M(String, disk, "", "Name of storage disk. Can be specified instead of storage policy.", 0) \ - M(Bool, allow_nullable_key, false, "Allow Nullable types as primary keys.", 0) \ - M(Bool, remove_empty_parts, true, "Remove empty parts after they were pruned by TTL, mutation, or collapsing merge algorithm.", 0) \ - M(Bool, assign_part_uuids, false, "Generate UUIDs for parts. Before enabling check that all replicas support new format.", 0) \ - M(Int64, max_partitions_to_read, -1, "Limit the max number of partitions that can be accessed in one query. <= 0 means unlimited. This setting is the default that can be overridden by the query-level setting with the same name.", 0) \ - M(UInt64, max_concurrent_queries, 0, "Max number of concurrently executed queries related to the MergeTree table (0 - disabled). Queries will still be limited by other max_concurrent_queries settings.", 0) \ - M(UInt64, min_marks_to_honor_max_concurrent_queries, 0, "Minimal number of marks to honor the MergeTree-level's max_concurrent_queries (0 - disabled). Queries will still be limited by other max_concurrent_queries settings.", 0) \ - M(UInt64, min_bytes_to_rebalance_partition_over_jbod, 0, "Minimal amount of bytes to enable part rebalance over JBOD array (0 - disabled).", 0) \ - M(Bool, check_sample_column_is_correct, true, "Check columns or columns by hash for sampling are unsigned integer.", 0) \ - M(Bool, allow_vertical_merges_from_compact_to_wide_parts, true, "Allows vertical merges from compact to wide parts. This settings must have the same value on all replicas", 0) \ - M(Bool, enable_the_endpoint_id_with_zookeeper_name_prefix, false, "Enable the endpoint id with zookeeper name prefix for the replicated merge tree table", 0) \ - M(UInt64, zero_copy_merge_mutation_min_parts_size_sleep_before_lock, 1ULL * 1024 * 1024 * 1024, "If zero copy replication is enabled sleep random amount of time before trying to lock depending on parts size for merge or mutation", 0) \ - M(Bool, allow_floating_point_partition_key, false, "Allow floating point as partition key", 0) \ - M(UInt64, sleep_before_loading_outdated_parts_ms, 0, "For testing. Do not change it.", 0) \ - M(Bool, always_use_copy_instead_of_hardlinks, false, "Always copy data instead of hardlinking during mutations/replaces/detaches and so on.", 0) \ - M(Bool, disable_freeze_partition_for_zero_copy_replication, true, "Disable FREEZE PARTITION query for zero copy replication.", 0) \ - M(Bool, disable_detach_partition_for_zero_copy_replication, true, "Disable DETACH PARTITION query for zero copy replication.", 0) \ - M(Bool, disable_fetch_partition_for_zero_copy_replication, true, "Disable FETCH PARTITION query for zero copy replication.", 0) \ - M(Bool, enable_block_number_column, false, "Enable persisting column _block_number for each row.", 0) ALIAS(allow_experimental_block_number_column) \ - M(Bool, enable_block_offset_column, false, "Enable persisting column _block_offset for each row.", 0) \ + DECLARE(Bool, allow_suspicious_indices, false, "Reject primary/secondary indexes and sorting keys with identical expressions", 0) \ + DECLARE(Bool, compatibility_allow_sampling_expression_not_in_primary_key, false, "Allow to create a table with sampling expression not in primary key. This is needed only to temporarily allow to run the server with wrong tables for backward compatibility.", 0) \ + DECLARE(Bool, use_minimalistic_checksums_in_zookeeper, true, "Use small format (dozens bytes) for part checksums in ZooKeeper instead of ordinary ones (dozens KB). Before enabling check that all replicas support new format.", 0) \ + DECLARE(Bool, use_minimalistic_part_header_in_zookeeper, true, "Store part header (checksums and columns) in a compact format and a single part znode instead of separate znodes (/columns and /checksums). This can dramatically reduce snapshot size in ZooKeeper. Before enabling check that all replicas support new format.", 0) \ + DECLARE(UInt64, finished_mutations_to_keep, 100, "How many records about mutations that are done to keep. If zero, then keep all of them.", 0) \ + DECLARE(UInt64, min_merge_bytes_to_use_direct_io, 10ULL * 1024 * 1024 * 1024, "Minimal amount of bytes to enable O_DIRECT in merge (0 - disabled).", 0) \ + DECLARE(UInt64, index_granularity_bytes, 10 * 1024 * 1024, "Approximate amount of bytes in single granule (0 - disabled).", 0) \ + DECLARE(UInt64, min_index_granularity_bytes, 1024, "Minimum amount of bytes in single granule.", 1024) \ + DECLARE(Int64, merge_with_ttl_timeout, 3600 * 4, "Minimal time in seconds, when merge with delete TTL can be repeated.", 0) \ + DECLARE(Int64, merge_with_recompression_ttl_timeout, 3600 * 4, "Minimal time in seconds, when merge with recompression TTL can be repeated.", 0) \ + DECLARE(Bool, ttl_only_drop_parts, false, "Only drop altogether the expired parts and not partially prune them.", 0) \ + DECLARE(Bool, materialize_ttl_recalculate_only, false, "Only recalculate ttl info when MATERIALIZE TTL", 0) \ + DECLARE(Bool, enable_mixed_granularity_parts, true, "Enable parts with adaptive and non adaptive granularity", 0) \ + DECLARE(UInt64, concurrent_part_removal_threshold, 100, "Activate concurrent part removal (see 'max_part_removal_threads') only if the number of inactive data parts is at least this.", 0) \ + DECLARE(UInt64, zero_copy_concurrent_part_removal_max_split_times, 5, "Max recursion depth for splitting independent Outdated parts ranges into smaller subranges (highly not recommended to change)", 0) \ + DECLARE(Float, zero_copy_concurrent_part_removal_max_postpone_ratio, static_cast(0.05), "Max percentage of top level parts to postpone removal in order to get smaller independent ranges (highly not recommended to change)", 0) \ + DECLARE(String, storage_policy, "default", "Name of storage disk policy", 0) \ + DECLARE(String, disk, "", "Name of storage disk. Can be specified instead of storage policy.", 0) \ + DECLARE(Bool, allow_nullable_key, false, "Allow Nullable types as primary keys.", 0) \ + DECLARE(Bool, remove_empty_parts, true, "Remove empty parts after they were pruned by TTL, mutation, or collapsing merge algorithm.", 0) \ + DECLARE(Bool, assign_part_uuids, false, "Generate UUIDs for parts. Before enabling check that all replicas support new format.", 0) \ + DECLARE(Int64, max_partitions_to_read, -1, "Limit the max number of partitions that can be accessed in one query. <= 0 means unlimited. This setting is the default that can be overridden by the query-level setting with the same name.", 0) \ + DECLARE(UInt64, max_concurrent_queries, 0, "Max number of concurrently executed queries related to the MergeTree table (0 - disabled). Queries will still be limited by other max_concurrent_queries settings.", 0) \ + DECLARE(UInt64, min_marks_to_honor_max_concurrent_queries, 0, "Minimal number of marks to honor the MergeTree-level's max_concurrent_queries (0 - disabled). Queries will still be limited by other max_concurrent_queries settings.", 0) \ + DECLARE(UInt64, min_bytes_to_rebalance_partition_over_jbod, 0, "Minimal amount of bytes to enable part rebalance over JBOD array (0 - disabled).", 0) \ + DECLARE(Bool, check_sample_column_is_correct, true, "Check columns or columns by hash for sampling are unsigned integer.", 0) \ + DECLARE(Bool, allow_vertical_merges_from_compact_to_wide_parts, true, "Allows vertical merges from compact to wide parts. This settings must have the same value on all replicas", 0) \ + DECLARE(Bool, enable_the_endpoint_id_with_zookeeper_name_prefix, false, "Enable the endpoint id with zookeeper name prefix for the replicated merge tree table", 0) \ + DECLARE(UInt64, zero_copy_merge_mutation_min_parts_size_sleep_before_lock, 1ULL * 1024 * 1024 * 1024, "If zero copy replication is enabled sleep random amount of time before trying to lock depending on parts size for merge or mutation", 0) \ + DECLARE(Bool, allow_floating_point_partition_key, false, "Allow floating point as partition key", 0) \ + DECLARE(UInt64, sleep_before_loading_outdated_parts_ms, 0, "For testing. Do not change it.", 0) \ + DECLARE(Bool, always_use_copy_instead_of_hardlinks, false, "Always copy data instead of hardlinking during mutations/replaces/detaches and so on.", 0) \ + DECLARE(Bool, disable_freeze_partition_for_zero_copy_replication, true, "Disable FREEZE PARTITION query for zero copy replication.", 0) \ + DECLARE(Bool, disable_detach_partition_for_zero_copy_replication, true, "Disable DETACH PARTITION query for zero copy replication.", 0) \ + DECLARE(Bool, disable_fetch_partition_for_zero_copy_replication, true, "Disable FETCH PARTITION query for zero copy replication.", 0) \ + DECLARE(Bool, enable_block_number_column, false, "Enable persisting column _block_number for each row.", 0) ALIAS(allow_experimental_block_number_column) \ + DECLARE(Bool, enable_block_offset_column, false, "Enable persisting column _block_offset for each row.", 0) \ \ /** Experimental/work in progress feature. Unsafe for production. */ \ - M(UInt64, part_moves_between_shards_enable, 0, "Experimental/Incomplete feature to move parts between shards. Does not take into account sharding expressions.", 0) \ - M(UInt64, part_moves_between_shards_delay_seconds, 30, "Time to wait before/after moving parts between shards.", 0) \ - M(Bool, allow_remote_fs_zero_copy_replication, false, "Don't use this setting in production, because it is not ready.", 0) \ - M(String, remote_fs_zero_copy_zookeeper_path, "/clickhouse/zero_copy", "ZooKeeper path for zero-copy table-independent info.", 0) \ - M(Bool, remote_fs_zero_copy_path_compatible_mode, false, "Run zero-copy in compatible mode during conversion process.", 0) \ - M(Bool, cache_populated_by_fetch, false, "Only available in ClickHouse Cloud", 0) \ - M(Bool, force_read_through_cache_for_merges, false, "Force read-through filesystem cache for merges", 0) \ - M(Bool, allow_experimental_replacing_merge_with_cleanup, false, "Allow experimental CLEANUP merges for ReplacingMergeTree with is_deleted column.", 0) \ + DECLARE(UInt64, part_moves_between_shards_enable, 0, "Experimental/Incomplete feature to move parts between shards. Does not take into account sharding expressions.", EXPERIMENTAL) \ + DECLARE(UInt64, part_moves_between_shards_delay_seconds, 30, "Time to wait before/after moving parts between shards.", EXPERIMENTAL) \ + DECLARE(Bool, allow_remote_fs_zero_copy_replication, false, "Don't use this setting in production, because it is not ready.", BETA) \ + DECLARE(String, remote_fs_zero_copy_zookeeper_path, "/clickhouse/zero_copy", "ZooKeeper path for zero-copy table-independent info.", EXPERIMENTAL) \ + DECLARE(Bool, remote_fs_zero_copy_path_compatible_mode, false, "Run zero-copy in compatible mode during conversion process.", EXPERIMENTAL) \ + DECLARE(Bool, cache_populated_by_fetch, false, "Only available in ClickHouse Cloud", EXPERIMENTAL) \ + DECLARE(Bool, force_read_through_cache_for_merges, false, "Force read-through filesystem cache for merges", EXPERIMENTAL) \ + DECLARE(Bool, allow_experimental_replacing_merge_with_cleanup, false, "Allow experimental CLEANUP merges for ReplacingMergeTree with is_deleted column.", EXPERIMENTAL) \ \ /** Compress marks and primary key. */ \ - M(Bool, compress_marks, true, "Marks support compression, reduce mark file size and speed up network transmission.", 0) \ - M(Bool, compress_primary_key, true, "Primary key support compression, reduce primary key file size and speed up network transmission.", 0) \ - M(String, marks_compression_codec, "ZSTD(3)", "Compression encoding used by marks, marks are small enough and cached, so the default compression is ZSTD(3).", 0) \ - M(String, primary_key_compression_codec, "ZSTD(3)", "Compression encoding used by primary, primary key is small enough and cached, so the default compression is ZSTD(3).", 0) \ - M(UInt64, marks_compress_block_size, 65536, "Mark compress block size, the actual size of the block to compress.", 0) \ - M(UInt64, primary_key_compress_block_size, 65536, "Primary compress block size, the actual size of the block to compress.", 0) \ - M(Bool, primary_key_lazy_load, true, "Load primary key in memory on first use instead of on table initialization. This can save memory in the presence of a large number of tables.", 0) \ - M(Float, primary_key_ratio_of_unique_prefix_values_to_skip_suffix_columns, 0.9f, "If the value of a column of the primary key in data part changes at least in this ratio of times, skip loading next columns in memory. This allows to save memory usage by not loading useless columns of the primary key.", 0) \ + DECLARE(Bool, compress_marks, true, "Marks support compression, reduce mark file size and speed up network transmission.", 0) \ + DECLARE(Bool, compress_primary_key, true, "Primary key support compression, reduce primary key file size and speed up network transmission.", 0) \ + DECLARE(String, marks_compression_codec, "ZSTD(3)", "Compression encoding used by marks, marks are small enough and cached, so the default compression is ZSTD(3).", 0) \ + DECLARE(String, primary_key_compression_codec, "ZSTD(3)", "Compression encoding used by primary, primary key is small enough and cached, so the default compression is ZSTD(3).", 0) \ + DECLARE(UInt64, marks_compress_block_size, 65536, "Mark compress block size, the actual size of the block to compress.", 0) \ + DECLARE(UInt64, primary_key_compress_block_size, 65536, "Primary compress block size, the actual size of the block to compress.", 0) \ + DECLARE(Bool, primary_key_lazy_load, true, "Load primary key in memory on first use instead of on table initialization. This can save memory in the presence of a large number of tables.", 0) \ + DECLARE(Float, primary_key_ratio_of_unique_prefix_values_to_skip_suffix_columns, 0.9f, "If the value of a column of the primary key in data part changes at least in this ratio of times, skip loading next columns in memory. This allows to save memory usage by not loading useless columns of the primary key.", 0) \ + DECLARE(Bool, prewarm_mark_cache, false, "If true mark cache will be prewarmed by saving marks to mark cache on inserts, merges, fetches and on startup of server", 0) \ + DECLARE(String, columns_to_prewarm_mark_cache, "", "List of columns to prewarm mark cache for (if enabled). Empty means all columns", 0) \ /** Projection settings. */ \ - M(UInt64, max_projections, 25, "The maximum number of merge tree projections.", 0) \ - M(LightweightMutationProjectionMode, lightweight_mutation_projection_mode, LightweightMutationProjectionMode::THROW, "When lightweight delete happens on a table with projection(s), the possible operations include throw the exception as projection exists, or drop projections of this table's relevant parts, or rebuild the projections.", 0) \ - M(DeduplicateMergeProjectionMode, deduplicate_merge_projection_mode, DeduplicateMergeProjectionMode::THROW, "Whether to allow create projection for the table with non-classic MergeTree. Ignore option is purely for compatibility which might result in incorrect answer. Otherwise, if allowed, what is the action when merge, drop or rebuild.", 0) \ + DECLARE(UInt64, max_projections, 25, "The maximum number of merge tree projections.", 0) \ + DECLARE(LightweightMutationProjectionMode, lightweight_mutation_projection_mode, LightweightMutationProjectionMode::THROW, "When lightweight delete happens on a table with projection(s), the possible operations include throw the exception as projection exists, or drop projections of this table's relevant parts, or rebuild the projections.", 0) \ + DECLARE(DeduplicateMergeProjectionMode, deduplicate_merge_projection_mode, DeduplicateMergeProjectionMode::THROW, "Whether to allow create projection for the table with non-classic MergeTree. Ignore option is purely for compatibility which might result in incorrect answer. Otherwise, if allowed, what is the action when merge, drop or rebuild.", 0) \ #define MAKE_OBSOLETE_MERGE_TREE_SETTING(M, TYPE, NAME, DEFAULT) \ - M(TYPE, NAME, DEFAULT, "Obsolete setting, does nothing.", BaseSettingsHelpers::Flags::OBSOLETE) + M(TYPE, NAME, DEFAULT, "Obsolete setting, does nothing.", SettingsTierType::OBSOLETE) #define OBSOLETE_MERGE_TREE_SETTINGS(M, ALIAS) \ /** Obsolete settings that do nothing but left for compatibility reasons. */ \ @@ -278,8 +281,9 @@ namespace ErrorCodes MERGE_TREE_SETTINGS(M, ALIAS) \ OBSOLETE_MERGE_TREE_SETTINGS(M, ALIAS) -DECLARE_SETTINGS_TRAITS(MergeTreeSettingsTraits, LIST_OF_MERGE_TREE_SETTINGS) +// clang-format on +DECLARE_SETTINGS_TRAITS(MergeTreeSettingsTraits, LIST_OF_MERGE_TREE_SETTINGS) /** Settings for the MergeTree family of engines. * Could be loaded from config or from a CREATE TABLE query (SETTINGS clause). @@ -333,7 +337,7 @@ void MergeTreeSettingsImpl::loadFromQuery(ASTStorage & storage_def, ContextPtr c else if (name == "storage_policy") found_storage_policy_setting = true; - if (found_disk_setting && found_storage_policy_setting) + if (!is_attach && found_disk_setting && found_storage_policy_setting) { throw Exception( ErrorCodes::BAD_ARGUMENTS, @@ -489,8 +493,7 @@ void MergeTreeColumnSettings::validate(const SettingsChanges & changes) } } -#define INITIALIZE_SETTING_EXTERN(TYPE, NAME, DEFAULT, DESCRIPTION, FLAGS) \ - MergeTreeSettings ## TYPE NAME = & MergeTreeSettings ## Impl :: NAME; +#define INITIALIZE_SETTING_EXTERN(TYPE, NAME, DEFAULT, DESCRIPTION, FLAGS) MergeTreeSettings##TYPE NAME = &MergeTreeSettingsImpl ::NAME; namespace MergeTreeSetting { @@ -514,18 +517,7 @@ MergeTreeSettings::MergeTreeSettings(MergeTreeSettings && settings) noexcept MergeTreeSettings::~MergeTreeSettings() = default; -#define IMPLEMENT_SETTING_SUBSCRIPT_OPERATOR(CLASS_NAME, TYPE) \ - const SettingField##TYPE & MergeTreeSettings::operator[](CLASS_NAME##TYPE t) const \ - { \ - return impl.get()->*t; \ - } \ -SettingField##TYPE & MergeTreeSettings::operator[](CLASS_NAME##TYPE t) \ - { \ - return impl.get()->*t; \ - } - MERGETREE_SETTINGS_SUPPORTED_TYPES(MergeTreeSettings, IMPLEMENT_SETTING_SUBSCRIPT_OPERATOR) -#undef IMPLEMENT_SETTING_SUBSCRIPT_OPERATOR bool MergeTreeSettings::has(std::string_view name) const { @@ -662,7 +654,8 @@ void MergeTreeSettings::dumpToSystemMergeTreeSettingsColumns(MutableColumnsAndCo res_columns[5]->insert(max); res_columns[6]->insert(writability == SettingConstraintWritability::CONST); res_columns[7]->insert(setting.getTypeName()); - res_columns[8]->insert(setting.isObsolete()); + res_columns[8]->insert(setting.getTier() == SettingsTierType::OBSOLETE); + res_columns[9]->insert(setting.getTier()); } } diff --git a/src/Storages/MergeTree/MergeTreeSink.cpp b/src/Storages/MergeTree/MergeTreeSink.cpp index 1e42f16736d..99852309c77 100644 --- a/src/Storages/MergeTree/MergeTreeSink.cpp +++ b/src/Storages/MergeTree/MergeTreeSink.cpp @@ -94,7 +94,7 @@ void MergeTreeSink::consume(Chunk & chunk) DelayedPartitions partitions; const Settings & settings = context->getSettingsRef(); - size_t streams = 0; + size_t total_streams = 0; bool support_parallel_write = false; auto token_info = chunk.getChunkInfos().get(); @@ -153,16 +153,18 @@ void MergeTreeSink::consume(Chunk & chunk) max_insert_delayed_streams_for_parallel_write = 0; /// In case of too much columns/parts in block, flush explicitly. - streams += temp_part.streams.size(); + size_t current_streams = 0; + for (const auto & stream : temp_part.streams) + current_streams += stream.stream->getNumberOfOpenStreams(); - if (streams > max_insert_delayed_streams_for_parallel_write) + if (total_streams + current_streams > max_insert_delayed_streams_for_parallel_write) { finishDelayedChunk(); delayed_chunk = std::make_unique(); delayed_chunk->partitions = std::move(partitions); finishDelayedChunk(); - streams = 0; + total_streams = 0; support_parallel_write = false; partitions = DelayedPartitions{}; } @@ -174,6 +176,8 @@ void MergeTreeSink::consume(Chunk & chunk) .block_dedup_token = block_dedup_token, .part_counters = std::move(part_counters), }); + + total_streams += current_streams; } if (need_to_define_dedup_token) @@ -243,6 +247,15 @@ void MergeTreeSink::finishDelayedChunk() /// Part can be deduplicated, so increment counters and add to part log only if it's really added if (added) { + if (auto * mark_cache = storage.getContext()->getMarkCache().get()) + { + for (const auto & stream : partition.temp_part.streams) + { + auto marks = stream.stream->releaseCachedMarks(); + addMarksToCache(*part, marks, mark_cache); + } + } + auto counters_snapshot = std::make_shared(partition.part_counters.getPartiallyAtomicSnapshot()); PartLog::addNewPart(storage.getContext(), PartLog::PartLogEntry(part, partition.elapsed_ns, counters_snapshot)); StorageMergeTree::incrementInsertedPartsProfileEvent(part->getType()); diff --git a/src/Storages/MergeTree/MergedBlockOutputStream.cpp b/src/Storages/MergeTree/MergedBlockOutputStream.cpp index 4ee68580d3f..77c34aae30a 100644 --- a/src/Storages/MergeTree/MergedBlockOutputStream.cpp +++ b/src/Storages/MergeTree/MergedBlockOutputStream.cpp @@ -25,6 +25,7 @@ MergedBlockOutputStream::MergedBlockOutputStream( CompressionCodecPtr default_codec_, TransactionID tid, bool reset_columns_, + bool save_marks_in_cache, bool blocks_are_granules_size, const WriteSettings & write_settings_, const MergeTreeIndexGranularity & computed_index_granularity) @@ -39,6 +40,7 @@ MergedBlockOutputStream::MergedBlockOutputStream( storage_settings, data_part->index_granularity_info.mark_type.adaptive, /* rewrite_primary_key = */ true, + save_marks_in_cache, blocks_are_granules_size); /// TODO: looks like isStoredOnDisk() is always true for MergeTreeDataPart diff --git a/src/Storages/MergeTree/MergedBlockOutputStream.h b/src/Storages/MergeTree/MergedBlockOutputStream.h index e212fe5bb5a..060778866e0 100644 --- a/src/Storages/MergeTree/MergedBlockOutputStream.h +++ b/src/Storages/MergeTree/MergedBlockOutputStream.h @@ -24,6 +24,7 @@ public: CompressionCodecPtr default_codec_, TransactionID tid, bool reset_columns_ = false, + bool save_marks_in_cache = false, bool blocks_are_granules_size = false, const WriteSettings & write_settings = {}, const MergeTreeIndexGranularity & computed_index_granularity = {}); diff --git a/src/Storages/MergeTree/MergedColumnOnlyOutputStream.cpp b/src/Storages/MergeTree/MergedColumnOnlyOutputStream.cpp index 05cd77dcd40..bed539dfe02 100644 --- a/src/Storages/MergeTree/MergedColumnOnlyOutputStream.cpp +++ b/src/Storages/MergeTree/MergedColumnOnlyOutputStream.cpp @@ -19,6 +19,7 @@ MergedColumnOnlyOutputStream::MergedColumnOnlyOutputStream( const MergeTreeIndices & indices_to_recalc, const ColumnsStatistics & stats_to_recalc_, WrittenOffsetColumns * offset_columns_, + bool save_marks_in_cache, const MergeTreeIndexGranularity & index_granularity, const MergeTreeIndexGranularityInfo * index_granularity_info) : IMergedBlockOutputStream(data_part->storage.getSettings(), data_part->getDataPartStoragePtr(), metadata_snapshot_, columns_list_, /*reset_columns=*/ true) @@ -30,7 +31,9 @@ MergedColumnOnlyOutputStream::MergedColumnOnlyOutputStream( data_part->storage.getContext()->getWriteSettings(), storage_settings, index_granularity_info ? index_granularity_info->mark_type.adaptive : data_part->storage.canUseAdaptiveGranularity(), - /* rewrite_primary_key = */ false); + /* rewrite_primary_key = */ false, + save_marks_in_cache, + /* blocks_are_granules_size = */ false); writer = createMergeTreeDataPartWriter( data_part->getType(), diff --git a/src/Storages/MergeTree/MergedColumnOnlyOutputStream.h b/src/Storages/MergeTree/MergedColumnOnlyOutputStream.h index e837a62743e..f6bf9e37a58 100644 --- a/src/Storages/MergeTree/MergedColumnOnlyOutputStream.h +++ b/src/Storages/MergeTree/MergedColumnOnlyOutputStream.h @@ -22,6 +22,7 @@ public: const MergeTreeIndices & indices_to_recalc_, const ColumnsStatistics & stats_to_recalc_, WrittenOffsetColumns * offset_columns_ = nullptr, + bool save_marks_in_cache = false, const MergeTreeIndexGranularity & index_granularity = {}, const MergeTreeIndexGranularityInfo * index_granularity_info_ = nullptr); diff --git a/src/Storages/MergeTree/MutateFromLogEntryTask.cpp b/src/Storages/MergeTree/MutateFromLogEntryTask.cpp index 54215cd2dba..6716144ce81 100644 --- a/src/Storages/MergeTree/MutateFromLogEntryTask.cpp +++ b/src/Storages/MergeTree/MutateFromLogEntryTask.cpp @@ -226,6 +226,10 @@ ReplicatedMergeMutateTaskBase::PrepareResult MutateFromLogEntryTask::prepare() future_mutated_part, task_context); + storage.writePartLog( + PartLogElement::MUTATE_PART_START, {}, 0, + entry.new_part_name, new_part, future_mutated_part->parts, merge_mutate_entry.get(), {}); + mutate_task = storage.merger_mutator.mutatePartToTemporaryPart( future_mutated_part, metadata_snapshot, commands, merge_mutate_entry.get(), entry.create_time, task_context, NO_TRANSACTION_PTR, reserved_space, table_lock_holder); diff --git a/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp b/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp index 53aef36404e..fbc20b282ca 100644 --- a/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp +++ b/src/Storages/MergeTree/MutatePlainMergeTreeTask.cpp @@ -39,6 +39,10 @@ void MutatePlainMergeTreeTask::prepare() future_part, task_context); + storage.writePartLog( + PartLogElement::MUTATE_PART_START, {}, 0, + future_part->name, new_part, future_part->parts, merge_list_entry.get(), {}); + stopwatch = std::make_unique(); write_part_log = [this] (const ExecutionStatus & execution_status) diff --git a/src/Storages/MergeTree/MutateTask.cpp b/src/Storages/MergeTree/MutateTask.cpp index 2e7847fc99f..936df7b0275 100644 --- a/src/Storages/MergeTree/MutateTask.cpp +++ b/src/Storages/MergeTree/MutateTask.cpp @@ -1623,6 +1623,7 @@ private: ctx->compression_codec, ctx->txn ? ctx->txn->tid : Tx::PrehistoricTID, /*reset_columns=*/ true, + /*save_marks_in_cache=*/ false, /*blocks_are_granules_size=*/ false, ctx->context->getWriteSettings(), computed_granularity); @@ -1851,6 +1852,7 @@ private: std::vector(ctx->indices_to_recalc.begin(), ctx->indices_to_recalc.end()), ColumnsStatistics(ctx->stats_to_recalc.begin(), ctx->stats_to_recalc.end()), nullptr, + /*save_marks_in_cache=*/ false, ctx->source_part->index_granularity, &ctx->source_part->index_granularity_info ); @@ -2164,6 +2166,7 @@ bool MutateTask::prepare() context_for_reading->setSetting("apply_mutations_on_fly", false); /// Skip using large sets in KeyCondition context_for_reading->setSetting("use_index_for_in_with_subqueries_max_values", 100000); + context_for_reading->setSetting("use_concurrency_control", false); for (const auto & command : *ctx->commands) if (!canSkipMutationCommandForPart(ctx->source_part, command, context_for_reading)) @@ -2286,7 +2289,7 @@ bool MutateTask::prepare() String tmp_part_dir_name = prefix + ctx->future_part->name; ctx->temporary_directory_lock = ctx->data->getTemporaryPartDirectoryHolder(tmp_part_dir_name); - auto builder = ctx->data->getDataPartBuilder(ctx->future_part->name, single_disk_volume, tmp_part_dir_name); + auto builder = ctx->data->getDataPartBuilder(ctx->future_part->name, single_disk_volume, tmp_part_dir_name, getReadSettings()); builder.withPartFormat(ctx->future_part->part_format); builder.withPartInfo(ctx->future_part->part_info); diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.cpp index 22b8ccca151..c258048354e 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.cpp @@ -1,5 +1,6 @@ #include #include +#include #include #include @@ -20,7 +21,6 @@ namespace ErrorCodes { extern const int SUPPORT_IS_DISABLED; extern const int REPLICA_STATUS_CHANGED; - extern const int LOGICAL_ERROR; } ReplicatedMergeTreeAttachThread::ReplicatedMergeTreeAttachThread(StorageReplicatedMergeTree & storage_) @@ -123,67 +123,6 @@ void ReplicatedMergeTreeAttachThread::checkHasReplicaMetadataInZooKeeper(const z } } -Int32 ReplicatedMergeTreeAttachThread::fixReplicaMetadataVersionIfNeeded(zkutil::ZooKeeperPtr zookeeper) -{ - const String & zookeeper_path = storage.zookeeper_path; - const String & replica_path = storage.replica_path; - const bool replica_readonly = storage.is_readonly; - - for (size_t i = 0; i != 2; ++i) - { - String replica_metadata_version_str; - const bool replica_metadata_version_exists = zookeeper->tryGet(replica_path + "/metadata_version", replica_metadata_version_str); - if (!replica_metadata_version_exists) - return -1; - - const Int32 metadata_version = parse(replica_metadata_version_str); - - if (metadata_version != 0 || replica_readonly) - { - /// No need to fix anything - return metadata_version; - } - - Coordination::Stat stat; - zookeeper->get(fs::path(zookeeper_path) / "metadata", &stat); - if (stat.version == 0) - { - /// No need to fix anything - return metadata_version; - } - - ReplicatedMergeTreeQueue & queue = storage.queue; - queue.pullLogsToQueue(zookeeper); - if (queue.getStatus().metadata_alters_in_queue != 0) - { - LOG_DEBUG(log, "No need to update metadata_version as there are ALTER_METADATA entries in the queue"); - return metadata_version; - } - - const Coordination::Requests ops = { - zkutil::makeSetRequest(fs::path(replica_path) / "metadata_version", std::to_string(stat.version), 0), - zkutil::makeCheckRequest(fs::path(zookeeper_path) / "metadata", stat.version), - }; - Coordination::Responses ops_responses; - const auto code = zookeeper->tryMulti(ops, ops_responses); - if (code == Coordination::Error::ZOK) - { - LOG_DEBUG(log, "Successfully set metadata_version to {}", stat.version); - return stat.version; - } - if (code != Coordination::Error::ZBADVERSION) - { - throw zkutil::KeeperException(code); - } - } - - /// Second attempt is only possible if metadata_version != 0 or metadata.version changed during the first attempt. - /// If metadata_version != 0, on second attempt we will return the new metadata_version. - /// If metadata.version changed, on second attempt we will either get metadata_version != 0 and return the new metadata_version or we will get metadata_alters_in_queue != 0 and return 0. - /// Either way, on second attempt this method should return. - throw Exception(ErrorCodes::LOGICAL_ERROR, "Failed to fix replica metadata_version in ZooKeeper after two attempts"); -} - void ReplicatedMergeTreeAttachThread::runImpl() { storage.setZooKeeper(); @@ -227,33 +166,6 @@ void ReplicatedMergeTreeAttachThread::runImpl() /// Just in case it was not removed earlier due to connection loss zookeeper->tryRemove(replica_path + "/flags/force_restore_data"); - const Int32 replica_metadata_version = fixReplicaMetadataVersionIfNeeded(zookeeper); - const bool replica_metadata_version_exists = replica_metadata_version != -1; - if (replica_metadata_version_exists) - { - storage.setInMemoryMetadata(metadata_snapshot->withMetadataVersion(replica_metadata_version)); - } - else - { - /// Table was created before 20.4 and was never altered, - /// let's initialize replica metadata version from global metadata version. - Coordination::Stat table_metadata_version_stat; - zookeeper->get(zookeeper_path + "/metadata", &table_metadata_version_stat); - - Coordination::Requests ops; - ops.emplace_back(zkutil::makeCheckRequest(zookeeper_path + "/metadata", table_metadata_version_stat.version)); - ops.emplace_back(zkutil::makeCreateRequest(replica_path + "/metadata_version", toString(table_metadata_version_stat.version), zkutil::CreateMode::Persistent)); - - Coordination::Responses res; - auto code = zookeeper->tryMulti(ops, res); - - if (code == Coordination::Error::ZBADVERSION) - throw Exception(ErrorCodes::REPLICA_STATUS_CHANGED, "Failed to initialize metadata_version " - "because table was concurrently altered, will retry"); - - zkutil::KeeperMultiException::check(code, ops, res); - } - storage.checkTableStructure(replica_path, metadata_snapshot); storage.checkParts(skip_sanity_checks); diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.h b/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.h index bfc97442598..250a5ed34d1 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.h +++ b/src/Storages/MergeTree/ReplicatedMergeTreeAttachThread.h @@ -48,8 +48,6 @@ private: void runImpl(); void finalizeInitialization(); - - Int32 fixReplicaMetadataVersionIfNeeded(zkutil::ZooKeeperPtr zookeeper); }; } diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp index 6b1581645f8..b1564b58a6c 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeQueue.cpp @@ -615,7 +615,7 @@ std::pair ReplicatedMergeTreeQueue::pullLogsToQueue(zkutil::Zo { std::lock_guard lock(pull_logs_to_queue_mutex); - if (reason != LOAD) + if (reason != LOAD && reason != FIX_METADATA_VERSION) { /// It's totally ok to load queue on readonly replica (that's what RestartingThread does on initialization). /// It's ok if replica became readonly due to connection loss after we got current zookeeper (in this case zookeeper must be expired). diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeQueue.h b/src/Storages/MergeTree/ReplicatedMergeTreeQueue.h index 9d3349663e2..6ec8818b0c6 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeQueue.h +++ b/src/Storages/MergeTree/ReplicatedMergeTreeQueue.h @@ -334,6 +334,7 @@ public: UPDATE, MERGE_PREDICATE, SYNC, + FIX_METADATA_VERSION, OTHER, }; diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp index 9d3e26cdc8d..93124e634bd 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp @@ -29,6 +29,8 @@ namespace MergeTreeSetting namespace ErrorCodes { extern const int REPLICA_IS_ALREADY_ACTIVE; + extern const int REPLICA_STATUS_CHANGED; + extern const int LOGICAL_ERROR; } namespace FailPoints @@ -207,6 +209,36 @@ bool ReplicatedMergeTreeRestartingThread::tryStartup() throw; } + const Int32 replica_metadata_version = fixReplicaMetadataVersionIfNeeded(zookeeper); + const bool replica_metadata_version_exists = replica_metadata_version != -1; + if (replica_metadata_version_exists) + { + storage.setInMemoryMetadata(storage.getInMemoryMetadataPtr()->withMetadataVersion(replica_metadata_version)); + } + else + { + /// Table was created before 20.4 and was never altered, + /// let's initialize replica metadata version from global metadata version. + + const String & zookeeper_path = storage.zookeeper_path, & replica_path = storage.replica_path; + + Coordination::Stat table_metadata_version_stat; + zookeeper->get(zookeeper_path + "/metadata", &table_metadata_version_stat); + + Coordination::Requests ops; + ops.emplace_back(zkutil::makeCheckRequest(zookeeper_path + "/metadata", table_metadata_version_stat.version)); + ops.emplace_back(zkutil::makeCreateRequest(replica_path + "/metadata_version", toString(table_metadata_version_stat.version), zkutil::CreateMode::Persistent)); + + Coordination::Responses res; + auto code = zookeeper->tryMulti(ops, res); + + if (code == Coordination::Error::ZBADVERSION) + throw Exception(ErrorCodes::REPLICA_STATUS_CHANGED, "Failed to initialize metadata_version " + "because table was concurrently altered, will retry"); + + zkutil::KeeperMultiException::check(code, ops, res); + } + storage.queue.removeCurrentPartsFromMutations(); storage.last_queue_update_finish_time.store(time(nullptr)); @@ -424,4 +456,64 @@ void ReplicatedMergeTreeRestartingThread::setNotReadonly() storage.readonly_start_time.store(0, std::memory_order_relaxed); } + +Int32 ReplicatedMergeTreeRestartingThread::fixReplicaMetadataVersionIfNeeded(zkutil::ZooKeeperPtr zookeeper) +{ + const String & zookeeper_path = storage.zookeeper_path; + const String & replica_path = storage.replica_path; + + const size_t num_attempts = 2; + for (size_t attempt = 0; attempt != num_attempts; ++attempt) + { + String replica_metadata_version_str; + Coordination::Stat replica_stat; + const bool replica_metadata_version_exists = zookeeper->tryGet(replica_path + "/metadata_version", replica_metadata_version_str, &replica_stat); + if (!replica_metadata_version_exists) + return -1; + + const Int32 metadata_version = parse(replica_metadata_version_str); + if (metadata_version != 0) + return metadata_version; + + Coordination::Stat table_stat; + zookeeper->get(fs::path(zookeeper_path) / "metadata", &table_stat); + if (table_stat.version == 0) + return metadata_version; + + ReplicatedMergeTreeQueue & queue = storage.queue; + queue.pullLogsToQueue(zookeeper, {}, ReplicatedMergeTreeQueue::FIX_METADATA_VERSION); + if (queue.getStatus().metadata_alters_in_queue != 0) + { + LOG_INFO(log, "Skipping updating metadata_version as there are ALTER_METADATA entries in the queue"); + return metadata_version; + } + + const Coordination::Requests ops = { + zkutil::makeSetRequest(fs::path(replica_path) / "metadata_version", std::to_string(table_stat.version), replica_stat.version), + zkutil::makeCheckRequest(fs::path(zookeeper_path) / "metadata", table_stat.version), + }; + Coordination::Responses ops_responses; + const Coordination::Error code = zookeeper->tryMulti(ops, ops_responses); + if (code == Coordination::Error::ZOK) + { + LOG_DEBUG(log, "Successfully set metadata_version to {}", table_stat.version); + return table_stat.version; + } + + if (code == Coordination::Error::ZBADVERSION) + { + LOG_WARNING(log, "Cannot fix metadata_version because either metadata.version or metadata_version.version changed, attempts left = {}", num_attempts - attempt - 1); + continue; + } + + throw zkutil::KeeperException(code); + } + + /// Second attempt is only possible if either metadata_version.version or metadata.version changed during the first attempt. + /// If metadata_version changed to non-zero value during the first attempt, on second attempt we will return the new metadata_version. + /// If metadata.version changed during first attempt, on second attempt we will either get metadata_version != 0 and return the new metadata_version or we will get metadata_alters_in_queue != 0 and return 0. + /// So either first or second attempt should return unless metadata_version was rewritten from 0 to 0 during the first attempt which is highly unlikely. + throw Exception(ErrorCodes::LOGICAL_ERROR, "Failed to fix replica metadata_version in ZooKeeper after two attempts"); +} + } diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.h b/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.h index d719505ae5e..6f450dc1d40 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.h +++ b/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.h @@ -6,6 +6,7 @@ #include #include #include +#include namespace DB @@ -68,6 +69,9 @@ private: /// Disable readonly mode for table void setNotReadonly(); + + /// Fix replica metadata_version if needed + Int32 fixReplicaMetadataVersionIfNeeded(zkutil::ZooKeeperPtr zookeeper); }; diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp index 95469337f8a..f3ae6e77ac3 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeSink.cpp @@ -3,8 +3,8 @@ #include #include #include -#include #include +#include #include #include #include @@ -341,7 +341,7 @@ void ReplicatedMergeTreeSinkImpl::consume(Chunk & chunk) using DelayedPartitions = std::vector; DelayedPartitions partitions; - size_t streams = 0; + size_t total_streams = 0; bool support_parallel_write = false; for (auto & current_block : part_blocks) @@ -418,15 +418,18 @@ void ReplicatedMergeTreeSinkImpl::consume(Chunk & chunk) max_insert_delayed_streams_for_parallel_write = 0; /// In case of too much columns/parts in block, flush explicitly. - streams += temp_part.streams.size(); - if (streams > max_insert_delayed_streams_for_parallel_write) + size_t current_streams = 0; + for (const auto & stream : temp_part.streams) + current_streams += stream.stream->getNumberOfOpenStreams(); + + if (total_streams + current_streams > max_insert_delayed_streams_for_parallel_write) { finishDelayedChunk(zookeeper); delayed_chunk = std::make_unique::DelayedChunk>(replicas_num); delayed_chunk->partitions = std::move(partitions); finishDelayedChunk(zookeeper); - streams = 0; + total_streams = 0; support_parallel_write = false; partitions = DelayedPartitions{}; } @@ -447,6 +450,8 @@ void ReplicatedMergeTreeSinkImpl::consume(Chunk & chunk) std::move(unmerged_block), std::move(part_counters) /// profile_events_scope must be reset here. )); + + total_streams += current_streams; } if (need_to_define_dedup_token) @@ -481,6 +486,17 @@ void ReplicatedMergeTreeSinkImpl::finishDelayedChunk(const ZooKeeperWithF /// Set a special error code if the block is duplicate int error = (deduplicate && deduplicated) ? ErrorCodes::INSERT_WAS_DEDUPLICATED : 0; + auto * mark_cache = storage.getContext()->getMarkCache().get(); + + if (!error && mark_cache) + { + for (const auto & stream : partition.temp_part.streams) + { + auto marks = stream.stream->releaseCachedMarks(); + addMarksToCache(*part, marks, mark_cache); + } + } + auto counters_snapshot = std::make_shared(partition.part_counters.getPartiallyAtomicSnapshot()); PartLog::addNewPart(storage.getContext(), PartLog::PartLogEntry(part, partition.elapsed_ns, counters_snapshot), ExecutionStatus(error)); StorageReplicatedMergeTree::incrementInsertedPartsProfileEvent(part->getType()); @@ -521,8 +537,18 @@ void ReplicatedMergeTreeSinkImpl::finishDelayedChunk(const ZooKeeperWithFa { partition.temp_part.finalize(); auto conflict_block_ids = commitPart(zookeeper, partition.temp_part.part, partition.block_id, delayed_chunk->replicas_num).first; + if (conflict_block_ids.empty()) { + if (auto * mark_cache = storage.getContext()->getMarkCache().get()) + { + for (const auto & stream : partition.temp_part.streams) + { + auto marks = stream.stream->releaseCachedMarks(); + addMarksToCache(*partition.temp_part.part, marks, mark_cache); + } + } + auto counters_snapshot = std::make_shared(partition.part_counters.getPartiallyAtomicSnapshot()); PartLog::addNewPart( storage.getContext(), diff --git a/src/Storages/MergeTree/checkDataPart.cpp b/src/Storages/MergeTree/checkDataPart.cpp index 2a1ddf32431..34e699bcef7 100644 --- a/src/Storages/MergeTree/checkDataPart.cpp +++ b/src/Storages/MergeTree/checkDataPart.cpp @@ -135,7 +135,6 @@ bool isRetryableException(std::exception_ptr exception_ptr) } } - static IMergeTreeDataPart::Checksums checkDataPart( MergeTreeData::DataPartPtr data_part, const IDataPartStorage & data_part_storage, @@ -422,6 +421,7 @@ IMergeTreeDataPart::Checksums checkDataPart( } ReadSettings read_settings; + read_settings.read_through_distributed_cache = false; read_settings.enable_filesystem_cache = false; read_settings.enable_filesystem_cache_log = false; read_settings.enable_filesystem_read_prefetches_log = false; From 863887cca5674088ebe15a95d07b2ee0aebf4597 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 8 Nov 2024 00:50:47 +0100 Subject: [PATCH 544/680] Reset WriteSettings to master --- src/IO/WriteSettings.h | 3 --- 1 file changed, 3 deletions(-) diff --git a/src/IO/WriteSettings.h b/src/IO/WriteSettings.h index 4eeb01b5acc..94410f787f0 100644 --- a/src/IO/WriteSettings.h +++ b/src/IO/WriteSettings.h @@ -4,7 +4,6 @@ #include #include - namespace DB { @@ -29,8 +28,6 @@ struct WriteSettings bool use_adaptive_write_buffer = false; size_t adaptive_write_buffer_initial_size = 16 * 1024; - size_t max_compression_threads = 1; - bool write_through_distributed_cache = false; DistributedCacheSettings distributed_cache_settings; From c78272871f26d541800bb783d886325efa3c3ee7 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 8 Nov 2024 00:51:32 +0100 Subject: [PATCH 545/680] Rollback some changes --- .../0_stateless/03254_parallel_compression.sql | 11 ----------- 1 file changed, 11 deletions(-) delete mode 100644 tests/queries/0_stateless/03254_parallel_compression.sql diff --git a/tests/queries/0_stateless/03254_parallel_compression.sql b/tests/queries/0_stateless/03254_parallel_compression.sql deleted file mode 100644 index a17deed7d8c..00000000000 --- a/tests/queries/0_stateless/03254_parallel_compression.sql +++ /dev/null @@ -1,11 +0,0 @@ -DROP TABLE IF EXISTS test2; - -CREATE TABLE test2 -( - k UInt64 -) ENGINE = MergeTree ORDER BY k SETTINGS min_compress_block_size = 10240, min_bytes_for_wide_part = 1, max_compression_threads = 64; - -INSERT INTO test2 SELECT number FROM numbers(20000); -SELECT sum(k) = (9999 * 10000 / 2 + 10000 * 9999) FROM test2 WHERE k > 10000; - -DROP TABLE test2; From 3fa72482a75551c3132fa2f7ee8770d4f6862f98 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Fri, 8 Nov 2024 00:56:38 +0100 Subject: [PATCH 546/680] Revert some changes --- src/Compression/ParallelCompressedWriteBuffer.cpp | 4 ---- src/Compression/ParallelCompressedWriteBuffer.h | 10 ---------- 2 files changed, 14 deletions(-) diff --git a/src/Compression/ParallelCompressedWriteBuffer.cpp b/src/Compression/ParallelCompressedWriteBuffer.cpp index 303e1ece68a..3831d07e91a 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.cpp +++ b/src/Compression/ParallelCompressedWriteBuffer.cpp @@ -44,8 +44,6 @@ void ParallelCompressedWriteBuffer::nextImpl() /// The buffer will be compressed and processed in the thread. current_buffer->busy = true; current_buffer->sequence_num = current_sequence_num; - current_buffer->out_callback = callback; - callback = {}; ++current_sequence_num; current_buffer->uncompressed_size = offset(); pool.scheduleOrThrowOnError([this, my_current_buffer = current_buffer, thread_group = CurrentThread::getGroup()] @@ -155,8 +153,6 @@ void ParallelCompressedWriteBuffer::compress(Iterator buffer) } std::unique_lock lock(mutex); - if (buffer->out_callback) - buffer->out_callback(); buffer->busy = false; cond.notify_all(); } diff --git a/src/Compression/ParallelCompressedWriteBuffer.h b/src/Compression/ParallelCompressedWriteBuffer.h index 8c5f249b06c..38a3a083e19 100644 --- a/src/Compression/ParallelCompressedWriteBuffer.h +++ b/src/Compression/ParallelCompressedWriteBuffer.h @@ -31,13 +31,6 @@ public: ~ParallelCompressedWriteBuffer() override; - /// This function will be called once after compressing the next data and sending it to the out. - /// It can be used to fill information about marks. - void setCompletionCallback(std::function callback_) - { - callback = callback_; - } - private: void nextImpl() override; void finalizeImpl() override; @@ -61,15 +54,12 @@ private: BufferPair * previous = nullptr; size_t sequence_num = 0; bool busy = false; - std::function out_callback; }; std::mutex mutex; std::condition_variable cond; std::list buffers; - std::function callback; - using Iterator = std::list::iterator; Iterator current_buffer; size_t current_sequence_num = 0; From f24dca21a56f97ce2d422bc3e411868eca5c751c Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Fri, 8 Nov 2024 09:15:15 +0100 Subject: [PATCH 547/680] Implement CLICKHOUSE_RUN_AS_ROOT instead of preser UID/GID --- docker/keeper/entrypoint.sh | 25 +++++++++++++++++-------- docker/server/entrypoint.sh | 25 +++++++++++++++++-------- 2 files changed, 34 insertions(+), 16 deletions(-) diff --git a/docker/keeper/entrypoint.sh b/docker/keeper/entrypoint.sh index 92b91a0f8c3..31e4c8b63da 100644 --- a/docker/keeper/entrypoint.sh +++ b/docker/keeper/entrypoint.sh @@ -5,19 +5,28 @@ set -eo pipefail shopt -s nullglob DO_CHOWN=1 -if [ "${CLICKHOUSE_DO_NOT_CHOWN:-0}" = "1" ]; then +if [[ "${CLICKHOUSE_DO_NOT_CHOWN:-0}" = "1" || "${CLICKHOUSE_RUN_AS_ROOT:=0}" = "1" ]]; then DO_CHOWN=0 fi +# CLICKHOUSE_UID and CLICKHOUSE_GID are kept for backward compatibility, but deprecated +# One must use either "docker run --user" or CLICKHOUSE_RUN_AS_ROOT=1 to run the process as +# FIXME: Remove ALL CLICKHOUSE_UID CLICKHOUSE_GID before 25.3 +if [[ "${CLICKHOUSE_UID:-}" || "${CLICKHOUSE_GID:-}" ]]; then + echo 'WARNING: Support for CLICKHOUSE_UID/CLICKHOUSE_GID will be removed in a couple of releases.' >&2 + echo 'WARNING: Either use a proper "docker run --user=xxx:xxxx" argument instead of CLICKHOUSE_UID/CLICKHOUSE_GID' >&2 + echo 'WARNING: or set "CLICKHOUSE_RUN_AS_ROOT=1" ENV to run the clickhouse-server as root:root' >&2 +fi + # support `docker run --user=xxx:xxxx` -if [ "$(id -u)" = "0" ]; then - # CLICKHOUSE_UID and CLICKHOUSE_GID are kept for backward compatibility - if [[ "${CLICKHOUSE_UID:-}" || "${CLICKHOUSE_GID:-}" ]]; then - echo 'WARNING: consider using a proper "--user=xxx:xxxx" running argument instead of CLICKHOUSE_UID/CLICKHOUSE_GID' >&2 - echo 'Support for CLICKHOUSE_UID/CLICKHOUSE_GID will be removed in a couple of releases' >&2 +if [[ "$(id -u)" = "0" ]]; then + if [[ "$CLICKHOUSE_RUN_AS_ROOT" = 1 ]]; then + USER=0 + GROUP=0 + else + USER="${CLICKHOUSE_UID:-"$(id -u clickhouse)"}" + GROUP="${CLICKHOUSE_GID:-"$(id -g clickhouse)"}" fi - USER="${CLICKHOUSE_UID:-"$(id -u clickhouse)"}" - GROUP="${CLICKHOUSE_GID:-"$(id -g clickhouse)"}" if command -v gosu &> /dev/null; then gosu="gosu $USER:$GROUP" elif command -v su-exec &> /dev/null; then diff --git a/docker/server/entrypoint.sh b/docker/server/entrypoint.sh index 5a91d54d32b..443bcd7a176 100755 --- a/docker/server/entrypoint.sh +++ b/docker/server/entrypoint.sh @@ -4,19 +4,28 @@ set -eo pipefail shopt -s nullglob DO_CHOWN=1 -if [ "${CLICKHOUSE_DO_NOT_CHOWN:-0}" = "1" ]; then +if [[ "${CLICKHOUSE_DO_NOT_CHOWN:-0}" = "1" || "${CLICKHOUSE_RUN_AS_ROOT:=0}" = "1" ]]; then DO_CHOWN=0 fi +# CLICKHOUSE_UID and CLICKHOUSE_GID are kept for backward compatibility, but deprecated +# One must use either "docker run --user" or CLICKHOUSE_RUN_AS_ROOT=1 to run the process as +# FIXME: Remove ALL CLICKHOUSE_UID CLICKHOUSE_GID before 25.3 +if [[ "${CLICKHOUSE_UID:-}" || "${CLICKHOUSE_GID:-}" ]]; then + echo 'WARNING: Support for CLICKHOUSE_UID/CLICKHOUSE_GID will be removed in a couple of releases.' >&2 + echo 'WARNING: Either use a proper "docker run --user=xxx:xxxx" argument instead of CLICKHOUSE_UID/CLICKHOUSE_GID' >&2 + echo 'WARNING: or set "CLICKHOUSE_RUN_AS_ROOT=1" ENV to run the clickhouse-server as root:root' >&2 +fi + # support `docker run --user=xxx:xxxx` -if [ "$(id -u)" = "0" ]; then - # CLICKHOUSE_UID and CLICKHOUSE_GID are kept for backward compatibility - if [[ "${CLICKHOUSE_UID:-}" || "${CLICKHOUSE_GID:-}" ]]; then - echo 'WARNING: consider using a proper "--user=xxx:xxxx" running argument instead of CLICKHOUSE_UID/CLICKHOUSE_GID' >&2 - echo 'Support for CLICKHOUSE_UID/CLICKHOUSE_GID will be removed in a couple of releases' >&2 +if [[ "$(id -u)" = "0" ]]; then + if [[ "$CLICKHOUSE_RUN_AS_ROOT" = 1 ]]; then + USER=0 + GROUP=0 + else + USER="${CLICKHOUSE_UID:-"$(id -u clickhouse)"}" + GROUP="${CLICKHOUSE_GID:-"$(id -g clickhouse)"}" fi - USER="${CLICKHOUSE_UID:-"$(id -u clickhouse)"}" - GROUP="${CLICKHOUSE_GID:-"$(id -g clickhouse)"}" else USER="$(id -u)" GROUP="$(id -g)" From 0f945cadc74aed12e6a1f05d7cde98aa02e369b7 Mon Sep 17 00:00:00 2001 From: Derek Chia Date: Fri, 8 Nov 2024 17:34:53 +0800 Subject: [PATCH 548/680] Update settings.md Remove duplicated `background_pool_size` description --- .../server-configuration-parameters/settings.md | 10 ---------- 1 file changed, 10 deletions(-) diff --git a/docs/en/operations/server-configuration-parameters/settings.md b/docs/en/operations/server-configuration-parameters/settings.md index 02fa5a8ca58..c5f92ccdf68 100644 --- a/docs/en/operations/server-configuration-parameters/settings.md +++ b/docs/en/operations/server-configuration-parameters/settings.md @@ -131,16 +131,6 @@ Type: UInt64 Default: 8 -## background_pool_size - -Sets the number of threads performing background merges and mutations for tables with MergeTree engines. You can only increase the number of threads at runtime. To lower the number of threads you have to restart the server. By adjusting this setting, you manage CPU and disk load. Smaller pool size utilizes less CPU and disk resources, but background processes advance slower which might eventually impact query performance. - -Before changing it, please also take a look at related MergeTree settings, such as `number_of_free_entries_in_pool_to_lower_max_size_of_merge` and `number_of_free_entries_in_pool_to_execute_mutation`. - -Type: UInt64 - -Default: 16 - ## background_schedule_pool_size The maximum number of threads that will be used for constantly executing some lightweight periodic operations for replicated tables, Kafka streaming, and DNS cache updates. From bd875401115fb8116302f446c2dec27835b5e958 Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Fri, 8 Nov 2024 09:45:51 +0000 Subject: [PATCH 549/680] Update tests/queries/0_stateless/03256_invalid_mutation_query.sql MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Co-authored-by: János Benjamin Antal --- .../0_stateless/03256_invalid_mutation_query.sql | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/tests/queries/0_stateless/03256_invalid_mutation_query.sql b/tests/queries/0_stateless/03256_invalid_mutation_query.sql index 2c554cabb9e..9b4e8f9a7ea 100644 --- a/tests/queries/0_stateless/03256_invalid_mutation_query.sql +++ b/tests/queries/0_stateless/03256_invalid_mutation_query.sql @@ -3,11 +3,11 @@ DROP TABLE IF EXISTS t2; CREATE TABLE t (x int) ENGINE = MergeTree() ORDER BY (); -DELETE FROM t WHERE y in (SELECT y FROM t); -- { serverError 47 } -DELETE FROM t WHERE x in (SELECT y FROM t); -- { serverError 47 } -DELETE FROM t WHERE x IN (SELECT * FROM t2); -- { serverError 60 } -ALTER TABLE t DELETE WHERE x in (SELECT y FROM t); -- { serverError 47 } -ALTER TABLE t UPDATE x = 1 WHERE x IN (SELECT y FROM t); -- { serverError 47 } +DELETE FROM t WHERE y in (SELECT x FROM t); -- { serverError UNKNOWN_IDENTIFIER } +DELETE FROM t WHERE x in (SELECT y FROM t); -- { serverError UNKNOWN_IDENTIFIER } +DELETE FROM t WHERE x IN (SELECT * FROM t2); -- { serverError UNKNOWN_TABLE } +ALTER TABLE t DELETE WHERE x in (SELECT y FROM t); -- { serverError UNKNOWN_IDENTIFIER } +ALTER TABLE t UPDATE x = 1 WHERE x IN (SELECT y FROM t); -- { serverError UNKNOWN_IDENTIFIER } DELETE FROM t WHERE x IN (SELECT foo FROM bar) SETTINGS validate_mutation_query = 0; From 2d70dd11d27837f2d73fa2b2496ac5d17c1c5a67 Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Fri, 8 Nov 2024 09:47:23 +0000 Subject: [PATCH 550/680] Make it work without the new analyzer --- src/Interpreters/MutationsInterpreter.cpp | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/src/Interpreters/MutationsInterpreter.cpp b/src/Interpreters/MutationsInterpreter.cpp index 589791ac871..a35353a6b2a 100644 --- a/src/Interpreters/MutationsInterpreter.cpp +++ b/src/Interpreters/MutationsInterpreter.cpp @@ -1387,9 +1387,17 @@ void MutationsInterpreter::validate() } } + // Make sure the mutation query is valid if (context->getSettingsRef()[Setting::validate_mutation_query]) - // Make sure the mutation query is valid - prepareQueryAffectedQueryTree(commands, source.getStorage(), context); + { + if (context->getSettingsRef()[Setting::allow_experimental_analyzer]) + prepareQueryAffectedQueryTree(commands, source.getStorage(), context); + else + { + ASTPtr select_query = prepareQueryAffectedAST(commands, source.getStorage(), context); + InterpreterSelectQuery(select_query, context, source.getStorage(), metadata_snapshot); + } + } QueryPlan plan; From d75a41f04ccb536b4083034b076e0f6a012e6d06 Mon Sep 17 00:00:00 2001 From: Yarik Briukhovetskyi <114298166+yariks5s@users.noreply.github.com> Date: Fri, 8 Nov 2024 11:24:28 +0100 Subject: [PATCH 551/680] init --- tests/queries/0_stateless/01287_max_execution_speed.sql | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tests/queries/0_stateless/01287_max_execution_speed.sql b/tests/queries/0_stateless/01287_max_execution_speed.sql index 0d132999481..89c3050a256 100644 --- a/tests/queries/0_stateless/01287_max_execution_speed.sql +++ b/tests/queries/0_stateless/01287_max_execution_speed.sql @@ -1,5 +1,8 @@ -- Tags: no-fasttest, no-debug, no-tsan, no-msan, no-asan +SET max_rows_to_read=0; +SET max_bytes_to_read=0; + SET min_execution_speed = 100000000000, timeout_before_checking_execution_speed = 0; SELECT count() FROM system.numbers; -- { serverError TOO_SLOW } SET min_execution_speed = 0; From cf1da69f93c4c8e982b89a73565c16642ab0f18f Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Fri, 8 Nov 2024 11:44:02 +0100 Subject: [PATCH 552/680] Make keeper entrypoint less verbose, like the in the server --- docker/keeper/entrypoint.sh | 1 - 1 file changed, 1 deletion(-) diff --git a/docker/keeper/entrypoint.sh b/docker/keeper/entrypoint.sh index 31e4c8b63da..2b96e4dd655 100644 --- a/docker/keeper/entrypoint.sh +++ b/docker/keeper/entrypoint.sh @@ -1,6 +1,5 @@ #!/bin/bash -set +x set -eo pipefail shopt -s nullglob From 11f3568f5b661330e3fa94fb1515807dd73d7e22 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Fri, 8 Nov 2024 11:45:12 +0100 Subject: [PATCH 553/680] First check the ROOT to assign the env --- docker/keeper/entrypoint.sh | 2 +- docker/server/entrypoint.sh | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/docker/keeper/entrypoint.sh b/docker/keeper/entrypoint.sh index 2b96e4dd655..934605b0b6f 100644 --- a/docker/keeper/entrypoint.sh +++ b/docker/keeper/entrypoint.sh @@ -4,7 +4,7 @@ set -eo pipefail shopt -s nullglob DO_CHOWN=1 -if [[ "${CLICKHOUSE_DO_NOT_CHOWN:-0}" = "1" || "${CLICKHOUSE_RUN_AS_ROOT:=0}" = "1" ]]; then +if [[ "${CLICKHOUSE_RUN_AS_ROOT:=0}" = "1" || "${CLICKHOUSE_DO_NOT_CHOWN:-0}" = "1" ]]; then DO_CHOWN=0 fi diff --git a/docker/server/entrypoint.sh b/docker/server/entrypoint.sh index 443bcd7a176..2f87008f2e5 100755 --- a/docker/server/entrypoint.sh +++ b/docker/server/entrypoint.sh @@ -4,7 +4,7 @@ set -eo pipefail shopt -s nullglob DO_CHOWN=1 -if [[ "${CLICKHOUSE_DO_NOT_CHOWN:-0}" = "1" || "${CLICKHOUSE_RUN_AS_ROOT:=0}" = "1" ]]; then +if [[ "${CLICKHOUSE_RUN_AS_ROOT:=0}" = "1" || "${CLICKHOUSE_DO_NOT_CHOWN:-0}" = "1" ]]; then DO_CHOWN=0 fi From dd1ca389dbc3b3f5c5f456bc0d070a972acca806 Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Fri, 8 Nov 2024 10:45:13 +0000 Subject: [PATCH 554/680] Trying to cast filter column. --- src/Processors/QueryPlan/FilterStep.cpp | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/src/Processors/QueryPlan/FilterStep.cpp b/src/Processors/QueryPlan/FilterStep.cpp index 64c46332c34..7613aac618e 100644 --- a/src/Processors/QueryPlan/FilterStep.cpp +++ b/src/Processors/QueryPlan/FilterStep.cpp @@ -5,6 +5,8 @@ #include #include #include +#include +#include #include #include #include @@ -46,7 +48,19 @@ static ActionsAndName splitSingleAndFilter(ActionsDAG & dag, const ActionsDAG::N auto name = filter_node->result_name; auto split_result = dag.split({filter_node}, true); dag = std::move(split_result.second); - split_result.first.getOutputs().emplace(split_result.first.getOutputs().begin(), split_result.split_nodes_mapping[filter_node]); + + const auto * split_filter_node = split_result.split_nodes_mapping[filter_node]; + auto filter_type = removeLowCardinality(split_filter_node->result_type); + if (!filter_type->onlyNull() && !isUInt8(removeNullable(filter_type))) + { + DataTypePtr cast_type = std::make_shared(); + if (filter_type->isNullable()) + cast_type = std::make_shared(std::move(cast_type)); + + split_result.first.addCast(*split_filter_node, cast_type, {}); + } + + split_result.first.getOutputs().emplace(split_result.first.getOutputs().begin(), split_filter_node); return ActionsAndName{std::move(split_result.first), std::move(name)}; } @@ -168,7 +182,7 @@ void FilterStep::describeActions(FormatSettings & settings) const for (auto & and_atom : and_atoms) { auto expression = std::make_shared(std::move(and_atom.dag)); - settings.out << prefix << "AND column: " << and_atom.name; + settings.out << prefix << "AND column: " << and_atom.name << '\n'; expression->describeActions(settings.out, prefix); } From b370fefb3c8e5583904ab5fe6b21e4ebbb7de5ad Mon Sep 17 00:00:00 2001 From: Christoph Wurm Date: Fri, 8 Nov 2024 10:53:30 +0000 Subject: [PATCH 555/680] Fix test 03173_forbid_qualify --- tests/queries/0_stateless/03173_forbid_qualify.reference | 1 - tests/queries/0_stateless/03173_forbid_qualify.sql | 3 +-- 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/tests/queries/0_stateless/03173_forbid_qualify.reference b/tests/queries/0_stateless/03173_forbid_qualify.reference index c2f595d8c4b..648ff45ff18 100644 --- a/tests/queries/0_stateless/03173_forbid_qualify.reference +++ b/tests/queries/0_stateless/03173_forbid_qualify.reference @@ -1,3 +1,2 @@ 100 49 -100 diff --git a/tests/queries/0_stateless/03173_forbid_qualify.sql b/tests/queries/0_stateless/03173_forbid_qualify.sql index 0a41385c52f..04c65fdab9c 100644 --- a/tests/queries/0_stateless/03173_forbid_qualify.sql +++ b/tests/queries/0_stateless/03173_forbid_qualify.sql @@ -7,5 +7,4 @@ select count() from test_qualify; -- 100 select * from test_qualify qualify row_number() over (order by number) = 50 SETTINGS enable_analyzer = 1; -- 49 select * from test_qualify qualify row_number() over (order by number) = 50 SETTINGS enable_analyzer = 0; -- { serverError NOT_IMPLEMENTED } -delete from test_qualify where number in (select number from test_qualify qualify row_number() over (order by number) = 50); -- { serverError UNFINISHED } -select count() from test_qualify; -- 100 +delete from test_qualify where number in (select number from test_qualify qualify row_number() over (order by number) = 50); -- { serverError NOT_IMPLEMENTED } From 5275c0a8c44fd3cae1d078411efe42e2f34df437 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Fri, 8 Nov 2024 11:53:46 +0100 Subject: [PATCH 556/680] Reverse order on implicit options --- src/Client/ClientBaseOptimizedParts.cpp | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/src/Client/ClientBaseOptimizedParts.cpp b/src/Client/ClientBaseOptimizedParts.cpp index afffe775029..6eaa3708df6 100644 --- a/src/Client/ClientBaseOptimizedParts.cpp +++ b/src/Client/ClientBaseOptimizedParts.cpp @@ -109,8 +109,8 @@ void ClientApplicationBase::parseAndCheckOptions(OptionsDescription & options_de && !op.original_tokens[0].empty() && !op.value.empty()) { /// Two special cases for better usability: - /// - if the option is a filesystem file, then it's likely a queries file (clickhouse repro.sql) /// - if the option contains a whitespace, it might be a query: clickhouse "SELECT 1" + /// - if the option is a filesystem file, then it's likely a queries file (clickhouse repro.sql) /// These are relevant for interactive usage - user-friendly, but questionable in general. /// In case of ambiguity or for scripts, prefer using proper options. @@ -119,10 +119,10 @@ void ClientApplicationBase::parseAndCheckOptions(OptionsDescription & options_de const char * option; std::error_code ec; - if (std::filesystem::is_regular_file(std::filesystem::path{token}, ec)) - option = "queries-file"; - else if (token.contains(' ')) + if (token.contains(' ')) option = "query"; + else if (std::filesystem::is_regular_file(std::filesystem::path{token}, ec)) + option = "queries-file"; else throw Exception(ErrorCodes::BAD_ARGUMENTS, "Positional option `{}` is not supported.", token); From 164e3c26677a209bc7990d326869e71eb3be3bef Mon Sep 17 00:00:00 2001 From: kssenii Date: Fri, 8 Nov 2024 11:54:43 +0100 Subject: [PATCH 557/680] Update settings changes history --- src/Core/SettingsChangesHistory.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 64964f294bd..efa47302343 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -74,6 +74,7 @@ static std::initializer_list Date: Fri, 8 Nov 2024 10:55:52 +0000 Subject: [PATCH 558/680] Better test fix --- tests/queries/0_stateless/03173_forbid_qualify.reference | 1 + tests/queries/0_stateless/03173_forbid_qualify.sql | 3 ++- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/03173_forbid_qualify.reference b/tests/queries/0_stateless/03173_forbid_qualify.reference index 648ff45ff18..c2f595d8c4b 100644 --- a/tests/queries/0_stateless/03173_forbid_qualify.reference +++ b/tests/queries/0_stateless/03173_forbid_qualify.reference @@ -1,2 +1,3 @@ 100 49 +100 diff --git a/tests/queries/0_stateless/03173_forbid_qualify.sql b/tests/queries/0_stateless/03173_forbid_qualify.sql index 04c65fdab9c..f7b05a1eb7e 100644 --- a/tests/queries/0_stateless/03173_forbid_qualify.sql +++ b/tests/queries/0_stateless/03173_forbid_qualify.sql @@ -7,4 +7,5 @@ select count() from test_qualify; -- 100 select * from test_qualify qualify row_number() over (order by number) = 50 SETTINGS enable_analyzer = 1; -- 49 select * from test_qualify qualify row_number() over (order by number) = 50 SETTINGS enable_analyzer = 0; -- { serverError NOT_IMPLEMENTED } -delete from test_qualify where number in (select number from test_qualify qualify row_number() over (order by number) = 50); -- { serverError NOT_IMPLEMENTED } +delete from test_qualify where number in (select number from test_qualify qualify row_number() over (order by number) = 50) SETTINGS validate_mutation_query = 0; -- { serverError UNFINISHED } +select count() from test_qualify; -- 100 From 87b9f5cb4ef65bd8c7313bd4f2563e41b974e951 Mon Sep 17 00:00:00 2001 From: alesapin Date: Fri, 8 Nov 2024 12:24:29 +0100 Subject: [PATCH 559/680] Add min_parts_to_merge_at_once setting --- .../MergeTree/MergeSelectors/SimpleMergeSelector.cpp | 5 ++++- src/Storages/MergeTree/MergeSelectors/SimpleMergeSelector.h | 2 ++ src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp | 2 ++ src/Storages/MergeTree/MergeTreeSettings.cpp | 1 + 4 files changed, 9 insertions(+), 1 deletion(-) diff --git a/src/Storages/MergeTree/MergeSelectors/SimpleMergeSelector.cpp b/src/Storages/MergeTree/MergeSelectors/SimpleMergeSelector.cpp index c393349ef32..4f786215cbe 100644 --- a/src/Storages/MergeTree/MergeSelectors/SimpleMergeSelector.cpp +++ b/src/Storages/MergeTree/MergeSelectors/SimpleMergeSelector.cpp @@ -116,7 +116,7 @@ bool allow( double sum_size, double max_size, double min_age, - double range_size, + size_t range_size, double partition_size, double min_size_to_lower_base_log, double max_size_to_lower_base_log, @@ -125,6 +125,9 @@ bool allow( if (settings.min_age_to_force_merge && min_age >= settings.min_age_to_force_merge) return true; + if (settings.min_parts_to_merge_at_once && range_size < settings.min_parts_to_merge_at_once) + return false; + /// Map size to 0..1 using logarithmic scale /// Use log(1 + x) instead of log1p(x) because our sum_size is always integer. /// Also log1p seems to be slow and significantly affect performance of merges assignment. diff --git a/src/Storages/MergeTree/MergeSelectors/SimpleMergeSelector.h b/src/Storages/MergeTree/MergeSelectors/SimpleMergeSelector.h index 2d4129b8bf8..1e7676c6aed 100644 --- a/src/Storages/MergeTree/MergeSelectors/SimpleMergeSelector.h +++ b/src/Storages/MergeTree/MergeSelectors/SimpleMergeSelector.h @@ -90,6 +90,8 @@ public: { /// Zero means unlimited. Can be overridden by the same merge tree setting. size_t max_parts_to_merge_at_once = 100; + /// Zero means no minimum. Can be overridden by the same merge tree setting. + size_t min_parts_to_merge_at_once = 0; /// Some sort of a maximum number of parts in partition. Can be overridden by the same merge tree setting. size_t parts_to_throw_insert = 3000; diff --git a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp index 37b6539755c..488f4b2390d 100644 --- a/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp +++ b/src/Storages/MergeTree/MergeTreeDataMergerMutator.cpp @@ -82,6 +82,7 @@ namespace MergeTreeSetting extern const MergeTreeSettingsMergeSelectorAlgorithm merge_selector_algorithm; extern const MergeTreeSettingsBool merge_selector_enable_heuristic_to_remove_small_parts_at_right; extern const MergeTreeSettingsFloat merge_selector_base; + extern const MergeTreeSettingsUInt64 min_parts_to_merge_at_once; } namespace ErrorCodes @@ -566,6 +567,7 @@ SelectPartsDecision MergeTreeDataMergerMutator::selectPartsToMergeFromRanges( simple_merge_settings.max_parts_to_merge_at_once = (*data_settings)[MergeTreeSetting::max_parts_to_merge_at_once]; simple_merge_settings.enable_heuristic_to_remove_small_parts_at_right = (*data_settings)[MergeTreeSetting::merge_selector_enable_heuristic_to_remove_small_parts_at_right]; simple_merge_settings.base = (*data_settings)[MergeTreeSetting::merge_selector_base]; + simple_merge_settings.min_parts_to_merge_at_once = (*data_settings)[MergeTreeSetting::min_parts_to_merge_at_once]; if (!(*data_settings)[MergeTreeSetting::min_age_to_force_merge_on_partition_only]) simple_merge_settings.min_age_to_force_merge = (*data_settings)[MergeTreeSetting::min_age_to_force_merge_seconds]; diff --git a/src/Storages/MergeTree/MergeTreeSettings.cpp b/src/Storages/MergeTree/MergeTreeSettings.cpp index 33910d1048d..fcd4e05cf00 100644 --- a/src/Storages/MergeTree/MergeTreeSettings.cpp +++ b/src/Storages/MergeTree/MergeTreeSettings.cpp @@ -102,6 +102,7 @@ namespace ErrorCodes DECLARE(MergeSelectorAlgorithm, merge_selector_algorithm, MergeSelectorAlgorithm::SIMPLE, "The algorithm to select parts for merges assignment", EXPERIMENTAL) \ DECLARE(Bool, merge_selector_enable_heuristic_to_remove_small_parts_at_right, true, "Enable heuristic for selecting parts for merge which removes parts from right side of range, if their size is less than specified ratio (0.01) of sum_size. Works for Simple and StochasticSimple merge selectors", 0) \ DECLARE(Float, merge_selector_base, 5.0, "Affects write amplification of assigned merges (expert level setting, don't change if you don't understand what it is doing). Works for Simple and StochasticSimple merge selectors", 0) \ + DECLARE(UInt64, min_parts_to_merge_at_once, 0, "Minimal amount of data parts which merge selector can pick to merge at once (expert level setting, don't change if you don't understand what it is doing). 0 - disabled. Works for Simple and StochasticSimple merge selectors.", 0) \ \ /** Inserts settings. */ \ DECLARE(UInt64, parts_to_delay_insert, 1000, "If table contains at least that many active parts in single partition, artificially slow down insert into table. Disabled if set to 0", 0) \ From b6cad9c913b304052939cd100ba4e9d35b44c47a Mon Sep 17 00:00:00 2001 From: alesapin Date: Fri, 8 Nov 2024 12:25:26 +0100 Subject: [PATCH 560/680] Add test --- ...03267_min_parts_to_merge_at_once.reference | 4 ++ .../03267_min_parts_to_merge_at_once.sh | 43 +++++++++++++++++++ 2 files changed, 47 insertions(+) create mode 100644 tests/queries/0_stateless/03267_min_parts_to_merge_at_once.reference create mode 100755 tests/queries/0_stateless/03267_min_parts_to_merge_at_once.sh diff --git a/tests/queries/0_stateless/03267_min_parts_to_merge_at_once.reference b/tests/queries/0_stateless/03267_min_parts_to_merge_at_once.reference new file mode 100644 index 00000000000..966a0980e59 --- /dev/null +++ b/tests/queries/0_stateless/03267_min_parts_to_merge_at_once.reference @@ -0,0 +1,4 @@ +2 +3 +4 +1 diff --git a/tests/queries/0_stateless/03267_min_parts_to_merge_at_once.sh b/tests/queries/0_stateless/03267_min_parts_to_merge_at_once.sh new file mode 100755 index 00000000000..e069b57bf86 --- /dev/null +++ b/tests/queries/0_stateless/03267_min_parts_to_merge_at_once.sh @@ -0,0 +1,43 @@ +#!/usr/bin/env bash + +CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CURDIR"/../shell_config.sh + +$CLICKHOUSE_CLIENT --query "DROP TABLE IF EXISTS t;" + +$CLICKHOUSE_CLIENT --query "CREATE TABLE t (key UInt64) ENGINE = MergeTree() ORDER BY tuple() SETTINGS min_parts_to_merge_at_once=5, merge_selector_base=1" + +$CLICKHOUSE_CLIENT --query "INSERT INTO t VALUES (1)" +$CLICKHOUSE_CLIENT --query "INSERT INTO t VALUES (2);" + +# doesn't make test flaky +sleep 1 + +$CLICKHOUSE_CLIENT --query "SELECT count() FROM system.parts WHERE active and database = currentDatabase() and table = 't'" + +$CLICKHOUSE_CLIENT --query "INSERT INTO t VALUES (3)" + +$CLICKHOUSE_CLIENT --query "SELECT count() FROM system.parts WHERE active and database = currentDatabase() and table = 't'" + +$CLICKHOUSE_CLIENT --query "INSERT INTO t VALUES (4)" + +$CLICKHOUSE_CLIENT --query "SELECT count() FROM system.parts WHERE active and database = currentDatabase() and table = 't'" + +$CLICKHOUSE_CLIENT --query "INSERT INTO t VALUES (5)" + +counter=0 retries=60 + +I=0 +while [[ $counter -lt $retries ]]; do + result=$($CLICKHOUSE_CLIENT --query "SELECT count() FROM system.parts WHERE active and database = currentDatabase() and table = 't'") + if [ "$result" -eq "1" ];then + break; + fi + sleep 0.5 + counter=$((counter + 1)) +done + +$CLICKHOUSE_CLIENT --query "SELECT count() FROM system.parts WHERE active and database = currentDatabase() and table = 't'" + +$CLICKHOUSE_CLIENT --query "DROP TABLE IF EXISTS t" From 4c644a98f5985a540ee75dc5a1f5ae31be39cc15 Mon Sep 17 00:00:00 2001 From: Pavel Kruglov <48961922+Avogar@users.noreply.github.com> Date: Fri, 8 Nov 2024 12:29:04 +0100 Subject: [PATCH 561/680] Fix broken 03247_ghdata_string_to_json_alter --- .../queries/0_stateless/03247_ghdata_string_to_json_alter.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh b/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh index 931d106120c..a2d1788cb5d 100755 --- a/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh +++ b/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh @@ -18,12 +18,12 @@ ${CLICKHOUSE_CLIENT} -q "SELECT count() FROM ghdata WHERE NOT ignore(*)" ${CLICKHOUSE_CLIENT} -q \ "SELECT data.repo.name, count() AS stars FROM ghdata \ - WHERE data.type = 'WatchEvent' GROUP BY data.repo.name ORDER BY stars DESC, data.repo.name LIMIT 5" + WHERE data.type = 'WatchEvent' GROUP BY data.repo.name ORDER BY stars DESC, data.repo.name LIMIT 5" --allow_suspicious_types_in_group_by=1, --allow_suspicious_types_in_order_by=1 ${CLICKHOUSE_CLIENT} --enable_analyzer=1 -q \ "SELECT data.payload.commits[].author.name AS name, count() AS c FROM ghdata \ ARRAY JOIN data.payload.commits[].author.name \ - GROUP BY name ORDER BY c DESC, name LIMIT 5" + GROUP BY name ORDER BY c DESC, name LIMIT 5" --allow_suspicious_types_in_group_by=1, --allow_suspicious_types_in_order_by=1 ${CLICKHOUSE_CLIENT} -q "SELECT max(data.payload.pull_request.assignees[].size0) FROM ghdata" From 96383d42b184df2e05e9d8aa5ee83dbce4105800 Mon Sep 17 00:00:00 2001 From: Pablo Marcos Date: Fri, 8 Nov 2024 11:38:43 +0000 Subject: [PATCH 562/680] Small refactor to ease debugging when something happens on the CI --- src/Interpreters/QueryMetricLog.cpp | 13 ++++++------- .../03203_system_query_metric_log.reference | 6 +++--- .../0_stateless/03203_system_query_metric_log.sh | 6 +++--- 3 files changed, 12 insertions(+), 13 deletions(-) diff --git a/src/Interpreters/QueryMetricLog.cpp b/src/Interpreters/QueryMetricLog.cpp index e784c357b29..4fbe4f9e1b5 100644 --- a/src/Interpreters/QueryMetricLog.cpp +++ b/src/Interpreters/QueryMetricLog.cpp @@ -148,7 +148,7 @@ void QueryMetricLog::startQuery(const String & query_id, TimePoint start_time, U { QueryMetricLogStatus query_status; query_status.interval_milliseconds = interval_milliseconds; - query_status.next_collect_time = start_time + std::chrono::milliseconds(interval_milliseconds); + query_status.next_collect_time = start_time; auto context = getContext(); const auto & process_list = context->getProcessList(); @@ -213,6 +213,7 @@ void QueryMetricLog::finishQuery(const String & query_id, TimePoint finish_time, void QueryMetricLogStatus::scheduleNext(String query_id) { + next_collect_time += std::chrono::milliseconds(interval_milliseconds); const auto now = std::chrono::system_clock::now(); if (next_collect_time > now) { @@ -229,8 +230,9 @@ void QueryMetricLogStatus::scheduleNext(String query_id) std::optional QueryMetricLogStatus::createLogMetricElement(const String & query_id, const QueryStatusInfo & query_info, TimePoint query_info_time, bool schedule_next) { - LOG_TRACE(logger, "Collecting query_metric_log for query {} and interval {} ms with QueryStatusInfo from {}. Schedule next: {}", - query_id, interval_milliseconds, timePointToString(query_info_time), schedule_next); + LOG_TRACE(logger, "Collecting query_metric_log for query {} and interval {} ms with QueryStatusInfo from {}. Next collection time: {}", + query_id, interval_milliseconds, timePointToString(query_info_time), + schedule_next ? timePointToString(next_collect_time + std::chrono::milliseconds(interval_milliseconds)) : "finished"); if (query_info_time <= last_collect_time) { @@ -276,15 +278,12 @@ std::optional QueryMetricLogStatus::createLogMetricElemen } else { - LOG_TRACE(logger, "Query {} has no profile counters", query_id); + LOG_WARNING(logger, "Query {} has no profile counters", query_id); elem.profile_events = std::vector(ProfileEvents::end()); } if (schedule_next) - { - next_collect_time += std::chrono::milliseconds(interval_milliseconds); scheduleNext(query_id); - } return elem; } diff --git a/tests/queries/0_stateless/03203_system_query_metric_log.reference b/tests/queries/0_stateless/03203_system_query_metric_log.reference index 940b0c4e178..fa8e27a7e90 100644 --- a/tests/queries/0_stateless/03203_system_query_metric_log.reference +++ b/tests/queries/0_stateless/03203_system_query_metric_log.reference @@ -23,8 +23,8 @@ --Interval 123: check that the SleepFunctionCalls, SleepFunctionMilliseconds and ProfileEvent_SleepFunctionElapsedMicroseconds are correct 1 --Check that a query_metric_log_interval=0 disables the collection -0 +1 -Check that a query which execution time is less than query_metric_log_interval is never collected -0 +1 --Check that there is a final event when queries finish -3 +1 diff --git a/tests/queries/0_stateless/03203_system_query_metric_log.sh b/tests/queries/0_stateless/03203_system_query_metric_log.sh index bf94be79d7c..abcd14c8e5d 100755 --- a/tests/queries/0_stateless/03203_system_query_metric_log.sh +++ b/tests/queries/0_stateless/03203_system_query_metric_log.sh @@ -84,17 +84,17 @@ check_log 123 # query_metric_log_interval=0 disables the collection altogether $CLICKHOUSE_CLIENT -m -q """ SELECT '--Check that a query_metric_log_interval=0 disables the collection'; - SELECT count() FROM system.query_metric_log WHERE event_date >= yesterday() AND query_id = '${query_prefix}_0' + SELECT count() == 0 FROM system.query_metric_log WHERE event_date >= yesterday() AND query_id = '${query_prefix}_0' """ # a quick query that takes less than query_metric_log_interval is never collected $CLICKHOUSE_CLIENT -m -q """ SELECT '-Check that a query which execution time is less than query_metric_log_interval is never collected'; - SELECT count() FROM system.query_metric_log WHERE event_date >= yesterday() AND query_id = '${query_prefix}_fast' + SELECT count() == 0 FROM system.query_metric_log WHERE event_date >= yesterday() AND query_id = '${query_prefix}_fast' """ # a query that takes more than query_metric_log_interval is collected including the final row $CLICKHOUSE_CLIENT -m -q """ SELECT '--Check that there is a final event when queries finish'; - SELECT count() FROM system.query_metric_log WHERE event_date >= yesterday() AND query_id = '${query_prefix}_1000' + SELECT count() > 2 FROM system.query_metric_log WHERE event_date >= yesterday() AND query_id = '${query_prefix}_1000' """ From 1bd6b9df95792e8917e1da744a0d8e7d586949ed Mon Sep 17 00:00:00 2001 From: alesapin Date: Fri, 8 Nov 2024 12:47:48 +0100 Subject: [PATCH 563/680] Fix style check --- tests/queries/0_stateless/03267_min_parts_to_merge_at_once.sh | 1 - 1 file changed, 1 deletion(-) diff --git a/tests/queries/0_stateless/03267_min_parts_to_merge_at_once.sh b/tests/queries/0_stateless/03267_min_parts_to_merge_at_once.sh index e069b57bf86..90b9d0339cf 100755 --- a/tests/queries/0_stateless/03267_min_parts_to_merge_at_once.sh +++ b/tests/queries/0_stateless/03267_min_parts_to_merge_at_once.sh @@ -28,7 +28,6 @@ $CLICKHOUSE_CLIENT --query "INSERT INTO t VALUES (5)" counter=0 retries=60 -I=0 while [[ $counter -lt $retries ]]; do result=$($CLICKHOUSE_CLIENT --query "SELECT count() FROM system.parts WHERE active and database = currentDatabase() and table = 't'") if [ "$result" -eq "1" ];then From 10329cbbf2da51925e5a4580a8ba9faf3315cd02 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Fri, 8 Nov 2024 12:07:30 +0100 Subject: [PATCH 564/680] Generate clickhouse/clickhouse-server README as in docker-library --- docker/server/README.md | 8 ++ docker/server/README.sh | 38 +++++ docker/server/README.src/README-short.txt | 1 + docker/server/README.src/content.md | 166 ++++++++++++++++++++++ docker/server/README.src/github-repo | 1 + docker/server/README.src/license.md | 1 + docker/server/README.src/logo.svg | 43 ++++++ docker/server/README.src/maintainer.md | 1 + docker/server/README.src/metadata.json | 7 + 9 files changed, 266 insertions(+) create mode 100755 docker/server/README.sh create mode 100644 docker/server/README.src/README-short.txt create mode 100644 docker/server/README.src/content.md create mode 100644 docker/server/README.src/github-repo create mode 100644 docker/server/README.src/license.md create mode 100644 docker/server/README.src/logo.svg create mode 100644 docker/server/README.src/maintainer.md create mode 100644 docker/server/README.src/metadata.json diff --git a/docker/server/README.md b/docker/server/README.md index 1dc636414ac..e8c60204c96 100644 --- a/docker/server/README.md +++ b/docker/server/README.md @@ -1,3 +1,11 @@ + + # ClickHouse Server Docker Image ## What is ClickHouse? diff --git a/docker/server/README.sh b/docker/server/README.sh new file mode 100755 index 00000000000..42fa72404d1 --- /dev/null +++ b/docker/server/README.sh @@ -0,0 +1,38 @@ +#!/usr/bin/env bash +set -ueo pipefail + +# A script to generate README.sh close to as it done in https://github.com/docker-library/docs + +WORKDIR=$(dirname "$0") +SCRIPT_NAME=$(basename "$0") +CONTENT=README.src/content.md +LICENSE=README.src/license.md +cd "$WORKDIR" + +R=README.md + +cat > "$R" < + +EOD + +cat "$CONTENT" >> "$R" + +cat >> "$R" <=2, Azure and GCP instances. Examples for unsupported devices are Raspberry Pi 4 (ARMv8.0-A) and Jetson AGX Xavier/Orin (ARMv8.2-A). +- Since the Clickhouse 24.11 Ubuntu images started using `ubuntu:22.04` as its base image. It requires docker version >= `20.10.10` containing [patch](https://github.com/moby/moby/commit/977283509f75303bc6612665a04abf76ff1d2468). As a workaround you could use `docker run [--privileged | --security-opt seccomp=unconfined]` instead, however that has security implications. + +## How to use this image + +### start server instance + +```bash +docker run -d --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server +``` + +By default, ClickHouse will be accessible only via the Docker network. See the [networking section below](#networking). + +By default, starting above server instance will be run as the `default` user without password. + +### connect to it from a native client + +```bash +docker run -it --rm --link some-clickhouse-server:clickhouse-server --entrypoint clickhouse-client clickhouse/clickhouse-server --host clickhouse-server +# OR +docker exec -it some-clickhouse-server clickhouse-client +``` + +More information about the [ClickHouse client](https://clickhouse.com/docs/en/interfaces/cli/). + +### connect to it using curl + +```bash +echo "SELECT 'Hello, ClickHouse!'" | docker run -i --rm --link some-clickhouse-server:clickhouse-server curlimages/curl 'http://clickhouse-server:8123/?query=' -s --data-binary @- +``` + +More information about the [ClickHouse HTTP Interface](https://clickhouse.com/docs/en/interfaces/http/). + +### stopping / removing the container + +```bash +docker stop some-clickhouse-server +docker rm some-clickhouse-server +``` + +### networking + +You can expose your ClickHouse running in docker by [mapping a particular port](https://docs.docker.com/config/containers/container-networking/) from inside the container using host ports: + +```bash +docker run -d -p 18123:8123 -p19000:9000 --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server +echo 'SELECT version()' | curl 'http://localhost:18123/' --data-binary @- +``` + +`22.6.3.35` + +or by allowing the container to use [host ports directly](https://docs.docker.com/network/host/) using `--network=host` (also allows achieving better network performance): + +```bash +docker run -d --network=host --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server +echo 'SELECT version()' | curl 'http://localhost:8123/' --data-binary @- +``` + +`22.6.3.35` + +### Volumes + +Typically you may want to mount the following folders inside your container to achieve persistency: + +- `/var/lib/clickhouse/` - main folder where ClickHouse stores the data +- `/var/log/clickhouse-server/` - logs + +```bash +docker run -d \ + -v $(realpath ./ch_data):/var/lib/clickhouse/ \ + -v $(realpath ./ch_logs):/var/log/clickhouse-server/ \ + --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server +``` + +You may also want to mount: + +- `/etc/clickhouse-server/config.d/*.xml` - files with server configuration adjustments +- `/etc/clickhouse-server/users.d/*.xml` - files with user settings adjustments +- `/docker-entrypoint-initdb.d/` - folder with database initialization scripts (see below). + +### Linux capabilities + +ClickHouse has some advanced functionality, which requires enabling several [Linux capabilities](https://man7.org/linux/man-pages/man7/capabilities.7.html). + +They are optional and can be enabled using the following [docker command-line arguments](https://docs.docker.com/engine/reference/run/#runtime-privilege-and-linux-capabilities): + +```bash +docker run -d \ + --cap-add=SYS_NICE --cap-add=NET_ADMIN --cap-add=IPC_LOCK \ + --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server +``` + +## Configuration + +The container exposes port 8123 for the [HTTP interface](https://clickhouse.com/docs/en/interfaces/http_interface/) and port 9000 for the [native client](https://clickhouse.com/docs/en/interfaces/tcp/). + +ClickHouse configuration is represented with a file "config.xml" ([documentation](https://clickhouse.com/docs/en/operations/configuration_files/)) + +### Start server instance with custom configuration + +```bash +docker run -d --name some-clickhouse-server --ulimit nofile=262144:262144 -v /path/to/your/config.xml:/etc/clickhouse-server/config.xml clickhouse/clickhouse-server +``` + +### Start server as custom user + +```bash +# $(pwd)/data/clickhouse should exist and be owned by current user +docker run --rm --user ${UID}:${GID} --name some-clickhouse-server --ulimit nofile=262144:262144 -v "$(pwd)/logs/clickhouse:/var/log/clickhouse-server" -v "$(pwd)/data/clickhouse:/var/lib/clickhouse" clickhouse/clickhouse-server +``` + +When you use the image with local directories mounted, you probably want to specify the user to maintain the proper file ownership. Use the `--user` argument and mount `/var/lib/clickhouse` and `/var/log/clickhouse-server` inside the container. Otherwise, the image will complain and not start. + +### Start server from root (useful in case of enabled user namespace) + +```bash +docker run --rm -e CLICKHOUSE_UID=0 -e CLICKHOUSE_GID=0 --name clickhouse-server-userns -v "$(pwd)/logs/clickhouse:/var/log/clickhouse-server" -v "$(pwd)/data/clickhouse:/var/lib/clickhouse" clickhouse/clickhouse-server +``` + +### How to create default database and user on starting + +Sometimes you may want to create a user (user named `default` is used by default) and database on a container start. You can do it using environment variables `CLICKHOUSE_DB`, `CLICKHOUSE_USER`, `CLICKHOUSE_DEFAULT_ACCESS_MANAGEMENT` and `CLICKHOUSE_PASSWORD`: + +```bash +docker run --rm -e CLICKHOUSE_DB=my_database -e CLICKHOUSE_USER=username -e CLICKHOUSE_DEFAULT_ACCESS_MANAGEMENT=1 -e CLICKHOUSE_PASSWORD=password -p 9000:9000/tcp clickhouse/clickhouse-server +``` + +## How to extend this image + +To perform additional initialization in an image derived from this one, add one or more `*.sql`, `*.sql.gz`, or `*.sh` scripts under `/docker-entrypoint-initdb.d`. After the entrypoint calls `initdb`, it will run any `*.sql` files, run any executable `*.sh` scripts, and source any non-executable `*.sh` scripts found in that directory to do further initialization before starting the service. +Also, you can provide environment variables `CLICKHOUSE_USER` & `CLICKHOUSE_PASSWORD` that will be used for clickhouse-client during initialization. + +For example, to add an additional user and database, add the following to `/docker-entrypoint-initdb.d/init-db.sh`: + +```bash +#!/bin/bash +set -e + +clickhouse client -n <<-EOSQL + CREATE DATABASE docker; + CREATE TABLE docker.docker (x Int32) ENGINE = Log; +EOSQL +``` diff --git a/docker/server/README.src/github-repo b/docker/server/README.src/github-repo new file mode 100644 index 00000000000..dc2b6635325 --- /dev/null +++ b/docker/server/README.src/github-repo @@ -0,0 +1 @@ +https://github.com/ClickHouse/docker-library diff --git a/docker/server/README.src/license.md b/docker/server/README.src/license.md new file mode 100644 index 00000000000..6be024edcde --- /dev/null +++ b/docker/server/README.src/license.md @@ -0,0 +1 @@ +View [license information](https://github.com/ClickHouse/ClickHouse/blob/master/LICENSE) for the software contained in this image. diff --git a/docker/server/README.src/logo.svg b/docker/server/README.src/logo.svg new file mode 100644 index 00000000000..a50dd81a164 --- /dev/null +++ b/docker/server/README.src/logo.svg @@ -0,0 +1,43 @@ + + + + + + + + + + + + + + + + + + + + + + + \ No newline at end of file diff --git a/docker/server/README.src/maintainer.md b/docker/server/README.src/maintainer.md new file mode 100644 index 00000000000..26c7db1a293 --- /dev/null +++ b/docker/server/README.src/maintainer.md @@ -0,0 +1 @@ +[ClickHouse Inc.](%%GITHUB-REPO%%) diff --git a/docker/server/README.src/metadata.json b/docker/server/README.src/metadata.json new file mode 100644 index 00000000000..3d3937b21fb --- /dev/null +++ b/docker/server/README.src/metadata.json @@ -0,0 +1,7 @@ +{ + "hub": { + "categories": [ + "databases-and-storage" + ] + } +} From aa15b912df09bb95e400a50aa007266948f75697 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Fri, 8 Nov 2024 12:58:50 +0100 Subject: [PATCH 565/680] Apply review comments from docker-library/docs --- docker/server/README.md | 20 +++++++++------- docker/server/README.src/content.md | 36 ++++++++++++++++------------- 2 files changed, 32 insertions(+), 24 deletions(-) diff --git a/docker/server/README.md b/docker/server/README.md index e8c60204c96..7403d5b0b2a 100644 --- a/docker/server/README.md +++ b/docker/server/README.md @@ -16,6 +16,7 @@ ClickHouse works 100-1000x faster than traditional database management systems, For more information and documentation see https://clickhouse.com/. + ## Versions - The `latest` tag points to the latest release of the latest stable branch. @@ -24,6 +25,7 @@ For more information and documentation see https://clickhouse.com/. - The tag `head` is built from the latest commit to the default branch. - Each tag has optional `-alpine` suffix to reflect that it's built on top of `alpine`. + ### Compatibility - The amd64 image requires support for [SSE3 instructions](https://en.wikipedia.org/wiki/SSE3). Virtually all x86 CPUs after 2005 support SSE3. @@ -38,7 +40,7 @@ For more information and documentation see https://clickhouse.com/. docker run -d --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server ``` -By default, ClickHouse will be accessible only via the Docker network. See the [networking section below](#networking). +By default, ClickHouse will be accessible only via the Docker network. See the **networking** section below. By default, starting above server instance will be run as the `default` user without password. @@ -55,7 +57,7 @@ More information about the [ClickHouse client](https://clickhouse.com/docs/en/in ### connect to it using curl ```bash -echo "SELECT 'Hello, ClickHouse!'" | docker run -i --rm --link some-clickhouse-server:clickhouse-server curlimages/curl 'http://clickhouse-server:8123/?query=' -s --data-binary @- +echo "SELECT 'Hello, ClickHouse!'" | docker run -i --rm --link some-clickhouse-server:clickhouse-server buildpack-deps:curl 'http://clickhouse-server:8123/?query=' -s --data-binary @- ``` More information about the [ClickHouse HTTP Interface](https://clickhouse.com/docs/en/interfaces/http/). @@ -78,7 +80,7 @@ echo 'SELECT version()' | curl 'http://localhost:18123/' --data-binary @- `22.6.3.35` -or by allowing the container to use [host ports directly](https://docs.docker.com/network/host/) using `--network=host` (also allows achieving better network performance): +Or by allowing the container to use [host ports directly](https://docs.docker.com/network/host/) using `--network=host` (also allows achieving better network performance): ```bash docker run -d --network=host --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server @@ -96,8 +98,8 @@ Typically you may want to mount the following folders inside your container to a ```bash docker run -d \ - -v $(realpath ./ch_data):/var/lib/clickhouse/ \ - -v $(realpath ./ch_logs):/var/log/clickhouse-server/ \ + -v "$PWD/ch_data:/var/lib/clickhouse/" \ + -v "$PWD/ch_logs:/var/log/clickhouse-server/" \ --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server ``` @@ -119,6 +121,8 @@ docker run -d \ --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server ``` +Read more in [knowledge base](https://clickhouse.com/docs/knowledgebase/configure_cap_ipc_lock_and_cap_sys_nice_in_docker). + ## Configuration The container exposes port 8123 for the [HTTP interface](https://clickhouse.com/docs/en/interfaces/http_interface/) and port 9000 for the [native client](https://clickhouse.com/docs/en/interfaces/tcp/). @@ -134,8 +138,8 @@ docker run -d --name some-clickhouse-server --ulimit nofile=262144:262144 -v /pa ### Start server as custom user ```bash -# $(pwd)/data/clickhouse should exist and be owned by current user -docker run --rm --user ${UID}:${GID} --name some-clickhouse-server --ulimit nofile=262144:262144 -v "$(pwd)/logs/clickhouse:/var/log/clickhouse-server" -v "$(pwd)/data/clickhouse:/var/lib/clickhouse" clickhouse/clickhouse-server +# $PWD/data/clickhouse should exist and be owned by current user +docker run --rm --user "${UID}:${GID}" --name some-clickhouse-server --ulimit nofile=262144:262144 -v "$PWD/logs/clickhouse:/var/log/clickhouse-server" -v "$PWD/data/clickhouse:/var/lib/clickhouse" clickhouse/clickhouse-server ``` When you use the image with local directories mounted, you probably want to specify the user to maintain the proper file ownership. Use the `--user` argument and mount `/var/lib/clickhouse` and `/var/log/clickhouse-server` inside the container. Otherwise, the image will complain and not start. @@ -143,7 +147,7 @@ When you use the image with local directories mounted, you probably want to spec ### Start server from root (useful in case of enabled user namespace) ```bash -docker run --rm -e CLICKHOUSE_UID=0 -e CLICKHOUSE_GID=0 --name clickhouse-server-userns -v "$(pwd)/logs/clickhouse:/var/log/clickhouse-server" -v "$(pwd)/data/clickhouse:/var/lib/clickhouse" clickhouse/clickhouse-server +docker run --rm -e CLICKHOUSE_RUN_AS_ROOT=1 --name clickhouse-server-userns -v "$PWD/logs/clickhouse:/var/log/clickhouse-server" -v "$PWD/data/clickhouse:/var/lib/clickhouse" clickhouse/clickhouse-server ``` ### How to create default database and user on starting diff --git a/docker/server/README.src/content.md b/docker/server/README.src/content.md index e790de41236..bfc1a271546 100644 --- a/docker/server/README.src/content.md +++ b/docker/server/README.src/content.md @@ -10,6 +10,7 @@ ClickHouse works 100-1000x faster than traditional database management systems, For more information and documentation see https://clickhouse.com/. + ## Versions - The `latest` tag points to the latest release of the latest stable branch. @@ -18,6 +19,7 @@ For more information and documentation see https://clickhouse.com/. - The tag `head` is built from the latest commit to the default branch. - Each tag has optional `-alpine` suffix to reflect that it's built on top of `alpine`. + ### Compatibility - The amd64 image requires support for [SSE3 instructions](https://en.wikipedia.org/wiki/SSE3). Virtually all x86 CPUs after 2005 support SSE3. @@ -29,17 +31,17 @@ For more information and documentation see https://clickhouse.com/. ### start server instance ```bash -docker run -d --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server +docker run -d --name some-clickhouse-server --ulimit nofile=262144:262144 %%IMAGE%% ``` -By default, ClickHouse will be accessible only via the Docker network. See the [networking section below](#networking). +By default, ClickHouse will be accessible only via the Docker network. See the **networking** section below. By default, starting above server instance will be run as the `default` user without password. ### connect to it from a native client ```bash -docker run -it --rm --link some-clickhouse-server:clickhouse-server --entrypoint clickhouse-client clickhouse/clickhouse-server --host clickhouse-server +docker run -it --rm --link some-clickhouse-server:clickhouse-server --entrypoint clickhouse-client %%IMAGE%% --host clickhouse-server # OR docker exec -it some-clickhouse-server clickhouse-client ``` @@ -49,7 +51,7 @@ More information about the [ClickHouse client](https://clickhouse.com/docs/en/in ### connect to it using curl ```bash -echo "SELECT 'Hello, ClickHouse!'" | docker run -i --rm --link some-clickhouse-server:clickhouse-server curlimages/curl 'http://clickhouse-server:8123/?query=' -s --data-binary @- +echo "SELECT 'Hello, ClickHouse!'" | docker run -i --rm --link some-clickhouse-server:clickhouse-server buildpack-deps:curl 'http://clickhouse-server:8123/?query=' -s --data-binary @- ``` More information about the [ClickHouse HTTP Interface](https://clickhouse.com/docs/en/interfaces/http/). @@ -66,16 +68,16 @@ docker rm some-clickhouse-server You can expose your ClickHouse running in docker by [mapping a particular port](https://docs.docker.com/config/containers/container-networking/) from inside the container using host ports: ```bash -docker run -d -p 18123:8123 -p19000:9000 --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server +docker run -d -p 18123:8123 -p19000:9000 --name some-clickhouse-server --ulimit nofile=262144:262144 %%IMAGE%% echo 'SELECT version()' | curl 'http://localhost:18123/' --data-binary @- ``` `22.6.3.35` -or by allowing the container to use [host ports directly](https://docs.docker.com/network/host/) using `--network=host` (also allows achieving better network performance): +Or by allowing the container to use [host ports directly](https://docs.docker.com/network/host/) using `--network=host` (also allows achieving better network performance): ```bash -docker run -d --network=host --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server +docker run -d --network=host --name some-clickhouse-server --ulimit nofile=262144:262144 %%IMAGE%% echo 'SELECT version()' | curl 'http://localhost:8123/' --data-binary @- ``` @@ -90,9 +92,9 @@ Typically you may want to mount the following folders inside your container to a ```bash docker run -d \ - -v $(realpath ./ch_data):/var/lib/clickhouse/ \ - -v $(realpath ./ch_logs):/var/log/clickhouse-server/ \ - --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server + -v "$PWD/ch_data:/var/lib/clickhouse/" \ + -v "$PWD/ch_logs:/var/log/clickhouse-server/" \ + --name some-clickhouse-server --ulimit nofile=262144:262144 %%IMAGE%% ``` You may also want to mount: @@ -110,9 +112,11 @@ They are optional and can be enabled using the following [docker command-line ar ```bash docker run -d \ --cap-add=SYS_NICE --cap-add=NET_ADMIN --cap-add=IPC_LOCK \ - --name some-clickhouse-server --ulimit nofile=262144:262144 clickhouse/clickhouse-server + --name some-clickhouse-server --ulimit nofile=262144:262144 %%IMAGE%% ``` +Read more in [knowledge base](https://clickhouse.com/docs/knowledgebase/configure_cap_ipc_lock_and_cap_sys_nice_in_docker). + ## Configuration The container exposes port 8123 for the [HTTP interface](https://clickhouse.com/docs/en/interfaces/http_interface/) and port 9000 for the [native client](https://clickhouse.com/docs/en/interfaces/tcp/). @@ -122,14 +126,14 @@ ClickHouse configuration is represented with a file "config.xml" ([documentation ### Start server instance with custom configuration ```bash -docker run -d --name some-clickhouse-server --ulimit nofile=262144:262144 -v /path/to/your/config.xml:/etc/clickhouse-server/config.xml clickhouse/clickhouse-server +docker run -d --name some-clickhouse-server --ulimit nofile=262144:262144 -v /path/to/your/config.xml:/etc/clickhouse-server/config.xml %%IMAGE%% ``` ### Start server as custom user ```bash -# $(pwd)/data/clickhouse should exist and be owned by current user -docker run --rm --user ${UID}:${GID} --name some-clickhouse-server --ulimit nofile=262144:262144 -v "$(pwd)/logs/clickhouse:/var/log/clickhouse-server" -v "$(pwd)/data/clickhouse:/var/lib/clickhouse" clickhouse/clickhouse-server +# $PWD/data/clickhouse should exist and be owned by current user +docker run --rm --user "${UID}:${GID}" --name some-clickhouse-server --ulimit nofile=262144:262144 -v "$PWD/logs/clickhouse:/var/log/clickhouse-server" -v "$PWD/data/clickhouse:/var/lib/clickhouse" %%IMAGE%% ``` When you use the image with local directories mounted, you probably want to specify the user to maintain the proper file ownership. Use the `--user` argument and mount `/var/lib/clickhouse` and `/var/log/clickhouse-server` inside the container. Otherwise, the image will complain and not start. @@ -137,7 +141,7 @@ When you use the image with local directories mounted, you probably want to spec ### Start server from root (useful in case of enabled user namespace) ```bash -docker run --rm -e CLICKHOUSE_UID=0 -e CLICKHOUSE_GID=0 --name clickhouse-server-userns -v "$(pwd)/logs/clickhouse:/var/log/clickhouse-server" -v "$(pwd)/data/clickhouse:/var/lib/clickhouse" clickhouse/clickhouse-server +docker run --rm -e CLICKHOUSE_RUN_AS_ROOT=1 --name clickhouse-server-userns -v "$PWD/logs/clickhouse:/var/log/clickhouse-server" -v "$PWD/data/clickhouse:/var/lib/clickhouse" %%IMAGE%% ``` ### How to create default database and user on starting @@ -145,7 +149,7 @@ docker run --rm -e CLICKHOUSE_UID=0 -e CLICKHOUSE_GID=0 --name clickhouse-server Sometimes you may want to create a user (user named `default` is used by default) and database on a container start. You can do it using environment variables `CLICKHOUSE_DB`, `CLICKHOUSE_USER`, `CLICKHOUSE_DEFAULT_ACCESS_MANAGEMENT` and `CLICKHOUSE_PASSWORD`: ```bash -docker run --rm -e CLICKHOUSE_DB=my_database -e CLICKHOUSE_USER=username -e CLICKHOUSE_DEFAULT_ACCESS_MANAGEMENT=1 -e CLICKHOUSE_PASSWORD=password -p 9000:9000/tcp clickhouse/clickhouse-server +docker run --rm -e CLICKHOUSE_DB=my_database -e CLICKHOUSE_USER=username -e CLICKHOUSE_DEFAULT_ACCESS_MANAGEMENT=1 -e CLICKHOUSE_PASSWORD=password -p 9000:9000/tcp %%IMAGE%% ``` ## How to extend this image From 0dbc041d8bc49d2760fe85a8a76431395571dfb8 Mon Sep 17 00:00:00 2001 From: Pablo Marcos Date: Fri, 8 Nov 2024 12:00:34 +0000 Subject: [PATCH 566/680] Log when the query finishes for system.query_metric_log ASAP There are logs where we can see that after the query finishes, executeQuery takes up to 2 seconds to call finishQuery. --- src/Interpreters/executeQuery.cpp | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/src/Interpreters/executeQuery.cpp b/src/Interpreters/executeQuery.cpp index 9250c069283..4507126b7b3 100644 --- a/src/Interpreters/executeQuery.cpp +++ b/src/Interpreters/executeQuery.cpp @@ -506,6 +506,7 @@ void logQueryFinish( auto time_now = std::chrono::system_clock::now(); QueryStatusInfo info = process_list_elem->getInfo(true, settings[Setting::log_profile_events]); + logQueryMetricLogFinish(context, internal, elem.client_info.current_query_id, time_now, std::make_shared(info)); elem.type = QueryLogElementType::QUERY_FINISH; addStatusInfoToQueryLogElement(elem, info, query_ast, context); @@ -551,6 +552,7 @@ void logQueryFinish( if (auto query_log = context->getQueryLog()) query_log->add(elem); } + if (settings[Setting::log_processors_profiles]) { if (auto processors_profile_log = context->getProcessorsProfileLog()) @@ -598,8 +600,6 @@ void logQueryFinish( } } } - - logQueryMetricLogFinish(context, internal, elem.client_info.current_query_id, time_now, std::make_shared(info)); } if (query_span) @@ -669,6 +669,7 @@ void logQueryException( { elem.query_duration_ms = start_watch.elapsedMilliseconds(); } + logQueryMetricLogFinish(context, internal, elem.client_info.current_query_id, time_now, info); elem.query_cache_usage = QueryCache::Usage::None; @@ -698,8 +699,6 @@ void logQueryException( query_span->addAttribute("clickhouse.exception_code", elem.exception_code); query_span->finish(); } - - logQueryMetricLogFinish(context, internal, elem.client_info.current_query_id, time_now, info); } void logExceptionBeforeStart( @@ -753,6 +752,8 @@ void logExceptionBeforeStart( elem.client_info = context->getClientInfo(); + logQueryMetricLogFinish(context, false, elem.client_info.current_query_id, std::chrono::system_clock::now(), nullptr); + elem.log_comment = settings[Setting::log_comment]; if (elem.log_comment.size() > settings[Setting::max_query_size]) elem.log_comment.resize(settings[Setting::max_query_size]); @@ -797,8 +798,6 @@ void logExceptionBeforeStart( ProfileEvents::increment(ProfileEvents::FailedInsertQuery); } } - - logQueryMetricLogFinish(context, false, elem.client_info.current_query_id, std::chrono::system_clock::now(), nullptr); } void validateAnalyzerSettings(ASTPtr ast, bool context_value) From 1dcd06f098fbd661d1327cd9ecdabd32f67831ce Mon Sep 17 00:00:00 2001 From: maxvostrikov Date: Fri, 8 Nov 2024 13:11:36 +0100 Subject: [PATCH 567/680] squash! Missing tests in several tests in 24.10 Added corner cases for tests for: to_utc_timestamp and from_utc_timestamp (more timezones, spetial timezones, epoch corners does not look right, raising a bug over that) arrayUnion (empty and big arrays) quantilesExactWeightedInterpolated (more data types) --- tests/queries/0_stateless/02812_from_to_utc_timestamp.sh | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh b/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh index 20ae224332c..9eb4484ace0 100755 --- a/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh +++ b/tests/queries/0_stateless/02812_from_to_utc_timestamp.sh @@ -16,6 +16,7 @@ $CLICKHOUSE_CLIENT -q "select x, to_utc_timestamp(toDateTime('2023-03-16 11:22:3 $CLICKHOUSE_CLIENT -q "select to_utc_timestamp(toDateTime('2024-02-24 11:22:33'), 'Europe/Madrid'), from_utc_timestamp(toDateTime('2024-02-24 11:22:33'), 'Europe/Madrid')" $CLICKHOUSE_CLIENT -q "select to_utc_timestamp(toDateTime('2024-10-24 11:22:33'), 'Europe/Madrid'), from_utc_timestamp(toDateTime('2024-10-24 11:22:33'), 'Europe/Madrid')" $CLICKHOUSE_CLIENT -q "select to_utc_timestamp(toDateTime('2024-10-24 11:22:33'), 'EST'), from_utc_timestamp(toDateTime('2024-10-24 11:22:33'), 'EST')" + $CLICKHOUSE_CLIENT -q "select 'leap year:', to_utc_timestamp(toDateTime('2024-02-29 11:22:33'), 'EST'), from_utc_timestamp(toDateTime('2024-02-29 11:22:33'), 'EST')" $CLICKHOUSE_CLIENT -q "select 'non-leap year:', to_utc_timestamp(toDateTime('2023-02-29 11:22:33'), 'EST'), from_utc_timestamp(toDateTime('2023-02-29 11:22:33'), 'EST')" $CLICKHOUSE_CLIENT -q "select 'leap year:', to_utc_timestamp(toDateTime('2024-02-28 23:22:33'), 'EST'), from_utc_timestamp(toDateTime('2024-03-01 00:22:33'), 'EST')" From 6f74b3236bef52beed01aca5007dad13df7a5ae4 Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Fri, 8 Nov 2024 12:22:57 +0000 Subject: [PATCH 568/680] Fix some tests. --- src/Core/SettingsChangesHistory.cpp | 1 - src/Processors/QueryPlan/FilterStep.cpp | 4 +- .../01655_plan_optimizations.reference | 5 +-- .../0_stateless/01655_plan_optimizations.sh | 4 +- .../02496_remove_redundant_sorting.reference | 13 +++---- ...rouping_sets_predicate_push_down.reference | 38 ++++++++----------- 6 files changed, 27 insertions(+), 38 deletions(-) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index dedf8279533..8f01bacf254 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -75,7 +75,6 @@ static std::initializer_listresult_name; auto split_result = dag.split({filter_node}, true); dag = std::move(split_result.second); @@ -57,10 +56,11 @@ static ActionsAndName splitSingleAndFilter(ActionsDAG & dag, const ActionsDAG::N if (filter_type->isNullable()) cast_type = std::make_shared(std::move(cast_type)); - split_result.first.addCast(*split_filter_node, cast_type, {}); + split_filter_node = &split_result.first.addCast(*split_filter_node, cast_type, {}); } split_result.first.getOutputs().emplace(split_result.first.getOutputs().begin(), split_filter_node); + auto name = split_filter_node->result_name; return ActionsAndName{std::move(split_result.first), std::move(name)}; } diff --git a/tests/queries/0_stateless/01655_plan_optimizations.reference b/tests/queries/0_stateless/01655_plan_optimizations.reference index edf93b4b39f..7fc7556e85b 100644 --- a/tests/queries/0_stateless/01655_plan_optimizations.reference +++ b/tests/queries/0_stateless/01655_plan_optimizations.reference @@ -82,12 +82,12 @@ Filter column: notEquals(__table1.y, 0_UInt8) 9 10 > one condition of filter should be pushed down after aggregating, other two conditions are ANDed Filter column -FUNCTION and(minus(s, 8) :: 5, minus(s, 4) :: 2) -> and(notEquals(y, 0), minus(s, 8), minus(s, 4)) +FUNCTION and(minus(s, 8) :: 3, minus(s, 4) :: 5) -> and(notEquals(y, 0), minus(s, 8), minus(s, 4)) Aggregating Filter column: notEquals(y, 0) > (analyzer) one condition of filter should be pushed down after aggregating, other two conditions are ANDed Filter column -FUNCTION and(minus(__table1.s, 8_UInt8) :: 1, minus(__table1.s, 4_UInt8) :: 2) -> and(notEquals(__table1.y, 0_UInt8), minus(__table1.s, 8_UInt8), minus(__table1.s, 4_UInt8)) +FUNCTION and(minus(__table1.s, 8_UInt8) :: 3, minus(__table1.s, 4_UInt8) :: 5) -> and(notEquals(__table1.y, 0_UInt8), minus(__table1.s, 8_UInt8), minus(__table1.s, 4_UInt8)) Aggregating Filter column: notEquals(__table1.y, 0_UInt8) 0 1 @@ -163,7 +163,6 @@ Filter column: notEquals(__table1.y, 2_UInt8) > filter is pushed down before CreatingSets CreatingSets Filter -Filter 1 3 > one condition of filter is pushed down before LEFT JOIN diff --git a/tests/queries/0_stateless/01655_plan_optimizations.sh b/tests/queries/0_stateless/01655_plan_optimizations.sh index 42cdac8c01f..04ab9bbd11c 100755 --- a/tests/queries/0_stateless/01655_plan_optimizations.sh +++ b/tests/queries/0_stateless/01655_plan_optimizations.sh @@ -89,14 +89,14 @@ $CLICKHOUSE_CLIENT --enable_analyzer=0 --convert_query_to_cnf=0 -q " select sum(x) as s, y from (select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 and s - 8 and s - 4 settings enable_optimize_predicate_expression=0" | - grep -o "Aggregating\|Filter column\|Filter column: notEquals(y, 0)\|FUNCTION and(minus(s, 8) :: 5, minus(s, 4) :: 2) -> and(notEquals(y, 0), minus(s, 8), minus(s, 4))" + grep -o "Aggregating\|Filter column\|Filter column: notEquals(y, 0)\|FUNCTION and(minus(s, 8) :: 3, minus(s, 4) :: 5) -> and(notEquals(y, 0), minus(s, 8), minus(s, 4))" echo "> (analyzer) one condition of filter should be pushed down after aggregating, other two conditions are ANDed" $CLICKHOUSE_CLIENT --enable_analyzer=1 --convert_query_to_cnf=0 -q " explain actions = 1 select s, y from ( select sum(x) as s, y from (select number as x, number + 1 as y from numbers(10)) group by y ) where y != 0 and s - 8 and s - 4 settings enable_optimize_predicate_expression=0" | - grep -o "Aggregating\|Filter column\|Filter column: notEquals(__table1.y, 0_UInt8)\|FUNCTION and(minus(__table1.s, 8_UInt8) :: 1, minus(__table1.s, 4_UInt8) :: 2) -> and(notEquals(__table1.y, 0_UInt8), minus(__table1.s, 8_UInt8), minus(__table1.s, 4_UInt8))" + grep -o "Aggregating\|Filter column\|Filter column: notEquals(__table1.y, 0_UInt8)\|FUNCTION and(minus(__table1.s, 8_UInt8) :: 3, minus(__table1.s, 4_UInt8) :: 5) -> and(notEquals(__table1.y, 0_UInt8), minus(__table1.s, 8_UInt8), minus(__table1.s, 4_UInt8))" $CLICKHOUSE_CLIENT -q " select s, y from ( select sum(x) as s, y from (select number as x, number + 1 as y from numbers(10)) group by y diff --git a/tests/queries/0_stateless/02496_remove_redundant_sorting.reference b/tests/queries/0_stateless/02496_remove_redundant_sorting.reference index 7824fd8cba9..00db41e8ac5 100644 --- a/tests/queries/0_stateless/02496_remove_redundant_sorting.reference +++ b/tests/queries/0_stateless/02496_remove_redundant_sorting.reference @@ -332,13 +332,12 @@ SETTINGS optimize_aggregators_of_group_by_keys=0 -- avoid removing any() as it d Expression (Projection) Sorting (Sorting for ORDER BY) Expression (Before ORDER BY) - Filter ((WHERE + (Projection + Before ORDER BY))) - Filter (HAVING) - Aggregating - Expression ((Before GROUP BY + Projection)) - Sorting (Sorting for ORDER BY) - Expression ((Before ORDER BY + (Projection + Before ORDER BY))) - ReadFromSystemNumbers + Filter (((WHERE + (Projection + Before ORDER BY)) + HAVING)) + Aggregating + Expression ((Before GROUP BY + Projection)) + Sorting (Sorting for ORDER BY) + Expression ((Before ORDER BY + (Projection + Before ORDER BY))) + ReadFromSystemNumbers -- execute 1 2 diff --git a/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.reference b/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.reference index 9bb0c022752..a382e14ce03 100644 --- a/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.reference +++ b/tests/queries/0_stateless/02554_fix_grouping_sets_predicate_push_down.reference @@ -28,21 +28,17 @@ WHERE type_1 = \'all\' (Expression) ExpressionTransform × 2 (Filter) - FilterTransform × 2 - (Filter) - FilterTransform × 2 - (Filter) - FilterTransform × 2 - (Aggregating) - ExpressionTransform × 2 - AggregatingTransform × 2 - Copy 1 → 2 - (Expression) - ExpressionTransform - (Expression) - ExpressionTransform - (ReadFromMergeTree) - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + FilterTransform × 6 + (Aggregating) + ExpressionTransform × 2 + AggregatingTransform × 2 + Copy 1 → 2 + (Expression) + ExpressionTransform + (Expression) + ExpressionTransform + (ReadFromMergeTree) + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 (Expression) ExpressionTransform × 2 (Filter) @@ -68,14 +64,10 @@ ExpressionTransform × 2 ExpressionTransform × 2 AggregatingTransform × 2 Copy 1 → 2 - (Filter) - FilterTransform - (Filter) - FilterTransform - (Expression) - ExpressionTransform - (ReadFromMergeTree) - MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 + (Expression) + ExpressionTransform + (ReadFromMergeTree) + MergeTreeSelect(pool: ReadPoolInOrder, algorithm: InOrder) 0 → 1 (Expression) ExpressionTransform × 2 (Aggregating) From da0e267278efa2f42e0f18bf5a4b78a5d16dbe99 Mon Sep 17 00:00:00 2001 From: Pavel Kruglov <48961922+Avogar@users.noreply.github.com> Date: Fri, 8 Nov 2024 13:30:21 +0100 Subject: [PATCH 569/680] Fix typo --- .../queries/0_stateless/03247_ghdata_string_to_json_alter.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh b/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh index a2d1788cb5d..e8368b6702a 100755 --- a/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh +++ b/tests/queries/0_stateless/03247_ghdata_string_to_json_alter.sh @@ -18,12 +18,12 @@ ${CLICKHOUSE_CLIENT} -q "SELECT count() FROM ghdata WHERE NOT ignore(*)" ${CLICKHOUSE_CLIENT} -q \ "SELECT data.repo.name, count() AS stars FROM ghdata \ - WHERE data.type = 'WatchEvent' GROUP BY data.repo.name ORDER BY stars DESC, data.repo.name LIMIT 5" --allow_suspicious_types_in_group_by=1, --allow_suspicious_types_in_order_by=1 + WHERE data.type = 'WatchEvent' GROUP BY data.repo.name ORDER BY stars DESC, data.repo.name LIMIT 5" --allow_suspicious_types_in_group_by=1 --allow_suspicious_types_in_order_by=1 ${CLICKHOUSE_CLIENT} --enable_analyzer=1 -q \ "SELECT data.payload.commits[].author.name AS name, count() AS c FROM ghdata \ ARRAY JOIN data.payload.commits[].author.name \ - GROUP BY name ORDER BY c DESC, name LIMIT 5" --allow_suspicious_types_in_group_by=1, --allow_suspicious_types_in_order_by=1 + GROUP BY name ORDER BY c DESC, name LIMIT 5" --allow_suspicious_types_in_group_by=1 --allow_suspicious_types_in_order_by=1 ${CLICKHOUSE_CLIENT} -q "SELECT max(data.payload.pull_request.assignees[].size0) FROM ghdata" From 955f537bd5ef2f4a29717ac4999ce2af47b4c039 Mon Sep 17 00:00:00 2001 From: Pablo Marcos Date: Fri, 8 Nov 2024 12:28:06 +0000 Subject: [PATCH 570/680] Add a new setting query_metric_log_debug to avoid the noise --- src/Core/Settings.cpp | 5 +++++ src/Core/SettingsChangesHistory.cpp | 1 + src/Interpreters/QueryMetricLog.cpp | 18 +++++++++++------- src/Interpreters/QueryMetricLog.h | 3 ++- src/Interpreters/executeQuery.cpp | 3 ++- .../03203_system_query_metric_log.sh | 10 +++++----- 6 files changed, 26 insertions(+), 14 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 081e07ca2ce..d07cd7352a1 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -2784,6 +2784,11 @@ If set to any negative value, it will take the value `collect_interval_milliseco To disable the collection of a single query, set `query_metric_log_interval` to 0. Default value: -1 + )", 0) \ + DECLARE(Bool, query_metric_log_debug, false, R"( +Turns on debugging traces for system.query_metric_log + +Default value: false )", 0) \ DECLARE(LogsLevel, send_logs_level, LogsLevel::fatal, R"( Send server text logs with specified minimum level to client. Valid values: 'trace', 'debug', 'information', 'warning', 'error', 'fatal', 'none' diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index ed87fde8b7e..a3e21aa670f 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -75,6 +75,7 @@ static std::initializer_listgetProcessList(); @@ -222,21 +223,24 @@ void QueryMetricLogStatus::scheduleNext(String query_id) } else { - LOG_TRACE(logger, "The next collecting task for query {} should have already run at {}. Scheduling it right now", - query_id, timePointToString(next_collect_time)); + if (debug_traces) + LOG_DEBUG(logger, "The next collecting task for query {} should have already run at {}. Scheduling it right now", + query_id, timePointToString(next_collect_time)); task->schedule(); } } std::optional QueryMetricLogStatus::createLogMetricElement(const String & query_id, const QueryStatusInfo & query_info, TimePoint query_info_time, bool schedule_next) { - LOG_TRACE(logger, "Collecting query_metric_log for query {} and interval {} ms with QueryStatusInfo from {}. Next collection time: {}", - query_id, interval_milliseconds, timePointToString(query_info_time), - schedule_next ? timePointToString(next_collect_time + std::chrono::milliseconds(interval_milliseconds)) : "finished"); + if (debug_traces) + LOG_DEBUG(logger, "Collecting query_metric_log for query {} and interval {} ms with QueryStatusInfo from {}. Next collection time: {}", + query_id, interval_milliseconds, timePointToString(query_info_time), + schedule_next ? timePointToString(next_collect_time + std::chrono::milliseconds(interval_milliseconds)) : "finished"); if (query_info_time <= last_collect_time) { - LOG_TRACE(logger, "Query {} has a more recent metrics collected. Skipping this one", query_id); + if (debug_traces) + LOG_DEBUG(logger, "Query {} has a more recent metrics collected. Skipping this one", query_id); return {}; } diff --git a/src/Interpreters/QueryMetricLog.h b/src/Interpreters/QueryMetricLog.h index 65764229b0a..5f301b2cd13 100644 --- a/src/Interpreters/QueryMetricLog.h +++ b/src/Interpreters/QueryMetricLog.h @@ -51,6 +51,7 @@ struct QueryMetricLogStatus std::chrono::system_clock::time_point next_collect_time TSA_GUARDED_BY(getMutex()); std::vector last_profile_events TSA_GUARDED_BY(getMutex()) = std::vector(ProfileEvents::end()); BackgroundSchedulePool::TaskHolder task TSA_GUARDED_BY(getMutex()); + bool debug_traces = false; /// We need to be able to move it for the hash map, so we need to add an indirection here. std::unique_ptr mutex = std::make_unique(); @@ -78,7 +79,7 @@ public: void shutdown() final; /// Both startQuery and finishQuery are called from the thread that executes the query. - void startQuery(const String & query_id, TimePoint start_time, UInt64 interval_milliseconds); + void startQuery(const String & query_id, TimePoint start_time, UInt64 interval_milliseconds, bool debug_traces = false); void finishQuery(const String & query_id, TimePoint finish_time, QueryStatusInfoPtr query_info = nullptr); private: diff --git a/src/Interpreters/executeQuery.cpp b/src/Interpreters/executeQuery.cpp index 4507126b7b3..794d3dab0e6 100644 --- a/src/Interpreters/executeQuery.cpp +++ b/src/Interpreters/executeQuery.cpp @@ -146,6 +146,7 @@ namespace Setting extern const SettingsQueryCacheSystemTableHandling query_cache_system_table_handling; extern const SettingsSeconds query_cache_ttl; extern const SettingsInt64 query_metric_log_interval; + extern const SettingsBool query_metric_log_debug; extern const SettingsOverflowMode read_overflow_mode; extern const SettingsOverflowMode read_overflow_mode_leaf; extern const SettingsOverflowMode result_overflow_mode; @@ -455,7 +456,7 @@ QueryLogElement logQueryStart( { auto interval_milliseconds = getQueryMetricLogInterval(context); if (interval_milliseconds > 0) - query_metric_log->startQuery(elem.client_info.current_query_id, query_start_time, interval_milliseconds); + query_metric_log->startQuery(elem.client_info.current_query_id, query_start_time, interval_milliseconds, settings[Setting::query_metric_log_debug]); } return elem; diff --git a/tests/queries/0_stateless/03203_system_query_metric_log.sh b/tests/queries/0_stateless/03203_system_query_metric_log.sh index abcd14c8e5d..4bc764b777c 100755 --- a/tests/queries/0_stateless/03203_system_query_metric_log.sh +++ b/tests/queries/0_stateless/03203_system_query_metric_log.sh @@ -6,11 +6,11 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) readonly query_prefix=$CLICKHOUSE_DATABASE -$CLICKHOUSE_CLIENT --query-id="${query_prefix}_1000" -q "SELECT sleep(2.5) FORMAT Null" & -$CLICKHOUSE_CLIENT --query-id="${query_prefix}_400" -q "SELECT sleep(2.5) SETTINGS query_metric_log_interval=400 FORMAT Null" & -$CLICKHOUSE_CLIENT --query-id="${query_prefix}_123" -q "SELECT sleep(2.5) SETTINGS query_metric_log_interval=123 FORMAT Null" & -$CLICKHOUSE_CLIENT --query-id="${query_prefix}_0" -q "SELECT sleep(2.5) SETTINGS query_metric_log_interval=0 FORMAT Null" & -$CLICKHOUSE_CLIENT --query-id="${query_prefix}_fast" -q "SELECT sleep(0.1) FORMAT Null" & +$CLICKHOUSE_CLIENT --query-id="${query_prefix}_1000" -q "SELECT sleep(2.5) SETTINGS query_metric_log_debug=true FORMAT Null" & +$CLICKHOUSE_CLIENT --query-id="${query_prefix}_400" -q "SELECT sleep(2.5) SETTINGS query_metric_log_debug=true, query_metric_log_interval=400 FORMAT Null" & +$CLICKHOUSE_CLIENT --query-id="${query_prefix}_123" -q "SELECT sleep(2.5) SETTINGS query_metric_log_debug=true, query_metric_log_interval=123 FORMAT Null" & +$CLICKHOUSE_CLIENT --query-id="${query_prefix}_0" -q "SELECT sleep(2.5) SETTINGS query_metric_log_debug=true, query_metric_log_interval=0 FORMAT Null" & +$CLICKHOUSE_CLIENT --query-id="${query_prefix}_fast" -q "SELECT sleep(0.1) SETTINGS query_metric_log_debug=true FORMAT Null" & wait From fd9f32708371246e36b289164cf402230bc860c6 Mon Sep 17 00:00:00 2001 From: kssenii Date: Fri, 8 Nov 2024 13:49:08 +0100 Subject: [PATCH 571/680] Allow to disable memory buffer increase for filesystem cache --- src/Core/Settings.cpp | 3 +++ src/Disks/ObjectStorages/DiskObjectStorage.cpp | 5 ++++- src/IO/ReadSettings.h | 1 + src/Interpreters/Context.cpp | 2 ++ src/Storages/ObjectStorage/StorageObjectStorageSource.cpp | 2 +- 5 files changed, 11 insertions(+), 2 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 6f0109fa300..9a821879c5b 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -4872,6 +4872,9 @@ Limit on size of a single batch of file segments that a read buffer can request )", 0) \ DECLARE(UInt64, filesystem_cache_reserve_space_wait_lock_timeout_milliseconds, 1000, R"( Wait time to lock cache for space reservation in filesystem cache +)", 0) \ + DECLARE(Bool, filesystem_cache_prefer_bigger_buffer_size, true, R"( +Prefer bigger buffer size if filesystem cache is enabled to avoid writing small file segments which detiriorate cache performance )", 0) \ DECLARE(UInt64, temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds, (10 * 60 * 1000), R"( Wait time to lock cache for space reservation for temporary data in filesystem cache diff --git a/src/Disks/ObjectStorages/DiskObjectStorage.cpp b/src/Disks/ObjectStorages/DiskObjectStorage.cpp index 3720c04a471..fba45d5a0c9 100644 --- a/src/Disks/ObjectStorages/DiskObjectStorage.cpp +++ b/src/Disks/ObjectStorages/DiskObjectStorage.cpp @@ -642,7 +642,10 @@ std::unique_ptr DiskObjectStorage::readFile( }; /// Avoid cache fragmentation by choosing bigger buffer size. - bool prefer_bigger_buffer_size = object_storage->supportsCache() && read_settings.enable_filesystem_cache; + bool prefer_bigger_buffer_size = read_settings.filesystem_cache_prefer_bigger_buffer_size + && object_storage->supportsCache() + && read_settings.enable_filesystem_cache; + size_t buffer_size = prefer_bigger_buffer_size ? std::max(settings.remote_fs_buffer_size, DBMS_DEFAULT_BUFFER_SIZE) : settings.remote_fs_buffer_size; diff --git a/src/IO/ReadSettings.h b/src/IO/ReadSettings.h index 6ed02212095..c1747314c76 100644 --- a/src/IO/ReadSettings.h +++ b/src/IO/ReadSettings.h @@ -61,6 +61,7 @@ struct ReadSettings bool filesystem_cache_allow_background_download = true; bool filesystem_cache_allow_background_download_for_metadata_files_in_packed_storage = true; bool filesystem_cache_allow_background_download_during_fetch = true; + bool filesystem_cache_prefer_bigger_buffer_size = true; bool use_page_cache_for_disks_without_file_cache = false; bool read_from_page_cache_if_exists_otherwise_bypass_cache = false; diff --git a/src/Interpreters/Context.cpp b/src/Interpreters/Context.cpp index c1fa2c8549a..d42002bf98d 100644 --- a/src/Interpreters/Context.cpp +++ b/src/Interpreters/Context.cpp @@ -196,6 +196,7 @@ namespace Setting extern const SettingsUInt64 filesystem_cache_segments_batch_size; extern const SettingsBool filesystem_cache_enable_background_download_for_metadata_files_in_packed_storage; extern const SettingsBool filesystem_cache_enable_background_download_during_fetch; + extern const SettingsBool filesystem_cache_prefer_bigger_buffer_size; extern const SettingsBool http_make_head_request; extern const SettingsUInt64 http_max_fields; extern const SettingsUInt64 http_max_field_name_size; @@ -5751,6 +5752,7 @@ ReadSettings Context::getReadSettings() const res.filesystem_cache_allow_background_download_for_metadata_files_in_packed_storage = settings_ref[Setting::filesystem_cache_enable_background_download_for_metadata_files_in_packed_storage]; res.filesystem_cache_allow_background_download_during_fetch = settings_ref[Setting::filesystem_cache_enable_background_download_during_fetch]; + res.filesystem_cache_prefer_bigger_buffer_size = settings_ref[Setting::filesystem_cache_prefer_bigger_buffer_size]; res.filesystem_cache_max_download_size = settings_ref[Setting::filesystem_cache_max_download_size]; res.skip_download_if_exceeds_query_cache = settings_ref[Setting::skip_download_if_exceeds_query_cache]; diff --git a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp index 563bdc44760..1ccf23ade90 100644 --- a/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp +++ b/src/Storages/ObjectStorage/StorageObjectStorageSource.cpp @@ -517,7 +517,7 @@ std::unique_ptr StorageObjectStorageSource::createReadBu LOG_TRACE(log, "Downloading object of size {} with initial prefetch", object_size); - bool prefer_bigger_buffer_size = impl->isCached(); + bool prefer_bigger_buffer_size = read_settings.filesystem_cache_prefer_bigger_buffer_size && impl->isCached(); size_t buffer_size = prefer_bigger_buffer_size ? std::max(read_settings.remote_fs_buffer_size, DBMS_DEFAULT_BUFFER_SIZE) : read_settings.remote_fs_buffer_size; From fe73c1880a67340b8eea8c7d27a4f0a58aa42cd9 Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Fri, 8 Nov 2024 14:06:59 +0100 Subject: [PATCH 572/680] Update src/Core/Settings.cpp Co-authored-by: Nikita Taranov --- src/Core/Settings.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 9a821879c5b..8feb758df0f 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -4874,7 +4874,7 @@ Limit on size of a single batch of file segments that a read buffer can request Wait time to lock cache for space reservation in filesystem cache )", 0) \ DECLARE(Bool, filesystem_cache_prefer_bigger_buffer_size, true, R"( -Prefer bigger buffer size if filesystem cache is enabled to avoid writing small file segments which detiriorate cache performance +Prefer bigger buffer size if filesystem cache is enabled to avoid writing small file segments which deteriorate cache performance )", 0) \ DECLARE(UInt64, temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds, (10 * 60 * 1000), R"( Wait time to lock cache for space reservation for temporary data in filesystem cache From 298b172c49493a88c87fde4d0e09a6413102de55 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Thu, 7 Nov 2024 22:30:45 +0100 Subject: [PATCH 573/680] Add fallback to getgrgid_r and getpwuid_r for UID and GID arguments of clickhouse-su --- programs/su/su.cpp | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) diff --git a/programs/su/su.cpp b/programs/su/su.cpp index 33d929898f4..40242d0687f 100644 --- a/programs/su/su.cpp +++ b/programs/su/su.cpp @@ -59,7 +59,13 @@ void setUserAndGroup(std::string arg_uid, std::string arg_gid) throw ErrnoException(ErrorCodes::SYSTEM_ERROR, "Cannot do 'getgrnam_r' to obtain gid from group name ({})", arg_gid); if (!result) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "Group {} is not found in the system", arg_gid); + { + if (0 != getgrgid_r(gid, &entry, buf.get(), buf_size, &result)) + throw ErrnoException(ErrorCodes::SYSTEM_ERROR, "Cannot do 'getgrnam_r' to obtain gid from group name ({})", arg_gid); + + if (!result) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "Group {} is not found in the system", arg_gid); + } gid = entry.gr_gid; } @@ -84,7 +90,13 @@ void setUserAndGroup(std::string arg_uid, std::string arg_gid) throw ErrnoException(ErrorCodes::SYSTEM_ERROR, "Cannot do 'getpwnam_r' to obtain uid from user name ({})", arg_uid); if (!result) - throw Exception(ErrorCodes::BAD_ARGUMENTS, "User {} is not found in the system", arg_uid); + { + if (0 != getpwuid_r(uid, &entry, buf.get(), buf_size, &result)) + throw ErrnoException(ErrorCodes::SYSTEM_ERROR, "Cannot do 'getpwuid_r' to obtain uid from user name ({})", uid); + + if (!result) + throw Exception(ErrorCodes::BAD_ARGUMENTS, "User {} is not found in the system", arg_uid); + } uid = entry.pw_uid; } From 69ae05210364cf03bddf62b13bd752857bcbbedc Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Fri, 8 Nov 2024 10:22:01 +0000 Subject: [PATCH 574/680] SimSIMD: Improve suppression for msan false positive --- contrib/SimSIMD | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/SimSIMD b/contrib/SimSIMD index ee3c9c9c00b..9e3cfc32d26 160000 --- a/contrib/SimSIMD +++ b/contrib/SimSIMD @@ -1 +1 @@ -Subproject commit ee3c9c9c00b51645f62a1a9e99611b78c0052a21 +Subproject commit 9e3cfc32d26fbeece91e34df8668db28c0ca006a From ba20032987a042d45b4073e93eb5279222aff4ac Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Fri, 8 Nov 2024 14:08:36 +0000 Subject: [PATCH 575/680] Fix build --- contrib/SimSIMD | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/SimSIMD b/contrib/SimSIMD index 9e3cfc32d26..bb0bd2e7137 160000 --- a/contrib/SimSIMD +++ b/contrib/SimSIMD @@ -1 +1 @@ -Subproject commit 9e3cfc32d26fbeece91e34df8668db28c0ca006a +Subproject commit bb0bd2e7137f02c555341d7c93124ed19f3c24fb From aeb2cbf934c76d082b01dc023b28562efb5d6e02 Mon Sep 17 00:00:00 2001 From: Kseniia Sumarokova <54203879+kssenii@users.noreply.github.com> Date: Fri, 8 Nov 2024 15:26:41 +0100 Subject: [PATCH 576/680] Update Settings.cpp --- src/Core/Settings.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 8feb758df0f..07a2c52d72f 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -4874,7 +4874,7 @@ Limit on size of a single batch of file segments that a read buffer can request Wait time to lock cache for space reservation in filesystem cache )", 0) \ DECLARE(Bool, filesystem_cache_prefer_bigger_buffer_size, true, R"( -Prefer bigger buffer size if filesystem cache is enabled to avoid writing small file segments which deteriorate cache performance +Prefer bigger buffer size if filesystem cache is enabled to avoid writing small file segments which deteriorate cache performance. On the other hand, enabling this setting might increase memory usage. )", 0) \ DECLARE(UInt64, temporary_data_in_cache_reserve_space_wait_lock_timeout_milliseconds, (10 * 60 * 1000), R"( Wait time to lock cache for space reservation for temporary data in filesystem cache From 0929f66516261dea7b31479b8f5eaac1b4b8e38a Mon Sep 17 00:00:00 2001 From: kssenii Date: Fri, 8 Nov 2024 17:01:09 +0100 Subject: [PATCH 577/680] Update test --- tests/integration/test_storage_s3_queue/test.py | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tests/integration/test_storage_s3_queue/test.py b/tests/integration/test_storage_s3_queue/test.py index 284b304c632..62afc0b1c1d 100644 --- a/tests/integration/test_storage_s3_queue/test.py +++ b/tests/integration/test_storage_s3_queue/test.py @@ -1000,6 +1000,9 @@ def test_max_set_age(started_cluster): assert "Cannot parse input" in node.query( f"SELECT exception FROM system.s3queue WHERE file_name ilike '%{file_with_error}'" ) + assert "Cannot parse input" in node.query( + f"SELECT exception FROM system.s3queue_log WHERE file_name ilike '%{file_with_error}' ORDER BY processing_end_time DESC LIMIT 1" + ) assert 1 == int( node.query( From 5d2e1547a89cb43d545c5847cf47565f319bbd75 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Fri, 8 Nov 2024 11:34:06 -0500 Subject: [PATCH 578/680] use `/var/log/mysql/` instead of `/mysql/` --- tests/integration/compose/docker_compose_mysql.yml | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/integration/compose/docker_compose_mysql.yml b/tests/integration/compose/docker_compose_mysql.yml index f45410bde78..91df21165ea 100644 --- a/tests/integration/compose/docker_compose_mysql.yml +++ b/tests/integration/compose/docker_compose_mysql.yml @@ -5,7 +5,7 @@ services: environment: MYSQL_ROOT_PASSWORD: clickhouse MYSQL_ROOT_HOST: ${MYSQL_ROOT_HOST} - DATADIR: /mysql/ + DATADIR: /var/log/mysql/ expose: - ${MYSQL_PORT:-3306} command: --server_id=100 @@ -14,11 +14,11 @@ services: --gtid-mode="ON" --enforce-gtid-consistency --log-error-verbosity=3 - --log-error=/mysql/error.log + --log-error=/var/log/mysql/error.log --general-log=ON - --general-log-file=/mysql/general.log + --general-log-file=/var/log/mysql/general.log volumes: - type: ${MYSQL_LOGS_FS:-tmpfs} source: ${MYSQL_LOGS:-} - target: /mysql/ + target: /var/log/mysql/ user: ${MYSQL_DOCKER_USER} From 97ec890b8e3e06d9914e573363965ef439f76d21 Mon Sep 17 00:00:00 2001 From: Yakov Olkhovskiy <99031427+yakov-olkhovskiy@users.noreply.github.com> Date: Fri, 8 Nov 2024 11:36:36 -0500 Subject: [PATCH 579/680] use `/var/log/mysql/` instead of `/mysql/`, fix `MYSQL_ROOT_HOST` env initialization --- tests/integration/compose/docker_compose_mysql_8_0.yml | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/tests/integration/compose/docker_compose_mysql_8_0.yml b/tests/integration/compose/docker_compose_mysql_8_0.yml index e1ff1633bc7..e1e2e241443 100644 --- a/tests/integration/compose/docker_compose_mysql_8_0.yml +++ b/tests/integration/compose/docker_compose_mysql_8_0.yml @@ -4,8 +4,8 @@ services: restart: always environment: MYSQL_ROOT_PASSWORD: clickhouse - MYSQL_ROOT_HOST: ${MYSQL_ROOT_HOST} - DATADIR: /mysql/ + MYSQL_ROOT_HOST: ${MYSQL8_ROOT_HOST} + DATADIR: /var/log/mysql/ expose: - ${MYSQL8_PORT:-3306} command: --server_id=100 --log-bin='mysql-bin-1.log' @@ -13,11 +13,11 @@ services: --default-time-zone='+3:00' --gtid-mode="ON" --enforce-gtid-consistency --log-error-verbosity=3 - --log-error=/mysql/error.log + --log-error=/var/log/mysql/error.log --general-log=ON - --general-log-file=/mysql/general.log + --general-log-file=/var/log/mysql/general.log volumes: - type: ${MYSQL8_LOGS_FS:-tmpfs} source: ${MYSQL8_LOGS:-} - target: /mysql/ + target: /var/log/mysql/ user: ${MYSQL8_DOCKER_USER} From fe39c4b65bfee09d9c7d5327963983fbd4cdd234 Mon Sep 17 00:00:00 2001 From: Tanya Bragin Date: Fri, 8 Nov 2024 08:55:20 -0800 Subject: [PATCH 580/680] Update README.md - Update meetups Add Stockholm --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index dcaeda13acd..abaf27abf11 100644 --- a/README.md +++ b/README.md @@ -47,6 +47,7 @@ Upcoming meetups * [Dubai Meetup](https://www.meetup.com/clickhouse-dubai-meetup-group/events/303096989/) - November 21 * [Paris Meetup](https://www.meetup.com/clickhouse-france-user-group/events/303096434) - November 26 * [Amsterdam Meetup](https://www.meetup.com/clickhouse-netherlands-user-group/events/303638814) - December 3 +* [Stockholm Meetup](https://www.meetup.com/clickhouse-stockholm-user-group/events/304382411) - December 9 * [New York Meetup](https://www.meetup.com/clickhouse-new-york-user-group/events/304268174) - December 9 * [San Francisco Meetup](https://www.meetup.com/clickhouse-silicon-valley-meetup-group/events/304286951/) - December 12 From 9dc4046b897bd7cd185c0f5e0e221dea7481f8a9 Mon Sep 17 00:00:00 2001 From: alesapin Date: Fri, 8 Nov 2024 18:02:41 +0100 Subject: [PATCH 581/680] Add index granularity size column to system.parts --- src/Storages/MergeTree/IMergeTreeDataPart.cpp | 9 +++++++++ src/Storages/MergeTree/IMergeTreeDataPart.h | 2 ++ .../MergeTree/MergeTreeIndexGranularity.cpp | 10 ++++++++++ src/Storages/MergeTree/MergeTreeIndexGranularity.h | 3 +++ src/Storages/System/StorageSystemParts.cpp | 6 ++++++ .../03268_system_parts_index_granularity.reference | 1 + .../03268_system_parts_index_granularity.sql | 14 ++++++++++++++ 7 files changed, 45 insertions(+) create mode 100644 tests/queries/0_stateless/03268_system_parts_index_granularity.reference create mode 100644 tests/queries/0_stateless/03268_system_parts_index_granularity.sql diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.cpp b/src/Storages/MergeTree/IMergeTreeDataPart.cpp index 7453d609fa9..51c445945e6 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.cpp +++ b/src/Storages/MergeTree/IMergeTreeDataPart.cpp @@ -624,6 +624,15 @@ UInt64 IMergeTreeDataPart::getIndexSizeInAllocatedBytes() const return res; } +UInt64 IMergeTreeDataPart::getIndexGranularityBytes() const +{ + return index_granularity.getBytesSize(); +} +UInt64 IMergeTreeDataPart::getIndexGranularityAllocatedBytes() const +{ + return index_granularity.getBytesAllocated(); +} + void IMergeTreeDataPart::assertState(const std::initializer_list & affordable_states) const { if (!checkState(affordable_states)) diff --git a/src/Storages/MergeTree/IMergeTreeDataPart.h b/src/Storages/MergeTree/IMergeTreeDataPart.h index b41a1d840e1..55f1265318c 100644 --- a/src/Storages/MergeTree/IMergeTreeDataPart.h +++ b/src/Storages/MergeTree/IMergeTreeDataPart.h @@ -380,6 +380,8 @@ public: /// For data in RAM ('index') UInt64 getIndexSizeInBytes() const; UInt64 getIndexSizeInAllocatedBytes() const; + UInt64 getIndexGranularityBytes() const; + UInt64 getIndexGranularityAllocatedBytes() const; UInt64 getMarksCount() const; UInt64 getIndexSizeFromFile() const; diff --git a/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp b/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp index c3e740bde84..bf0ba17d473 100644 --- a/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp +++ b/src/Storages/MergeTree/MergeTreeIndexGranularity.cpp @@ -128,4 +128,14 @@ void MergeTreeIndexGranularity::shrinkToFitInMemory() marks_rows_partial_sums.shrink_to_fit(); } +uint64_t MergeTreeIndexGranularity::getBytesSize() const +{ + return marks_rows_partial_sums.size() * sizeof(size_t); +} +uint64_t MergeTreeIndexGranularity::getBytesAllocated() const +{ + return marks_rows_partial_sums.capacity() * sizeof(size_t); +} + + } diff --git a/src/Storages/MergeTree/MergeTreeIndexGranularity.h b/src/Storages/MergeTree/MergeTreeIndexGranularity.h index 9b8375dd2d8..c616d2ac49a 100644 --- a/src/Storages/MergeTree/MergeTreeIndexGranularity.h +++ b/src/Storages/MergeTree/MergeTreeIndexGranularity.h @@ -102,6 +102,9 @@ public: std::string describe() const; void shrinkToFitInMemory(); + + uint64_t getBytesSize() const; + uint64_t getBytesAllocated() const; }; } diff --git a/src/Storages/System/StorageSystemParts.cpp b/src/Storages/System/StorageSystemParts.cpp index 56a45d7b51d..d0e34842198 100644 --- a/src/Storages/System/StorageSystemParts.cpp +++ b/src/Storages/System/StorageSystemParts.cpp @@ -75,6 +75,8 @@ StorageSystemParts::StorageSystemParts(const StorageID & table_id_) {"data_version", std::make_shared(), "Number that is used to determine which mutations should be applied to the data part (mutations with a version higher than data_version)."}, {"primary_key_bytes_in_memory", std::make_shared(), "The amount of memory (in bytes) used by primary key values."}, {"primary_key_bytes_in_memory_allocated", std::make_shared(), "The amount of memory (in bytes) reserved for primary key values."}, + {"index_granularity_bytes_in_memory", std::make_shared(), "The amount of memory (in bytes) used by index granularity values."}, + {"index_granularity_bytes_in_memory_allocated", std::make_shared(), "The amount of memory (in bytes) reserved for index granularity values."}, {"is_frozen", std::make_shared(), "Flag that shows that a partition data backup exists. 1, the backup exists. 0, the backup does not exist. "}, {"database", std::make_shared(), "Name of the database."}, @@ -216,6 +218,10 @@ void StorageSystemParts::processNextStorage( columns[res_index++]->insert(part->getIndexSizeInBytes()); if (columns_mask[src_index++]) columns[res_index++]->insert(part->getIndexSizeInAllocatedBytes()); + if (columns_mask[src_index++]) + columns[res_index++]->insert(part->getIndexGranularityBytes()); + if (columns_mask[src_index++]) + columns[res_index++]->insert(part->getIndexGranularityAllocatedBytes()); if (columns_mask[src_index++]) columns[res_index++]->insert(part->is_frozen.load(std::memory_order_relaxed)); diff --git a/tests/queries/0_stateless/03268_system_parts_index_granularity.reference b/tests/queries/0_stateless/03268_system_parts_index_granularity.reference new file mode 100644 index 00000000000..f301cd54ad2 --- /dev/null +++ b/tests/queries/0_stateless/03268_system_parts_index_granularity.reference @@ -0,0 +1 @@ +88 88 diff --git a/tests/queries/0_stateless/03268_system_parts_index_granularity.sql b/tests/queries/0_stateless/03268_system_parts_index_granularity.sql new file mode 100644 index 00000000000..009a15d0825 --- /dev/null +++ b/tests/queries/0_stateless/03268_system_parts_index_granularity.sql @@ -0,0 +1,14 @@ +DROP TABLE IF EXISTS t; + +CREATE TABLE t ( + key UInt64, + value String +) +ENGINE MergeTree() +ORDER by key SETTINGS index_granularity = 10, index_granularity_bytes = '1024K'; + +INSERT INTO t SELECT number, toString(number) FROM numbers(100); + +SELECT index_granularity_bytes_in_memory, index_granularity_bytes_in_memory_allocated FROM system.parts where table = 't' and database = currentDatabase(); + +DROP TABLE IF EXISTS t; From 6c223c92bd852b56c713aff768b07c4adb90d5dc Mon Sep 17 00:00:00 2001 From: alesapin Date: Fri, 8 Nov 2024 18:13:29 +0100 Subject: [PATCH 582/680] btter --- .../queries/0_stateless/03268_system_parts_index_granularity.sql | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/03268_system_parts_index_granularity.sql b/tests/queries/0_stateless/03268_system_parts_index_granularity.sql index 009a15d0825..1bab7840856 100644 --- a/tests/queries/0_stateless/03268_system_parts_index_granularity.sql +++ b/tests/queries/0_stateless/03268_system_parts_index_granularity.sql @@ -1,3 +1,4 @@ +-- Tags: no-random-settings, no-random-merge-tree-settings DROP TABLE IF EXISTS t; CREATE TABLE t ( From 6d2504662a45e0c35758698ec60ac265309c0f6b Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 01:01:46 +0100 Subject: [PATCH 583/680] Update tests --- tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql | 1 + tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql | 1 + tests/queries/0_stateless/01958_partial_hour_timezone.sql | 2 ++ tests/queries/0_stateless/02125_query_views_log.sql | 2 ++ 4 files changed, 6 insertions(+) diff --git a/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql b/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql index 0154265ef72..907a8283396 100644 --- a/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql +++ b/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql @@ -1,5 +1,6 @@ SET session_timezone = 'UTC'; -- disable timezone randomization SET enable_analyzer = 1; -- The old path formats the result with different whitespaces +SET output_format_pretty_highlight_digit_groups = 0; SELECT '-- Negative tests'; SELECT dateTimeToSnowflakeID(); -- {serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH} diff --git a/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql b/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql index 41e5beb9c16..1f62f3d36da 100644 --- a/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql +++ b/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql @@ -1,5 +1,6 @@ SET session_timezone = 'UTC'; -- disable timezone randomization SET enable_analyzer = 1; -- The old path formats the result with different whitespaces +SET output_format_pretty_highlight_digit_groups = 0; SELECT '-- Negative tests'; SELECT snowflakeIDToDateTime(); -- {serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH} diff --git a/tests/queries/0_stateless/01958_partial_hour_timezone.sql b/tests/queries/0_stateless/01958_partial_hour_timezone.sql index 26350e55620..b72adfd9d58 100644 --- a/tests/queries/0_stateless/01958_partial_hour_timezone.sql +++ b/tests/queries/0_stateless/01958_partial_hour_timezone.sql @@ -1,3 +1,5 @@ +SET output_format_pretty_highlight_digit_groups = 0; + -- Appeared in https://github.com/ClickHouse/ClickHouse/pull/26978#issuecomment-890889362 WITH toDateTime('1970-06-17 07:39:21', 'Africa/Monrovia') as t SELECT toUnixTimestamp(t), diff --git a/tests/queries/0_stateless/02125_query_views_log.sql b/tests/queries/0_stateless/02125_query_views_log.sql index ba50902ebea..96170efedd6 100644 --- a/tests/queries/0_stateless/02125_query_views_log.sql +++ b/tests/queries/0_stateless/02125_query_views_log.sql @@ -1,3 +1,5 @@ +SET output_format_pretty_highlight_digit_groups = 0; + drop table if exists src; drop table if exists dst; drop table if exists mv1; From 19ca58e95203ed2eed71a6ef0ab88677d9bb6b93 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 02:29:44 +0100 Subject: [PATCH 584/680] Fix #71677 --- src/Functions/nested.cpp | 19 +++++++++++++------ .../03268_nested_analyzer.reference | 3 +++ .../0_stateless/03268_nested_analyzer.sql | 16 ++++++++++++++++ 3 files changed, 32 insertions(+), 6 deletions(-) create mode 100644 tests/queries/0_stateless/03268_nested_analyzer.reference create mode 100644 tests/queries/0_stateless/03268_nested_analyzer.sql diff --git a/src/Functions/nested.cpp b/src/Functions/nested.cpp index 85c342b5e7c..29d99b8a6df 100644 --- a/src/Functions/nested.cpp +++ b/src/Functions/nested.cpp @@ -108,27 +108,29 @@ public: { size_t arguments_size = arguments.size(); - const auto * lhs_array = assert_cast(arguments.at(1).column.get()); + ColumnPtr first_array_materialized = arguments[1].column->convertToFullColumnIfConst(); + const ColumnArray & first_array = assert_cast(*first_array_materialized); Columns data_columns; data_columns.reserve(arguments_size); - data_columns.push_back(lhs_array->getDataPtr()); + data_columns.push_back(first_array.getDataPtr()); for (size_t i = 2; i < arguments_size; ++i) { - const auto * rhs_array = assert_cast(arguments[i].column.get()); + ColumnPtr other_array_materialized = arguments[i].column->convertToFullColumnIfConst(); + const ColumnArray & other_array = assert_cast(*other_array_materialized); - if (!lhs_array->hasEqualOffsets(*rhs_array)) + if (!first_array.hasEqualOffsets(other_array)) throw Exception(ErrorCodes::SIZES_OF_ARRAYS_DONT_MATCH, "The argument 2 and argument {} of function {} have different array offsets", i + 1, getName()); - data_columns.push_back(rhs_array->getDataPtr()); + data_columns.push_back(other_array.getDataPtr()); } auto tuple_column = ColumnTuple::create(std::move(data_columns)); - auto array_column = ColumnArray::create(std::move(tuple_column), lhs_array->getOffsetsPtr()); + auto array_column = ColumnArray::create(std::move(tuple_column), first_array.getOffsetsPtr()); return array_column; } @@ -168,7 +170,12 @@ REGISTER_FUNCTION(Nested) { factory.registerFunction(FunctionDocumentation{ .description=R"( +This is a function used internally by the ClickHouse engine and not meant to be used directly. + Returns the array of tuples from multiple arrays. + +The first argument must be a constant array of Strings determining the names of the resulting Tuple. +The other arguments must be arrays of the same size. )", .examples{{"nested", "SELECT nested(['keys', 'values'], ['key_1', 'key_2'], ['value_1','value_2'])", ""}}, .categories{"OtherFunctions"} diff --git a/tests/queries/0_stateless/03268_nested_analyzer.reference b/tests/queries/0_stateless/03268_nested_analyzer.reference new file mode 100644 index 00000000000..01dabfe4ba7 --- /dev/null +++ b/tests/queries/0_stateless/03268_nested_analyzer.reference @@ -0,0 +1,3 @@ +[(1,3),(2,4)] +0 0 +0 0 1 diff --git a/tests/queries/0_stateless/03268_nested_analyzer.sql b/tests/queries/0_stateless/03268_nested_analyzer.sql new file mode 100644 index 00000000000..920cf2b3174 --- /dev/null +++ b/tests/queries/0_stateless/03268_nested_analyzer.sql @@ -0,0 +1,16 @@ +SELECT nested(['a', 'b'], [1, 2], materialize([3, 4])); + +DROP TABLE IF EXISTS test; +CREATE TABLE test +( + x UInt8, + “struct.x” DEFAULT [0], + “struct.y” ALIAS [1], +) +ENGINE = Memory; + +insert into test (x) values (0); +select * from test array join struct; +select x, struct.x, struct.y from test array join struct; + +DROP TABLE test; From b5237313adaac770c95b8c9415a01c23b1372f66 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 02:49:27 +0100 Subject: [PATCH 585/680] Fix tests --- tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql | 2 +- tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql | 2 +- tests/queries/0_stateless/01958_partial_hour_timezone.sql | 2 +- tests/queries/0_stateless/02125_query_views_log.sql | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql b/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql index 907a8283396..aeaf48716dc 100644 --- a/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql +++ b/tests/queries/0_stateless/01942_dateTimeToSnowflakeID.sql @@ -1,6 +1,6 @@ SET session_timezone = 'UTC'; -- disable timezone randomization SET enable_analyzer = 1; -- The old path formats the result with different whitespaces -SET output_format_pretty_highlight_digit_groups = 0; +SET output_format_pretty_single_large_number_tip_threshold = 0; SELECT '-- Negative tests'; SELECT dateTimeToSnowflakeID(); -- {serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH} diff --git a/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql b/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql index 1f62f3d36da..e9b32607837 100644 --- a/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql +++ b/tests/queries/0_stateless/01942_snowflakeIDToDateTime.sql @@ -1,6 +1,6 @@ SET session_timezone = 'UTC'; -- disable timezone randomization SET enable_analyzer = 1; -- The old path formats the result with different whitespaces -SET output_format_pretty_highlight_digit_groups = 0; +SET output_format_pretty_single_large_number_tip_threshold = 0; SELECT '-- Negative tests'; SELECT snowflakeIDToDateTime(); -- {serverError NUMBER_OF_ARGUMENTS_DOESNT_MATCH} diff --git a/tests/queries/0_stateless/01958_partial_hour_timezone.sql b/tests/queries/0_stateless/01958_partial_hour_timezone.sql index b72adfd9d58..3eecaaf97e6 100644 --- a/tests/queries/0_stateless/01958_partial_hour_timezone.sql +++ b/tests/queries/0_stateless/01958_partial_hour_timezone.sql @@ -1,4 +1,4 @@ -SET output_format_pretty_highlight_digit_groups = 0; +SET output_format_pretty_single_large_number_tip_threshold = 0; -- Appeared in https://github.com/ClickHouse/ClickHouse/pull/26978#issuecomment-890889362 WITH toDateTime('1970-06-17 07:39:21', 'Africa/Monrovia') as t diff --git a/tests/queries/0_stateless/02125_query_views_log.sql b/tests/queries/0_stateless/02125_query_views_log.sql index 96170efedd6..08e9c73a165 100644 --- a/tests/queries/0_stateless/02125_query_views_log.sql +++ b/tests/queries/0_stateless/02125_query_views_log.sql @@ -1,4 +1,4 @@ -SET output_format_pretty_highlight_digit_groups = 0; +SET output_format_pretty_single_large_number_tip_threshold = 0; drop table if exists src; drop table if exists dst; From 959c4633f9e8cbc3f41def853fa62618fba604c6 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 02:53:32 +0100 Subject: [PATCH 586/680] Apply review suggestion --- src/Formats/PrettyFormatHelpers.cpp | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/src/Formats/PrettyFormatHelpers.cpp b/src/Formats/PrettyFormatHelpers.cpp index 6e2af036651..4ee4b49521d 100644 --- a/src/Formats/PrettyFormatHelpers.cpp +++ b/src/Formats/PrettyFormatHelpers.cpp @@ -5,6 +5,11 @@ #include +static constexpr const char * GRAY_COLOR = "\033[90m"; +static constexpr const char * UNDERSCORE = "\033[4m"; +static constexpr const char * RESET_COLOR = "\033[0m"; + + namespace DB { @@ -25,11 +30,11 @@ void writeReadableNumberTip(WriteBuffer & out, const IColumn & column, size_t ro if (threshold && isFinite(value) && abs(value) > threshold) { if (color) - writeCString("\033[90m", out); + writeCString(GRAY_COLOR, out); writeCString(" -- ", out); formatReadableQuantity(value, out, 2); if (color) - writeCString("\033[0m", out); + writeCString(RESET_COLOR, out); } } @@ -76,9 +81,9 @@ String highlightDigitGroups(String source) size_t offset = num_digits_before_decimal - digit_num; if (offset && offset % 3 == 0) { - result += "\033[4m"; + result += UNDERSCORE; result += c; - result += "\033[0m"; + result += RESET_COLOR; } else { From ef0ec74d2bfe8ae61a06e0c3fa4e33fb3c094ef6 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 04:50:18 +0100 Subject: [PATCH 587/680] Fix build --- src/Processors/Formats/Impl/PrettyBlockOutputFormat.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h b/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h index 824a2fd2e6f..81bd0e6632d 100644 --- a/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h +++ b/src/Processors/Formats/Impl/PrettyBlockOutputFormat.h @@ -55,8 +55,6 @@ protected: } bool color; - -protected: bool readable_number_tip = false; private: From bf58f468082917f871dc706f8596020d0364b43e Mon Sep 17 00:00:00 2001 From: Amos Bird Date: Sat, 9 Nov 2024 13:04:39 +0800 Subject: [PATCH 588/680] Fix empty tuple ALTER --- src/Functions/FunctionsComparison.h | 3 +++ src/Functions/if.cpp | 3 +++ .../0_stateless/03268_empty_tuple_update.reference | 1 + .../queries/0_stateless/03268_empty_tuple_update.sql | 11 +++++++++++ 4 files changed, 18 insertions(+) create mode 100644 tests/queries/0_stateless/03268_empty_tuple_update.reference create mode 100644 tests/queries/0_stateless/03268_empty_tuple_update.sql diff --git a/src/Functions/FunctionsComparison.h b/src/Functions/FunctionsComparison.h index be0875581a5..9b2328065fc 100644 --- a/src/Functions/FunctionsComparison.h +++ b/src/Functions/FunctionsComparison.h @@ -1033,6 +1033,9 @@ private: size_t tuple_size, size_t input_rows_count) const { + if (0 == tuple_size) + throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Comparison of zero-sized tuples is not implemented"); + ColumnsWithTypeAndName less_columns(tuple_size); ColumnsWithTypeAndName equal_columns(tuple_size - 1); ColumnsWithTypeAndName tmp_columns(2); diff --git a/src/Functions/if.cpp b/src/Functions/if.cpp index e03b27b3c39..5e1e7067e86 100644 --- a/src/Functions/if.cpp +++ b/src/Functions/if.cpp @@ -668,6 +668,9 @@ private: temporary_columns[0] = arguments[0]; size_t tuple_size = type1.getElements().size(); + if (tuple_size == 0) + return ColumnTuple::create(input_rows_count); + Columns tuple_columns(tuple_size); for (size_t i = 0; i < tuple_size; ++i) diff --git a/tests/queries/0_stateless/03268_empty_tuple_update.reference b/tests/queries/0_stateless/03268_empty_tuple_update.reference new file mode 100644 index 00000000000..30bc45d7a18 --- /dev/null +++ b/tests/queries/0_stateless/03268_empty_tuple_update.reference @@ -0,0 +1 @@ +() 2 diff --git a/tests/queries/0_stateless/03268_empty_tuple_update.sql b/tests/queries/0_stateless/03268_empty_tuple_update.sql new file mode 100644 index 00000000000..343117719fc --- /dev/null +++ b/tests/queries/0_stateless/03268_empty_tuple_update.sql @@ -0,0 +1,11 @@ +DROP TABLE IF EXISTS t0; + +CREATE TABLE t0 (c0 Tuple(), c1 int) ENGINE = Memory(); + +INSERT INTO t0 VALUES ((), 1); + +ALTER TABLE t0 UPDATE c0 = (), c1 = 2 WHERE EXISTS (SELECT 1); + +SELECT * FROM t0; + +DROP TABLE t0; From a888db338e1c79166a3ff6993b71d2fd17dc8736 Mon Sep 17 00:00:00 2001 From: Pablo Marcos Date: Sat, 9 Nov 2024 08:23:25 +0100 Subject: [PATCH 589/680] Revert "Add a new setting query_metric_log_debug to avoid the noise" This reverts commit 955f537bd5ef2f4a29717ac4999ce2af47b4c039. --- src/Core/Settings.cpp | 5 ----- src/Core/SettingsChangesHistory.cpp | 1 - src/Interpreters/QueryMetricLog.cpp | 18 +++++++----------- src/Interpreters/QueryMetricLog.h | 3 +-- src/Interpreters/executeQuery.cpp | 3 +-- .../03203_system_query_metric_log.sh | 10 +++++----- 6 files changed, 14 insertions(+), 26 deletions(-) diff --git a/src/Core/Settings.cpp b/src/Core/Settings.cpp index 2677bde4d55..6f0109fa300 100644 --- a/src/Core/Settings.cpp +++ b/src/Core/Settings.cpp @@ -2787,11 +2787,6 @@ If set to any negative value, it will take the value `collect_interval_milliseco To disable the collection of a single query, set `query_metric_log_interval` to 0. Default value: -1 - )", 0) \ - DECLARE(Bool, query_metric_log_debug, false, R"( -Turns on debugging traces for system.query_metric_log - -Default value: false )", 0) \ DECLARE(LogsLevel, send_logs_level, LogsLevel::fatal, R"( Send server text logs with specified minimum level to client. Valid values: 'trace', 'debug', 'information', 'warning', 'error', 'fatal', 'none' diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index 008980aae11..c6223bef2b2 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -76,7 +76,6 @@ static std::initializer_listgetProcessList(); @@ -223,24 +222,21 @@ void QueryMetricLogStatus::scheduleNext(String query_id) } else { - if (debug_traces) - LOG_DEBUG(logger, "The next collecting task for query {} should have already run at {}. Scheduling it right now", - query_id, timePointToString(next_collect_time)); + LOG_TRACE(logger, "The next collecting task for query {} should have already run at {}. Scheduling it right now", + query_id, timePointToString(next_collect_time)); task->schedule(); } } std::optional QueryMetricLogStatus::createLogMetricElement(const String & query_id, const QueryStatusInfo & query_info, TimePoint query_info_time, bool schedule_next) { - if (debug_traces) - LOG_DEBUG(logger, "Collecting query_metric_log for query {} and interval {} ms with QueryStatusInfo from {}. Next collection time: {}", - query_id, interval_milliseconds, timePointToString(query_info_time), - schedule_next ? timePointToString(next_collect_time + std::chrono::milliseconds(interval_milliseconds)) : "finished"); + LOG_TRACE(logger, "Collecting query_metric_log for query {} and interval {} ms with QueryStatusInfo from {}. Next collection time: {}", + query_id, interval_milliseconds, timePointToString(query_info_time), + schedule_next ? timePointToString(next_collect_time + std::chrono::milliseconds(interval_milliseconds)) : "finished"); if (query_info_time <= last_collect_time) { - if (debug_traces) - LOG_DEBUG(logger, "Query {} has a more recent metrics collected. Skipping this one", query_id); + LOG_TRACE(logger, "Query {} has a more recent metrics collected. Skipping this one", query_id); return {}; } diff --git a/src/Interpreters/QueryMetricLog.h b/src/Interpreters/QueryMetricLog.h index 5f301b2cd13..65764229b0a 100644 --- a/src/Interpreters/QueryMetricLog.h +++ b/src/Interpreters/QueryMetricLog.h @@ -51,7 +51,6 @@ struct QueryMetricLogStatus std::chrono::system_clock::time_point next_collect_time TSA_GUARDED_BY(getMutex()); std::vector last_profile_events TSA_GUARDED_BY(getMutex()) = std::vector(ProfileEvents::end()); BackgroundSchedulePool::TaskHolder task TSA_GUARDED_BY(getMutex()); - bool debug_traces = false; /// We need to be able to move it for the hash map, so we need to add an indirection here. std::unique_ptr mutex = std::make_unique(); @@ -79,7 +78,7 @@ public: void shutdown() final; /// Both startQuery and finishQuery are called from the thread that executes the query. - void startQuery(const String & query_id, TimePoint start_time, UInt64 interval_milliseconds, bool debug_traces = false); + void startQuery(const String & query_id, TimePoint start_time, UInt64 interval_milliseconds); void finishQuery(const String & query_id, TimePoint finish_time, QueryStatusInfoPtr query_info = nullptr); private: diff --git a/src/Interpreters/executeQuery.cpp b/src/Interpreters/executeQuery.cpp index 794d3dab0e6..4507126b7b3 100644 --- a/src/Interpreters/executeQuery.cpp +++ b/src/Interpreters/executeQuery.cpp @@ -146,7 +146,6 @@ namespace Setting extern const SettingsQueryCacheSystemTableHandling query_cache_system_table_handling; extern const SettingsSeconds query_cache_ttl; extern const SettingsInt64 query_metric_log_interval; - extern const SettingsBool query_metric_log_debug; extern const SettingsOverflowMode read_overflow_mode; extern const SettingsOverflowMode read_overflow_mode_leaf; extern const SettingsOverflowMode result_overflow_mode; @@ -456,7 +455,7 @@ QueryLogElement logQueryStart( { auto interval_milliseconds = getQueryMetricLogInterval(context); if (interval_milliseconds > 0) - query_metric_log->startQuery(elem.client_info.current_query_id, query_start_time, interval_milliseconds, settings[Setting::query_metric_log_debug]); + query_metric_log->startQuery(elem.client_info.current_query_id, query_start_time, interval_milliseconds); } return elem; diff --git a/tests/queries/0_stateless/03203_system_query_metric_log.sh b/tests/queries/0_stateless/03203_system_query_metric_log.sh index 4bc764b777c..abcd14c8e5d 100755 --- a/tests/queries/0_stateless/03203_system_query_metric_log.sh +++ b/tests/queries/0_stateless/03203_system_query_metric_log.sh @@ -6,11 +6,11 @@ CURDIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) readonly query_prefix=$CLICKHOUSE_DATABASE -$CLICKHOUSE_CLIENT --query-id="${query_prefix}_1000" -q "SELECT sleep(2.5) SETTINGS query_metric_log_debug=true FORMAT Null" & -$CLICKHOUSE_CLIENT --query-id="${query_prefix}_400" -q "SELECT sleep(2.5) SETTINGS query_metric_log_debug=true, query_metric_log_interval=400 FORMAT Null" & -$CLICKHOUSE_CLIENT --query-id="${query_prefix}_123" -q "SELECT sleep(2.5) SETTINGS query_metric_log_debug=true, query_metric_log_interval=123 FORMAT Null" & -$CLICKHOUSE_CLIENT --query-id="${query_prefix}_0" -q "SELECT sleep(2.5) SETTINGS query_metric_log_debug=true, query_metric_log_interval=0 FORMAT Null" & -$CLICKHOUSE_CLIENT --query-id="${query_prefix}_fast" -q "SELECT sleep(0.1) SETTINGS query_metric_log_debug=true FORMAT Null" & +$CLICKHOUSE_CLIENT --query-id="${query_prefix}_1000" -q "SELECT sleep(2.5) FORMAT Null" & +$CLICKHOUSE_CLIENT --query-id="${query_prefix}_400" -q "SELECT sleep(2.5) SETTINGS query_metric_log_interval=400 FORMAT Null" & +$CLICKHOUSE_CLIENT --query-id="${query_prefix}_123" -q "SELECT sleep(2.5) SETTINGS query_metric_log_interval=123 FORMAT Null" & +$CLICKHOUSE_CLIENT --query-id="${query_prefix}_0" -q "SELECT sleep(2.5) SETTINGS query_metric_log_interval=0 FORMAT Null" & +$CLICKHOUSE_CLIENT --query-id="${query_prefix}_fast" -q "SELECT sleep(0.1) FORMAT Null" & wait From 516300e733c8b3a116139a0306797f779b818f56 Mon Sep 17 00:00:00 2001 From: Pablo Marcos Date: Sat, 9 Nov 2024 08:28:47 +0100 Subject: [PATCH 590/680] Demote log from warning to debug to avoid failing the test --- src/Interpreters/QueryMetricLog.cpp | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/src/Interpreters/QueryMetricLog.cpp b/src/Interpreters/QueryMetricLog.cpp index 4fbe4f9e1b5..62700f49605 100644 --- a/src/Interpreters/QueryMetricLog.cpp +++ b/src/Interpreters/QueryMetricLog.cpp @@ -107,6 +107,7 @@ void QueryMetricLog::collectMetric(const ProcessList & process_list, String quer const auto query_info = process_list.getQueryInfo(query_id, false, true, false); if (!query_info) { + /// TODO: remove trace before 24.11 release after checking everything is fine on the CI LOG_TRACE(logger, "Query {} is not running anymore, so we couldn't get its QueryStatusInfo", query_id); return; } @@ -118,6 +119,7 @@ void QueryMetricLog::collectMetric(const ProcessList & process_list, String quer if (it == queries.end()) { global_lock.unlock(); + /// TODO: remove trace before 24.11 release after checking everything is fine on the CI LOG_TRACE(logger, "Query {} not found in the list. Finished while this collecting task was running", query_id); return; } @@ -126,6 +128,7 @@ void QueryMetricLog::collectMetric(const ProcessList & process_list, String quer if (!query_status.mutex) { global_lock.unlock(); + /// TODO: remove trace before 24.11 release after checking everything is fine on the CI LOG_TRACE(logger, "Query {} finished while this collecting task was running", query_id); return; } @@ -230,12 +233,14 @@ void QueryMetricLogStatus::scheduleNext(String query_id) std::optional QueryMetricLogStatus::createLogMetricElement(const String & query_id, const QueryStatusInfo & query_info, TimePoint query_info_time, bool schedule_next) { + /// TODO: remove trace before 24.11 release after checking everything is fine on the CI LOG_TRACE(logger, "Collecting query_metric_log for query {} and interval {} ms with QueryStatusInfo from {}. Next collection time: {}", query_id, interval_milliseconds, timePointToString(query_info_time), schedule_next ? timePointToString(next_collect_time + std::chrono::milliseconds(interval_milliseconds)) : "finished"); if (query_info_time <= last_collect_time) { + /// TODO: remove trace before 24.11 release after checking everything is fine on the CI LOG_TRACE(logger, "Query {} has a more recent metrics collected. Skipping this one", query_id); return {}; } @@ -278,7 +283,8 @@ std::optional QueryMetricLogStatus::createLogMetricElemen } else { - LOG_WARNING(logger, "Query {} has no profile counters", query_id); + /// TODO: remove trace before 24.11 release after checking everything is fine on the CI + LOG_DEBUG(logger, "Query {} has no profile counters", query_id); elem.profile_events = std::vector(ProfileEvents::end()); } From e50bbc433e5c57c96bbf71e22b900a28eb5be6c5 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Fri, 8 Nov 2024 22:08:08 +0100 Subject: [PATCH 591/680] Another review round for docker-library/docs --- docker/server/README.md | 4 ++-- docker/server/README.src/content.md | 4 ++-- docker/server/README.src/github-repo | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/docker/server/README.md b/docker/server/README.md index 7403d5b0b2a..5f6144d0633 100644 --- a/docker/server/README.md +++ b/docker/server/README.md @@ -30,7 +30,7 @@ For more information and documentation see https://clickhouse.com/. - The amd64 image requires support for [SSE3 instructions](https://en.wikipedia.org/wiki/SSE3). Virtually all x86 CPUs after 2005 support SSE3. - The arm64 image requires support for the [ARMv8.2-A architecture](https://en.wikipedia.org/wiki/AArch64#ARMv8.2-A) and additionally the Load-Acquire RCpc register. The register is optional in version ARMv8.2-A and mandatory in [ARMv8.3-A](https://en.wikipedia.org/wiki/AArch64#ARMv8.3-A). Supported in Graviton >=2, Azure and GCP instances. Examples for unsupported devices are Raspberry Pi 4 (ARMv8.0-A) and Jetson AGX Xavier/Orin (ARMv8.2-A). -- Since the Clickhouse 24.11 Ubuntu images started using `ubuntu:22.04` as its base image. It requires docker version >= `20.10.10` containing [patch](https://github.com/moby/moby/commit/977283509f75303bc6612665a04abf76ff1d2468). As a workaround you could use `docker run [--privileged | --security-opt seccomp=unconfined]` instead, however that has security implications. +- Since the Clickhouse 24.11 Ubuntu images started using `ubuntu:22.04` as its base image. It requires docker version >= `20.10.10` containing [patch](https://github.com/moby/moby/commit/977283509f75303bc6612665a04abf76ff1d2468). As a workaround you could use `docker run --security-opt seccomp=unconfined` instead, however that has security implications. ## How to use this image @@ -57,7 +57,7 @@ More information about the [ClickHouse client](https://clickhouse.com/docs/en/in ### connect to it using curl ```bash -echo "SELECT 'Hello, ClickHouse!'" | docker run -i --rm --link some-clickhouse-server:clickhouse-server buildpack-deps:curl 'http://clickhouse-server:8123/?query=' -s --data-binary @- +echo "SELECT 'Hello, ClickHouse!'" | docker run -i --rm --link some-clickhouse-server:clickhouse-server buildpack-deps:curl curl 'http://clickhouse-server:8123/?query=' -s --data-binary @- ``` More information about the [ClickHouse HTTP Interface](https://clickhouse.com/docs/en/interfaces/http/). diff --git a/docker/server/README.src/content.md b/docker/server/README.src/content.md index bfc1a271546..df0b6718d69 100644 --- a/docker/server/README.src/content.md +++ b/docker/server/README.src/content.md @@ -24,7 +24,7 @@ For more information and documentation see https://clickhouse.com/. - The amd64 image requires support for [SSE3 instructions](https://en.wikipedia.org/wiki/SSE3). Virtually all x86 CPUs after 2005 support SSE3. - The arm64 image requires support for the [ARMv8.2-A architecture](https://en.wikipedia.org/wiki/AArch64#ARMv8.2-A) and additionally the Load-Acquire RCpc register. The register is optional in version ARMv8.2-A and mandatory in [ARMv8.3-A](https://en.wikipedia.org/wiki/AArch64#ARMv8.3-A). Supported in Graviton >=2, Azure and GCP instances. Examples for unsupported devices are Raspberry Pi 4 (ARMv8.0-A) and Jetson AGX Xavier/Orin (ARMv8.2-A). -- Since the Clickhouse 24.11 Ubuntu images started using `ubuntu:22.04` as its base image. It requires docker version >= `20.10.10` containing [patch](https://github.com/moby/moby/commit/977283509f75303bc6612665a04abf76ff1d2468). As a workaround you could use `docker run [--privileged | --security-opt seccomp=unconfined]` instead, however that has security implications. +- Since the Clickhouse 24.11 Ubuntu images started using `ubuntu:22.04` as its base image. It requires docker version >= `20.10.10` containing [patch](https://github.com/moby/moby/commit/977283509f75303bc6612665a04abf76ff1d2468). As a workaround you could use `docker run --security-opt seccomp=unconfined` instead, however that has security implications. ## How to use this image @@ -51,7 +51,7 @@ More information about the [ClickHouse client](https://clickhouse.com/docs/en/in ### connect to it using curl ```bash -echo "SELECT 'Hello, ClickHouse!'" | docker run -i --rm --link some-clickhouse-server:clickhouse-server buildpack-deps:curl 'http://clickhouse-server:8123/?query=' -s --data-binary @- +echo "SELECT 'Hello, ClickHouse!'" | docker run -i --rm --link some-clickhouse-server:clickhouse-server buildpack-deps:curl curl 'http://clickhouse-server:8123/?query=' -s --data-binary @- ``` More information about the [ClickHouse HTTP Interface](https://clickhouse.com/docs/en/interfaces/http/). diff --git a/docker/server/README.src/github-repo b/docker/server/README.src/github-repo index dc2b6635325..70a009ec958 100644 --- a/docker/server/README.src/github-repo +++ b/docker/server/README.src/github-repo @@ -1 +1 @@ -https://github.com/ClickHouse/docker-library +https://github.com/ClickHouse/ClickHouse From aa4d37f72cbea7834c8a8c8d6668f3b3b01b80a7 Mon Sep 17 00:00:00 2001 From: alesapin Date: Sat, 9 Nov 2024 13:41:08 +0100 Subject: [PATCH 592/680] Fix test --- .../0_stateless/02117_show_create_table_system.reference | 2 ++ 1 file changed, 2 insertions(+) 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 2ea62444cff..ef5a2c6665f 100644 --- a/tests/queries/0_stateless/02117_show_create_table_system.reference +++ b/tests/queries/0_stateless/02117_show_create_table_system.reference @@ -485,6 +485,8 @@ CREATE TABLE system.parts `data_version` UInt64, `primary_key_bytes_in_memory` UInt64, `primary_key_bytes_in_memory_allocated` UInt64, + `index_granularity_bytes_in_memory` UInt64, + `index_granularity_bytes_in_memory_allocated` UInt64, `is_frozen` UInt8, `database` String, `table` String, From 7b1c72729a4f8e37d7c6ddf4fc8894149085e3e3 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 14:22:43 +0100 Subject: [PATCH 593/680] Fix upgrade check --- tests/docker_scripts/upgrade_runner.sh | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/tests/docker_scripts/upgrade_runner.sh b/tests/docker_scripts/upgrade_runner.sh index ece75ebf782..92484f88ece 100755 --- a/tests/docker_scripts/upgrade_runner.sh +++ b/tests/docker_scripts/upgrade_runner.sh @@ -135,7 +135,7 @@ IS_SANITIZED=$(clickhouse-local --query "SELECT value LIKE '%-fsanitize=%' FROM if [ "${IS_SANITIZED}" -eq "0" ] then save_settings_clean 'new_settings.native' - clickhouse-local -nmq " + clickhouse-local --implicit-select 0 -nmq " CREATE TABLE old_settings AS file('old_settings.native'); CREATE TABLE old_version AS file('old_version.native'); CREATE TABLE new_settings AS file('new_settings.native'); @@ -147,7 +147,6 @@ then FROM new_settings LEFT JOIN old_settings ON new_settings.name = old_settings.name WHERE (new_value != old_value) - AND NOT (startsWith(new_value, 'auto(') AND old_value LIKE '%auto(%') AND (name NOT IN ( SELECT arrayJoin(tupleElement(changes, 'name')) FROM From 93d586876092ae662f510d67e1ad00e6d1d55bdf Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 14:30:46 +0100 Subject: [PATCH 594/680] Fix tests --- .../02751_ip_types_aggregate_functions_states.sql.j2 | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/queries/0_stateless/02751_ip_types_aggregate_functions_states.sql.j2 b/tests/queries/0_stateless/02751_ip_types_aggregate_functions_states.sql.j2 index 7d030d4be2d..602b98e576b 100644 --- a/tests/queries/0_stateless/02751_ip_types_aggregate_functions_states.sql.j2 +++ b/tests/queries/0_stateless/02751_ip_types_aggregate_functions_states.sql.j2 @@ -1,5 +1,7 @@ -- Tags: no-parallel, no-fasttest +SET output_format_pretty_single_large_number_tip_threshold = 0; + {# this test checks backward compatibility of aggregate functions States against IPv4, IPv6 types #} {% set ip4_generator = "select num::UInt32::IPv4 ip from (select arrayJoin(range(999999999, number)) as num from numbers(999999999,50)) order by ip" %} From 016c122af9d85da32726d9bb0d2b318eda06c2e8 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 16:33:37 +0100 Subject: [PATCH 595/680] Update PULL_REQUEST_TEMPLATE.md --- .github/PULL_REQUEST_TEMPLATE.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.github/PULL_REQUEST_TEMPLATE.md b/.github/PULL_REQUEST_TEMPLATE.md index 3dcce68ab46..976c69d3c34 100644 --- a/.github/PULL_REQUEST_TEMPLATE.md +++ b/.github/PULL_REQUEST_TEMPLATE.md @@ -12,7 +12,7 @@ tests/ci/cancel_and_rerun_workflow_lambda/app.py - Backward Incompatible Change - Build/Testing/Packaging Improvement - Documentation (changelog entry is not required) -- Critical Bug Fix (crash, LOGICAL_ERROR, data loss, RBAC) +- Critical Bug Fix (crash, data loss, RBAC) - Bug Fix (user-visible misbehavior in an official stable release) - CI Fix or Improvement (changelog entry is not required) - Not for changelog (changelog entry is not required) From a898f163546b3cb3d607443e9881c88337867e83 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 17:21:53 +0100 Subject: [PATCH 596/680] Fix tests --- tests/queries/0_stateless/02184_hash_functions_and_ip_types.sql | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/queries/0_stateless/02184_hash_functions_and_ip_types.sql b/tests/queries/0_stateless/02184_hash_functions_and_ip_types.sql index e7d1909cae6..22b59a16255 100644 --- a/tests/queries/0_stateless/02184_hash_functions_and_ip_types.sql +++ b/tests/queries/0_stateless/02184_hash_functions_and_ip_types.sql @@ -1,5 +1,6 @@ -- Tags: no-fasttest +SET output_format_pretty_single_large_number_tip_threshold = 0; SET enable_analyzer = 1; SELECT From 7849a9ce16d8a8ed9d97e3b57541c108bb00d044 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 18:12:11 +0100 Subject: [PATCH 597/680] Fix error --- tests/docker_scripts/upgrade_runner.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/docker_scripts/upgrade_runner.sh b/tests/docker_scripts/upgrade_runner.sh index 92484f88ece..15c5ab69521 100755 --- a/tests/docker_scripts/upgrade_runner.sh +++ b/tests/docker_scripts/upgrade_runner.sh @@ -63,7 +63,7 @@ install_packages previous_release_package_folder function save_settings_clean() { local out=$1 && shift - script -q -c "clickhouse-local -q \"select * from system.settings into outfile '$out'\"" --log-out /dev/null + script -q -c "clickhouse-local --implicit-select 0 -q \"select * from system.settings into outfile '$out'\"" --log-out /dev/null } # We save the (numeric) version of the old server to compare setting changes between the 2 @@ -135,7 +135,7 @@ IS_SANITIZED=$(clickhouse-local --query "SELECT value LIKE '%-fsanitize=%' FROM if [ "${IS_SANITIZED}" -eq "0" ] then save_settings_clean 'new_settings.native' - clickhouse-local --implicit-select 0 -nmq " + clickhouse-local -nmq " CREATE TABLE old_settings AS file('old_settings.native'); CREATE TABLE old_version AS file('old_version.native'); CREATE TABLE new_settings AS file('new_settings.native'); From 979b2128067e44e92bb738b491ebabaa0f41cbeb Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 1 Jan 2024 19:36:35 +0100 Subject: [PATCH 598/680] Make higher order functions constant expressions --- src/Functions/FunctionsMiscellaneous.h | 14 +++++++++++++- ...961_higher_order_constant_expressions.reference | 8 ++++++++ .../02961_higher_order_constant_expressions.sql | 11 +++++++++++ 3 files changed, 32 insertions(+), 1 deletion(-) create mode 100644 tests/queries/0_stateless/02961_higher_order_constant_expressions.reference create mode 100644 tests/queries/0_stateless/02961_higher_order_constant_expressions.sql diff --git a/src/Functions/FunctionsMiscellaneous.h b/src/Functions/FunctionsMiscellaneous.h index fb5109eaa88..4b189279651 100644 --- a/src/Functions/FunctionsMiscellaneous.h +++ b/src/Functions/FunctionsMiscellaneous.h @@ -6,6 +6,7 @@ #include #include #include +#include #include @@ -122,12 +123,19 @@ public: String getName() const override { return "FunctionCapture"; } bool useDefaultImplementationForNulls() const override { return false; } + /// It's possible if expression_actions contains function that don't use /// default implementation for Nothing and one of captured columns can be Nothing /// Example: SELECT arrayMap(x -> [x, arrayElement(y, 0)], []), [] as y bool useDefaultImplementationForNothing() const override { return false; } bool useDefaultImplementationForLowCardinalityColumns() const override { return false; } + /// If all the captured arguments are constant, let's also return ColumnConst (with ColumnFunction inside it). + /// Consequently, it allows to treat higher order functions with constant arrays and constant captured columns + /// as constant expressions. + /// Consequently, it allows its usage in contexts requiring constants, such as the right hand side of IN. + bool useDefaultImplementationForConstants() const override { return true; } + ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { Names names; @@ -148,7 +156,11 @@ public: auto function = std::make_unique(expression_actions, types, names, capture->return_type, capture->return_name); - return ColumnFunction::create(input_rows_count, std::move(function), arguments); + /// If there are no captured columns, the result is constant. + if (arguments.empty()) + return ColumnConst::create(ColumnFunction::create(1, std::move(function), arguments), input_rows_count); + else + return ColumnFunction::create(input_rows_count, std::move(function), arguments); } private: diff --git a/tests/queries/0_stateless/02961_higher_order_constant_expressions.reference b/tests/queries/0_stateless/02961_higher_order_constant_expressions.reference new file mode 100644 index 00000000000..058d23ad850 --- /dev/null +++ b/tests/queries/0_stateless/02961_higher_order_constant_expressions.reference @@ -0,0 +1,8 @@ +[1,2,3] 1 +[2,3,4] 1 +[2,4,6] 1 +[5,7,9] 1 +[1,1,1] 1 +[1,2,3] 0 +[0,0,0] 0 +3 1 diff --git a/tests/queries/0_stateless/02961_higher_order_constant_expressions.sql b/tests/queries/0_stateless/02961_higher_order_constant_expressions.sql new file mode 100644 index 00000000000..47480010751 --- /dev/null +++ b/tests/queries/0_stateless/02961_higher_order_constant_expressions.sql @@ -0,0 +1,11 @@ +SELECT arrayMap(x -> x, [1, 2, 3]) AS x, isConstant(x); +SELECT arrayMap(x -> x + 1, [1, 2, 3]) AS x, isConstant(x); +SELECT arrayMap(x -> x + x, [1, 2, 3]) AS x, isConstant(x); +SELECT arrayMap((x, y) -> x + y, [1, 2, 3], [4, 5, 6]) AS x, isConstant(x); +SELECT arrayMap(x -> 1, [1, 2, 3]) AS x, isConstant(x); +SELECT arrayMap(x -> x + number, [1, 2, 3]) AS x, isConstant(x) FROM numbers(1); +SELECT arrayMap(x -> number, [1, 2, 3]) AS x, isConstant(x) FROM numbers(1); +SELECT arrayMax([1, 2, 3]) AS x, isConstant(x); + +-- Does not work yet: +-- SELECT [1, 2, 3] IN arrayMap(x -> x, [1, 2, 3]); From 3f2f358fb9c7e8b01e32632684c1dea24c1fc67a Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 1 Jan 2024 20:32:27 +0100 Subject: [PATCH 599/680] Support constant lambda functions --- src/Functions/array/FunctionArrayMapped.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/Functions/array/FunctionArrayMapped.h b/src/Functions/array/FunctionArrayMapped.h index f4832431f04..e51c465f883 100644 --- a/src/Functions/array/FunctionArrayMapped.h +++ b/src/Functions/array/FunctionArrayMapped.h @@ -282,7 +282,9 @@ public: if (!column_with_type_and_name.column) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "First argument for function {} must be a function.", getName()); - const auto * column_function = typeid_cast(column_with_type_and_name.column.get()); + auto column_function_materialized = column_with_type_and_name.column->convertToFullColumnIfConst(); + + const auto * column_function = typeid_cast(column_function_materialized.get()); if (!column_function) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "First argument for function {} must be a function.", getName()); From 6c1016568c4e76e2285a5a73d5bbfc7c3d0824a5 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 03:35:21 +0100 Subject: [PATCH 600/680] Better implementation --- src/Functions/FunctionsMiscellaneous.h | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) diff --git a/src/Functions/FunctionsMiscellaneous.h b/src/Functions/FunctionsMiscellaneous.h index 4b189279651..6e89c4dd65d 100644 --- a/src/Functions/FunctionsMiscellaneous.h +++ b/src/Functions/FunctionsMiscellaneous.h @@ -130,12 +130,6 @@ public: bool useDefaultImplementationForNothing() const override { return false; } bool useDefaultImplementationForLowCardinalityColumns() const override { return false; } - /// If all the captured arguments are constant, let's also return ColumnConst (with ColumnFunction inside it). - /// Consequently, it allows to treat higher order functions with constant arrays and constant captured columns - /// as constant expressions. - /// Consequently, it allows its usage in contexts requiring constants, such as the right hand side of IN. - bool useDefaultImplementationForConstants() const override { return true; } - ColumnPtr executeImpl(const ColumnsWithTypeAndName & arguments, const DataTypePtr &, size_t input_rows_count) const override { Names names; @@ -156,8 +150,13 @@ public: auto function = std::make_unique(expression_actions, types, names, capture->return_type, capture->return_name); - /// If there are no captured columns, the result is constant. - if (arguments.empty()) + /// If all the captured arguments are constant, let's also return ColumnConst (with ColumnFunction inside it). + /// Consequently, it allows to treat higher order functions with constant arrays and constant captured columns + /// as constant expressions. + /// Consequently, it allows its usage in contexts requiring constants, such as the right hand side of IN. + bool all_arguments_are_constant = std::all_of(arguments.begin(), arguments.end(), [](const auto & arg) { return arg.column->isConst(); }); + + if (all_arguments_are_constant) return ColumnConst::create(ColumnFunction::create(1, std::move(function), arguments), input_rows_count); else return ColumnFunction::create(input_rows_count, std::move(function), arguments); From f1777b957946a5ab32851d8e1c7000ee3657e3d7 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 03:55:18 +0100 Subject: [PATCH 601/680] Fix error --- src/Functions/FunctionsMiscellaneous.h | 10 +++++++++- src/Processors/Formats/IOutputFormat.cpp | 2 +- src/Processors/Transforms/DistinctTransform.cpp | 2 +- src/QueryPipeline/RemoteQueryExecutor.cpp | 2 +- 4 files changed, 12 insertions(+), 4 deletions(-) diff --git a/src/Functions/FunctionsMiscellaneous.h b/src/Functions/FunctionsMiscellaneous.h index 6e89c4dd65d..62b43386db5 100644 --- a/src/Functions/FunctionsMiscellaneous.h +++ b/src/Functions/FunctionsMiscellaneous.h @@ -157,9 +157,17 @@ public: bool all_arguments_are_constant = std::all_of(arguments.begin(), arguments.end(), [](const auto & arg) { return arg.column->isConst(); }); if (all_arguments_are_constant) - return ColumnConst::create(ColumnFunction::create(1, std::move(function), arguments), input_rows_count); + { + ColumnsWithTypeAndName arguments_resized = arguments; + for (auto & elem : arguments_resized) + elem.column = elem.column->cloneResized(1); + + return ColumnConst::create(ColumnFunction::create(1, std::move(function), arguments_resized), input_rows_count); + } else + { return ColumnFunction::create(input_rows_count, std::move(function), arguments); + } } private: diff --git a/src/Processors/Formats/IOutputFormat.cpp b/src/Processors/Formats/IOutputFormat.cpp index 97628778adb..947de45f852 100644 --- a/src/Processors/Formats/IOutputFormat.cpp +++ b/src/Processors/Formats/IOutputFormat.cpp @@ -55,7 +55,7 @@ static Chunk prepareTotals(Chunk chunk) /// Skip rows except the first one. auto columns = chunk.detachColumns(); for (auto & column : columns) - column = column->cut(0, 1); + column = column->cloneResized(1); chunk.setColumns(std::move(columns), 1); } diff --git a/src/Processors/Transforms/DistinctTransform.cpp b/src/Processors/Transforms/DistinctTransform.cpp index d528303a642..53ee2c52884 100644 --- a/src/Processors/Transforms/DistinctTransform.cpp +++ b/src/Processors/Transforms/DistinctTransform.cpp @@ -64,7 +64,7 @@ void DistinctTransform::transform(Chunk & chunk) if (unlikely(key_columns_pos.empty())) { for (auto & column : columns) - column = column->cut(0, 1); + column = column->cloneResized(1); chunk.setColumns(std::move(columns), 1); stopReading(); diff --git a/src/QueryPipeline/RemoteQueryExecutor.cpp b/src/QueryPipeline/RemoteQueryExecutor.cpp index 5faae03bc8f..401b3d36f83 100644 --- a/src/QueryPipeline/RemoteQueryExecutor.cpp +++ b/src/QueryPipeline/RemoteQueryExecutor.cpp @@ -327,7 +327,7 @@ static Block adaptBlockStructure(const Block & block, const Block & header) /// TODO: check that column contains the same value. /// TODO: serialize const columns. auto col = block.getByName(elem.name); - col.column = block.getByName(elem.name).column->cut(0, 1); + col.column = block.getByName(elem.name).column->cloneResized(1); column = castColumn(col, elem.type); From 4ae7d589f7f09db1488618f47c8137ac6aca0d01 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 03:56:23 +0100 Subject: [PATCH 602/680] Fix error --- src/Processors/Formats/IOutputFormat.cpp | 2 +- src/Processors/Transforms/DistinctTransform.cpp | 2 +- src/QueryPipeline/RemoteQueryExecutor.cpp | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/Processors/Formats/IOutputFormat.cpp b/src/Processors/Formats/IOutputFormat.cpp index 947de45f852..97628778adb 100644 --- a/src/Processors/Formats/IOutputFormat.cpp +++ b/src/Processors/Formats/IOutputFormat.cpp @@ -55,7 +55,7 @@ static Chunk prepareTotals(Chunk chunk) /// Skip rows except the first one. auto columns = chunk.detachColumns(); for (auto & column : columns) - column = column->cloneResized(1); + column = column->cut(0, 1); chunk.setColumns(std::move(columns), 1); } diff --git a/src/Processors/Transforms/DistinctTransform.cpp b/src/Processors/Transforms/DistinctTransform.cpp index 53ee2c52884..d528303a642 100644 --- a/src/Processors/Transforms/DistinctTransform.cpp +++ b/src/Processors/Transforms/DistinctTransform.cpp @@ -64,7 +64,7 @@ void DistinctTransform::transform(Chunk & chunk) if (unlikely(key_columns_pos.empty())) { for (auto & column : columns) - column = column->cloneResized(1); + column = column->cut(0, 1); chunk.setColumns(std::move(columns), 1); stopReading(); diff --git a/src/QueryPipeline/RemoteQueryExecutor.cpp b/src/QueryPipeline/RemoteQueryExecutor.cpp index 401b3d36f83..5faae03bc8f 100644 --- a/src/QueryPipeline/RemoteQueryExecutor.cpp +++ b/src/QueryPipeline/RemoteQueryExecutor.cpp @@ -327,7 +327,7 @@ static Block adaptBlockStructure(const Block & block, const Block & header) /// TODO: check that column contains the same value. /// TODO: serialize const columns. auto col = block.getByName(elem.name); - col.column = block.getByName(elem.name).column->cloneResized(1); + col.column = block.getByName(elem.name).column->cut(0, 1); column = castColumn(col, elem.type); From 1e64b56a0f7f9cf0e7d209db4af33511d440954d Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 04:18:32 +0100 Subject: [PATCH 603/680] Support constexpr functions in arrayFold --- src/Functions/array/arrayFold.cpp | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/Functions/array/arrayFold.cpp b/src/Functions/array/arrayFold.cpp index 483a5d6404b..a9635f82db4 100644 --- a/src/Functions/array/arrayFold.cpp +++ b/src/Functions/array/arrayFold.cpp @@ -87,7 +87,9 @@ public: if (!lambda_function_with_type_and_name.column) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "First argument for function {} must be a function", getName()); - const auto * lambda_function = typeid_cast(lambda_function_with_type_and_name.column.get()); + auto lambda_function_materialized = lambda_function_with_type_and_name.column->convertToFullColumnIfConst(); + + const auto * lambda_function = typeid_cast(lambda_function_materialized.get()); if (!lambda_function) throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "First argument for function {} must be a function", getName()); From 4334a149735742c154e9381ea9b997bb25f03dbd Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 04:20:13 +0100 Subject: [PATCH 604/680] Fix test --- tests/queries/0_stateless/01284_fuzz_bits.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/queries/0_stateless/01284_fuzz_bits.sql b/tests/queries/0_stateless/01284_fuzz_bits.sql index 95a07c7bd44..1055d2aa580 100644 --- a/tests/queries/0_stateless/01284_fuzz_bits.sql +++ b/tests/queries/0_stateless/01284_fuzz_bits.sql @@ -18,7 +18,7 @@ FROM reinterpretAsUInt8( substring( fuzzBits( - arrayStringConcat(arrayMap(x -> toString('\0'), range(10000))), + materialize(arrayStringConcat(arrayMap(x -> toString('\0'), range(10000)))), 0.3 ), id + 1, From f2d45ba43b1d846dfb27d9f1b15b15dddb59c930 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 04:35:42 +0100 Subject: [PATCH 605/680] Fix tests --- src/Storages/MergeTree/KeyCondition.cpp | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/src/Storages/MergeTree/KeyCondition.cpp b/src/Storages/MergeTree/KeyCondition.cpp index 17723d341fb..c6497660386 100644 --- a/src/Storages/MergeTree/KeyCondition.cpp +++ b/src/Storages/MergeTree/KeyCondition.cpp @@ -597,12 +597,15 @@ static const ActionsDAG::Node & cloneASTWithInversionPushDown( case (ActionsDAG::ActionType::COLUMN): { String name; - if (const auto * column_const = typeid_cast(node.column.get())) + if (const auto * column_const = typeid_cast(node.column.get()); + column_const && column_const->getDataType() != TypeIndex::Function) + { /// 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. + /// DAG from the query (with enabled analyzer) uses suffixes for constants, like 1_UInt8. + /// DAG from the PK does not use it. This breaks matching by column name sometimes. /// Ideally, we should not compare names, but DAG subtrees instead. - name = ASTLiteral(column_const->getDataColumn()[0]).getColumnName(); + name = ASTLiteral(column_const->getField()).getColumnName(); + } else name = node.result_name; From ce8ffaf5c344c924f01697c543632068895d3bb9 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sat, 9 Nov 2024 23:54:37 +0100 Subject: [PATCH 606/680] Miscellaneous --- src/Analyzer/Resolve/QueryAnalyzer.cpp | 16 +++++++--------- src/Columns/ColumnFunction.cpp | 20 ++++++++++++++++++++ src/Columns/ColumnFunction.h | 10 ++-------- src/Functions/IFunction.h | 2 +- 4 files changed, 30 insertions(+), 18 deletions(-) diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index 4bb283cbf3e..390418494e7 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -51,7 +51,6 @@ #include #include #include -#include #include #include #include @@ -3023,9 +3022,10 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi argument_column.name = arguments_projection_names[function_argument_index]; /** If function argument is lambda, save lambda argument index and initialize argument type as DataTypeFunction - * where function argument types are initialized with empty array of lambda arguments size. + * where function argument types are initialized with empty arrays of lambda arguments size. */ - if (const auto * lambda_node = function_argument->as()) + const auto * lambda_node = function_argument->as(); + if (lambda_node) { size_t lambda_arguments_size = lambda_node->getArguments().getNodes().size(); argument_column.type = std::make_shared(DataTypes(lambda_arguments_size, nullptr), nullptr); @@ -3497,15 +3497,11 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi else function_base = function->build(argument_columns); - /// Do not constant fold get scalar functions - // bool disable_constant_folding = function_name == "__getScalar" || function_name == "shardNum" || - // function_name == "shardCount" || function_name == "hostName" || function_name == "tcpPort"; - /** If function is suitable for constant folding try to convert it to constant. * Example: SELECT plus(1, 1); * Result: SELECT 2; */ - if (function_base->isSuitableForConstantFolding()) // && !disable_constant_folding) + if (function_base->isSuitableForConstantFolding()) { auto result_type = function_base->getResultType(); auto executable_function = function_base->prepare(argument_columns); @@ -3514,7 +3510,9 @@ ProjectionNames QueryAnalyzer::resolveFunction(QueryTreeNodePtr & node, Identifi if (all_arguments_constants) { - size_t num_rows = function_arguments.empty() ? 0 : argument_columns.front().column->size(); + size_t num_rows = 0; + if (!argument_columns.empty()) + num_rows = argument_columns.front().column->size(); column = executable_function->execute(argument_columns, result_type, num_rows, true); } else diff --git a/src/Columns/ColumnFunction.cpp b/src/Columns/ColumnFunction.cpp index 18c343c6ca6..cc80d04444e 100644 --- a/src/Columns/ColumnFunction.cpp +++ b/src/Columns/ColumnFunction.cpp @@ -72,6 +72,26 @@ ColumnPtr ColumnFunction::cut(size_t start, size_t length) const return ColumnFunction::create(length, function, capture, is_short_circuit_argument, is_function_compiled); } +Field ColumnFunction::operator[](size_t n) const +{ + Field res; + get(n, res); + return res; +} + +void ColumnFunction::get(size_t n, Field & res) const +{ + const size_t tuple_size = captured_columns.size(); + + res = Tuple(); + Tuple & res_tuple = res.safeGet(); + res_tuple.reserve(tuple_size); + + for (size_t i = 0; i < tuple_size; ++i) + res_tuple.push_back((*captured_columns[i].column)[n]); +} + + #if !defined(DEBUG_OR_SANITIZER_BUILD) void ColumnFunction::insertFrom(const IColumn & src, size_t n) #else diff --git a/src/Columns/ColumnFunction.h b/src/Columns/ColumnFunction.h index b62c6bf70eb..8df9e23c0e8 100644 --- a/src/Columns/ColumnFunction.h +++ b/src/Columns/ColumnFunction.h @@ -60,15 +60,9 @@ public: void appendArguments(const ColumnsWithTypeAndName & columns); ColumnWithTypeAndName reduce() const; - Field operator[](size_t) const override - { - throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Cannot get value from {}", getName()); - } + Field operator[](size_t n) const override; - void get(size_t, Field &) const override - { - throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Cannot get value from {}", getName()); - } + void get(size_t n, Field & res) const override; StringRef getDataAt(size_t) const override { diff --git a/src/Functions/IFunction.h b/src/Functions/IFunction.h index c3ba4be7419..d0d6b02e69d 100644 --- a/src/Functions/IFunction.h +++ b/src/Functions/IFunction.h @@ -184,7 +184,7 @@ public: /** If function isSuitableForConstantFolding then, this method will be called during query analysis * if some arguments are constants. For example logical functions (AndFunction, OrFunction) can - * return they result based on some constant arguments. + * return the result based on some constant arguments. * Arguments are passed without modifications, useDefaultImplementationForNulls, useDefaultImplementationForNothing, * useDefaultImplementationForConstants, useDefaultImplementationForLowCardinality are not applied. */ From bc79d9bad3569a94b2755ab9ea3549ff8202148a Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 10 Nov 2024 00:01:42 +0100 Subject: [PATCH 607/680] Only with analyzer --- src/Functions/FunctionsMiscellaneous.h | 11 ++++++++--- src/Interpreters/ActionsVisitor.cpp | 2 +- src/Planner/PlannerActionsVisitor.cpp | 2 +- 3 files changed, 10 insertions(+), 5 deletions(-) diff --git a/src/Functions/FunctionsMiscellaneous.h b/src/Functions/FunctionsMiscellaneous.h index 62b43386db5..cea11cfe677 100644 --- a/src/Functions/FunctionsMiscellaneous.h +++ b/src/Functions/FunctionsMiscellaneous.h @@ -113,6 +113,7 @@ public: NamesAndTypesList lambda_arguments; String return_name; DataTypePtr return_type; + bool allow_constant_folding; }; using CapturePtr = std::shared_ptr; @@ -154,9 +155,11 @@ public: /// Consequently, it allows to treat higher order functions with constant arrays and constant captured columns /// as constant expressions. /// Consequently, it allows its usage in contexts requiring constants, such as the right hand side of IN. - bool all_arguments_are_constant = std::all_of(arguments.begin(), arguments.end(), [](const auto & arg) { return arg.column->isConst(); }); + bool constant_folding = capture->allow_constant_folding + && std::all_of(arguments.begin(), arguments.end(), + [](const auto & arg) { return arg.column->isConst(); }); - if (all_arguments_are_constant) + if (constant_folding) { ColumnsWithTypeAndName arguments_resized = arguments; for (auto & elem : arguments_resized) @@ -222,7 +225,8 @@ public: const Names & captured_names, const NamesAndTypesList & lambda_arguments, const DataTypePtr & function_return_type, - const String & expression_return_name) + const String & expression_return_name, + bool allow_constant_folding) : expression_actions(std::move(expression_actions_)) { /// Check that expression does not contain unusual actions that will break columns structure. @@ -265,6 +269,7 @@ public: .lambda_arguments = lambda_arguments, .return_name = expression_return_name, .return_type = function_return_type, + .allow_constant_folding = allow_constant_folding, }); } diff --git a/src/Interpreters/ActionsVisitor.cpp b/src/Interpreters/ActionsVisitor.cpp index 65c3fe8cfcf..696021b418c 100644 --- a/src/Interpreters/ActionsVisitor.cpp +++ b/src/Interpreters/ActionsVisitor.cpp @@ -1308,7 +1308,7 @@ void ActionsMatcher::visit(const ASTFunction & node, const ASTPtr & ast, Data & String lambda_name = data.getUniqueName("__lambda"); auto function_capture = std::make_shared( - lambda_actions, captured, lambda_arguments, result_type, result_name); + lambda_actions, captured, lambda_arguments, result_type, result_name, false); data.addFunction(function_capture, captured, lambda_name); argument_types[i] = std::make_shared(lambda_type->getArgumentTypes(), result_type); diff --git a/src/Planner/PlannerActionsVisitor.cpp b/src/Planner/PlannerActionsVisitor.cpp index aa233109fa9..2cb2a242c35 100644 --- a/src/Planner/PlannerActionsVisitor.cpp +++ b/src/Planner/PlannerActionsVisitor.cpp @@ -804,7 +804,7 @@ PlannerActionsVisitorImpl::NodeNameAndNodeMinLevel PlannerActionsVisitorImpl::vi auto lambda_node_name = calculateActionNodeName(node, *planner_context); auto function_capture = std::make_shared( - lambda_actions, captured_column_names, lambda_arguments_names_and_types, lambda_node.getExpression()->getResultType(), lambda_expression_node_name); + lambda_actions, captured_column_names, lambda_arguments_names_and_types, lambda_node.getExpression()->getResultType(), lambda_expression_node_name, true); // TODO: Pass IFunctionBase here not FunctionCaptureOverloadResolver. const auto * actions_node = actions_stack[level].addFunctionIfNecessary(lambda_node_name, std::move(lambda_children), function_capture); From b70f39879d3740e671d5111854daad2b9397adc1 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Sun, 10 Nov 2024 00:02:24 +0100 Subject: [PATCH 608/680] Only with analyzer --- .../0_stateless/02961_higher_order_constant_expressions.sql | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/queries/0_stateless/02961_higher_order_constant_expressions.sql b/tests/queries/0_stateless/02961_higher_order_constant_expressions.sql index 47480010751..23b0b72f48f 100644 --- a/tests/queries/0_stateless/02961_higher_order_constant_expressions.sql +++ b/tests/queries/0_stateless/02961_higher_order_constant_expressions.sql @@ -1,3 +1,5 @@ +SET enable_analyzer = 1; + SELECT arrayMap(x -> x, [1, 2, 3]) AS x, isConstant(x); SELECT arrayMap(x -> x + 1, [1, 2, 3]) AS x, isConstant(x); SELECT arrayMap(x -> x + x, [1, 2, 3]) AS x, isConstant(x); From 55540c2119ca0dbc5d4eb51763155f27883df0b9 Mon Sep 17 00:00:00 2001 From: Eduard Karacharov Date: Sun, 10 Nov 2024 11:18:12 +0200 Subject: [PATCH 609/680] fix: transform set constant only if allowed --- src/Storages/MergeTree/KeyCondition.cpp | 7 +- src/Storages/MergeTree/KeyCondition.h | 1 + .../03269_partition_key_not_in_set.reference | 13 +++ .../03269_partition_key_not_in_set.sql | 81 +++++++++++++++++++ 4 files changed, 100 insertions(+), 2 deletions(-) create mode 100644 tests/queries/0_stateless/03269_partition_key_not_in_set.reference create mode 100644 tests/queries/0_stateless/03269_partition_key_not_in_set.sql diff --git a/src/Storages/MergeTree/KeyCondition.cpp b/src/Storages/MergeTree/KeyCondition.cpp index 17723d341fb..a2783ff4efe 100644 --- a/src/Storages/MergeTree/KeyCondition.cpp +++ b/src/Storages/MergeTree/KeyCondition.cpp @@ -1158,6 +1158,7 @@ bool KeyCondition::tryPrepareSetIndex( const RPNBuilderFunctionTreeNode & func, RPNElement & out, size_t & out_key_column_num, + bool & allow_constant_transformation, bool & is_constant_transformed) { const auto & left_arg = func.getArgumentAt(0); @@ -1184,7 +1185,9 @@ bool KeyCondition::tryPrepareSetIndex( set_transforming_chains.push_back(set_transforming_chain); } // For partition index, checking if set can be transformed to prune any partitions - else if (single_point && canSetValuesBeWrappedByFunctions(node, index_mapping.key_index, data_type, set_transforming_chain)) + else if ( + single_point && allow_constant_transformation + && canSetValuesBeWrappedByFunctions(node, index_mapping.key_index, data_type, set_transforming_chain)) { indexes_mapping.push_back(index_mapping); data_types.push_back(data_type); @@ -1954,7 +1957,7 @@ bool KeyCondition::extractAtomFromTree(const RPNBuilderTreeNode & node, RPNEleme if (functionIsInOrGlobalInOperator(func_name)) { - if (tryPrepareSetIndex(func, out, key_column_num, is_constant_transformed)) + if (tryPrepareSetIndex(func, out, key_column_num, allow_constant_transformation, is_constant_transformed)) { key_arg_pos = 0; is_set_const = true; diff --git a/src/Storages/MergeTree/KeyCondition.h b/src/Storages/MergeTree/KeyCondition.h index 8c946bd3bbd..20b40271dc2 100644 --- a/src/Storages/MergeTree/KeyCondition.h +++ b/src/Storages/MergeTree/KeyCondition.h @@ -312,6 +312,7 @@ private: const RPNBuilderFunctionTreeNode & func, RPNElement & out, size_t & out_key_column_num, + bool & allow_constant_transformation, bool & is_constant_transformed); /// Checks that the index can not be used. diff --git a/tests/queries/0_stateless/03269_partition_key_not_in_set.reference b/tests/queries/0_stateless/03269_partition_key_not_in_set.reference new file mode 100644 index 00000000000..1e34df0c77e --- /dev/null +++ b/tests/queries/0_stateless/03269_partition_key_not_in_set.reference @@ -0,0 +1,13 @@ +-- Monotonic function in partition key +48 +48 +-- Non-monotonic function in partition key +48 +48 +-- Multiple partition columns +50 +50 +96 +96 +98 +98 diff --git a/tests/queries/0_stateless/03269_partition_key_not_in_set.sql b/tests/queries/0_stateless/03269_partition_key_not_in_set.sql new file mode 100644 index 00000000000..562521fb7ee --- /dev/null +++ b/tests/queries/0_stateless/03269_partition_key_not_in_set.sql @@ -0,0 +1,81 @@ +-- Related to https://github.com/ClickHouse/ClickHouse/issues/69829 +-- +-- The main goal of the test is to assert that constant transformation +-- for set constant while partition pruning won't be performed +-- if it's not allowed (NOT IN operator case) + +DROP TABLE IF EXISTS 03269_filters; +CREATE TABLE 03269_filters ( + id Int32, + dt Date +) +engine = MergeTree +order by id; + +INSERT INTO 03269_filters +SELECT 6, '2020-01-01' +UNION ALL +SELECT 38, '2021-01-01'; + +SELECT '-- Monotonic function in partition key'; + +DROP TABLE IF EXISTS 03269_single_monotonic; +CREATE TABLE 03269_single_monotonic( + id Int32 +) +ENGINE = MergeTree +PARTITION BY intDiv(id, 10) +ORDER BY id; + +INSERT INTO 03269_single_monotonic SELECT number FROM numbers(50); + +SELECT count() FROM 03269_single_monotonic WHERE id NOT IN (6, 38); +SELECT count() FROM 03269_single_monotonic WHERE id NOT IN ( + SELECT id FROM 03269_filters +); + +DROP TABLE 03269_single_monotonic; + +SELECT '-- Non-monotonic function in partition key'; + +DROP TABLE IF EXISTS 03269_single_non_monotonic; +CREATE TABLE 03269_single_non_monotonic ( + id Int32 +) +ENGINE = MergeTree +PARTITION BY id % 10 +ORDER BY id; + +INSERT INTO 03269_single_non_monotonic SELECT number FROM numbers(50); + +SELECT count() FROM 03269_single_non_monotonic WHERE id NOT IN (6, 38); +SELECT count() FROM 03269_single_non_monotonic WHERE id NOT IN (SELECT id FROM 03269_filters); + +DROP TABLE 03269_single_non_monotonic; + +SELECT '-- Multiple partition columns'; + +DROP TABLE IF EXISTS 03269_multiple_part_cols; +CREATE TABLE 03269_multiple_part_cols ( + id Int32, + dt Date, +) +ENGINE = MergeTree +PARTITION BY (dt, intDiv(id, 10)) +ORDER BY id; + +INSERT INTO 03269_multiple_part_cols +SELECT number, '2020-01-01' FROM numbers(50) +UNION ALL +SELECT number, '2021-01-01' FROM numbers(50); + +SELECT count() FROM 03269_multiple_part_cols WHERE dt NOT IN ('2020-01-01'); +SELECT count() FROM 03269_multiple_part_cols WHERE dt NOT IN (SELECT dt FROM 03269_filters WHERE dt < '2021-01-01'); + +SELECT count() FROM 03269_multiple_part_cols WHERE id NOT IN (6, 38); +SELECT count() FROM 03269_multiple_part_cols WHERE id NOT IN (SELECT id FROM 03269_filters); + +SELECT count() FROM 03269_multiple_part_cols WHERE (id, dt) NOT IN ((6, '2020-01-01'), (38, '2021-01-01')); +SELECT count() FROM 03269_multiple_part_cols WHERE (id, dt) NOT IN (SELECT id, dt FROM 03269_filters); + +DROP TABLE 03269_multiple_part_cols; From f2d6b1db7fb8b8eee52e2a33ce6f88648fe1c863 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Sun, 10 Nov 2024 12:39:10 +0000 Subject: [PATCH 610/680] Better --- contrib/SimSIMD | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/SimSIMD b/contrib/SimSIMD index bb0bd2e7137..fa60f1b8e35 160000 --- a/contrib/SimSIMD +++ b/contrib/SimSIMD @@ -1 +1 @@ -Subproject commit bb0bd2e7137f02c555341d7c93124ed19f3c24fb +Subproject commit fa60f1b8e3582c50978f0ae86c2ebb6c9af957f3 From d1e638da6e65a2f0de4aa72b78fd894c090606de Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Sun, 10 Nov 2024 15:12:21 +0100 Subject: [PATCH 611/680] Let's name cherry-pick branches the same way as backports --- tests/ci/cherry_pick.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/ci/cherry_pick.py b/tests/ci/cherry_pick.py index a796f63de6c..9bdc184f661 100644 --- a/tests/ci/cherry_pick.py +++ b/tests/ci/cherry_pick.py @@ -97,7 +97,7 @@ close it. self.pr = pr self.repo = repo - self.cherrypick_branch = f"cherrypick/{name}/{pr.merge_commit_sha}" + self.cherrypick_branch = f"cherrypick/{name}/{pr.number}" self.backport_branch = f"backport/{name}/{pr.number}" self.cherrypick_pr = None # type: Optional[PullRequest] self.backport_pr = None # type: Optional[PullRequest] From b6b850a2f11301272ee28fe2274733c2cdb0c7c6 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Sun, 10 Nov 2024 17:03:35 +0000 Subject: [PATCH 612/680] Docs: Add row and byte sizes of tables --- docs/en/getting-started/example-datasets/tpch.md | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/docs/en/getting-started/example-datasets/tpch.md b/docs/en/getting-started/example-datasets/tpch.md index 5fa0d779ecd..379b92cbe9a 100644 --- a/docs/en/getting-started/example-datasets/tpch.md +++ b/docs/en/getting-started/example-datasets/tpch.md @@ -33,6 +33,21 @@ Then, generate the data. Parameter `-s` specifies the scale factor. For example, ./dbgen -s 100 ``` +Detailed table sizes with scale factor 100: + +| Table | size (in rows) | size (compressed in ClickHouse) | +|----------|----------------|---------------------------------| +| nation | 25 | 2 kB | +| region | 5 | 1 kB | +| part | 20.000.000 | 895 MB | +| supplier | 1.000.000 | 75 MB | +| partsupp | 80.000.000 | 4.37 GB | +| customer | 15.000.000 | 1.19 GB | +| orders | 150.000.000 | 6.15 GB | +| lineitem | 600.00.00 | 26.69 GB | + +(The table sizes in ClickHouse are taken from `system.tables.total_bytes` and based on below table definitions. + Now create tables in ClickHouse. We stick as closely as possible to the rules of the TPC-H specification: From 3668a78589821d89f8f7cce92e6c2bc54fff6ea3 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Sun, 10 Nov 2024 17:24:00 +0000 Subject: [PATCH 613/680] Fix spelling --- utils/check-style/aspell-ignore/en/aspell-dict.txt | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt index a08143467cd..a58b5e9ff58 100644 --- a/utils/check-style/aspell-ignore/en/aspell-dict.txt +++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt @@ -186,7 +186,6 @@ ComplexKeyCache ComplexKeyDirect ComplexKeyHashed Composable -composable ConcurrencyControlAcquired ConcurrencyControlSoftLimit Config @@ -405,12 +404,12 @@ ITION Identifiant IdentifierQuotingRule IdentifierQuotingStyle -Incrementing -IndexesAreNeighbors -InfluxDB InJodaSyntax InJodaSyntaxOrNull InJodaSyntaxOrZero +Incrementing +IndexesAreNeighbors +InfluxDB Instana IntN Integrations @@ -1475,6 +1474,7 @@ combinator combinators comparising composable +composable compressability concat concatAssumeInjective @@ -2355,6 +2355,7 @@ parsedatetime parsers partitionID partitionId +partsupp pathFull pclmulqdq pcre From f9fa5ed515daaf6bb30ed13fd882a4c92cb84351 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Sun, 10 Nov 2024 20:38:51 +0000 Subject: [PATCH 614/680] Docs: Steps to populate TPC-H tables from S3 --- .../getting-started/example-datasets/tpch.md | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) diff --git a/docs/en/getting-started/example-datasets/tpch.md b/docs/en/getting-started/example-datasets/tpch.md index 379b92cbe9a..c0bf54a5a7c 100644 --- a/docs/en/getting-started/example-datasets/tpch.md +++ b/docs/en/getting-started/example-datasets/tpch.md @@ -166,10 +166,26 @@ clickhouse-client --format_csv_delimiter '|' --query "INSERT INTO orders FORMAT clickhouse-client --format_csv_delimiter '|' --query "INSERT INTO lineitem FORMAT CSV" < lineitem.tbl ``` -The queries are generated by `./qgen -s `. Example queries for `s = 100`: +:::note +Instead of using tpch-kit and generating the tables by yourself, you can alternatively import the data from a public S3 bucket. Make sure +to create empty tables first using above `CREATE` statements. + +```sql +INSERT INTO nation SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/nation.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; +INSERT INTO region SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/region.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; +INSERT INTO part SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/part.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; +INSERT INTO supplier SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/supplier.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; +INSERT INTO partsupp SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/partsupp.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; +INSERT INTO customer SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/customer.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; +INSERT INTO orders SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/orders.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; +INSERT INTO lineitem SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/lineitem.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; +```` +::: ## Queries +The queries are generated by `./qgen -s `. Example queries for `s = 100`: + **Correctness** The result of the queries agrees with the official results unless mentioned otherwise. To verify, generate a TPC-H database with scale From 892d43bd7d57faed05cdcef77e684a09dbad3e36 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Sun, 10 Nov 2024 20:50:07 +0000 Subject: [PATCH 615/680] SF 1 vs. 100 --- .../getting-started/example-datasets/tpch.md | 27 +++++++++++++------ 1 file changed, 19 insertions(+), 8 deletions(-) diff --git a/docs/en/getting-started/example-datasets/tpch.md b/docs/en/getting-started/example-datasets/tpch.md index c0bf54a5a7c..de2c425b402 100644 --- a/docs/en/getting-started/example-datasets/tpch.md +++ b/docs/en/getting-started/example-datasets/tpch.md @@ -171,14 +171,25 @@ Instead of using tpch-kit and generating the tables by yourself, you can alterna to create empty tables first using above `CREATE` statements. ```sql -INSERT INTO nation SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/nation.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; -INSERT INTO region SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/region.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; -INSERT INTO part SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/part.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; -INSERT INTO supplier SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/supplier.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; -INSERT INTO partsupp SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/partsupp.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; -INSERT INTO customer SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/customer.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; -INSERT INTO orders SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/orders.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; -INSERT INTO lineitem SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/lineitem.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter='|', input_format_defaults_for_omitted_fields=1, input_format_csv_empty_as_default=1; +-- Scaling factor 1 +INSERT INTO nation SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/1/nation.tbl', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO region SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/1/region.tbl', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO part SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/1/part.tbl', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO supplier SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/1/supplier.tbl', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO partsupp SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/1/partsupp.tbl', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO customer SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/1/customer.tbl', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO orders SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/1/orders.tbl', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO lineitem SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/1/lineitem.tbl', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; + +-- Scaling factor 100 +INSERT INTO nation SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/nation.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO region SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/region.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO part SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/part.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO supplier SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/supplier.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO partsupp SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/partsupp.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO customer SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/customer.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO orders SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/orders.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; +INSERT INTO lineitem SELECT * FROM s3('https://clickhouse-datasets.s3.amazonaws.com/h/100/lineitem.tbl.gz', NOSIGN, CSV) SETTINGS format_csv_delimiter = '|', input_format_defaults_for_omitted_fields = 1, input_format_csv_empty_as_default = 1; ```` ::: From a74f491df3c217bf4132b08118e4708b05d3bf60 Mon Sep 17 00:00:00 2001 From: Shaun Struwig <41984034+Blargian@users.noreply.github.com> Date: Sun, 10 Nov 2024 22:02:01 +0100 Subject: [PATCH 616/680] Fix typo --- docs/en/sql-reference/functions/index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/sql-reference/functions/index.md b/docs/en/sql-reference/functions/index.md index c0256ba4735..04a87c369ab 100644 --- a/docs/en/sql-reference/functions/index.md +++ b/docs/en/sql-reference/functions/index.md @@ -24,7 +24,7 @@ All expressions in a query that have the same AST (the same record or same resul ## Types of Results -All functions return a single return as the result (not several values, and not zero values). The type of result is usually defined only by the types of arguments, not by the values. Exceptions are the tupleElement function (the a.N operator), and the toFixedString function. +All functions return a single value as the result (not several values, and not zero values). The type of result is usually defined only by the types of arguments, not by the values. Exceptions are the tupleElement function (the a.N operator), and the toFixedString function. ## Constants From 866e4daeecb301030e3f89eb56395c1156cb840d Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 11 Nov 2024 01:10:49 +0100 Subject: [PATCH 617/680] Update index.md --- .../aggregate-functions/reference/index.md | 174 +++++++++--------- 1 file changed, 85 insertions(+), 89 deletions(-) diff --git a/docs/en/sql-reference/aggregate-functions/reference/index.md b/docs/en/sql-reference/aggregate-functions/reference/index.md index 2dce0afe2e1..d7b287f764b 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/index.md +++ b/docs/en/sql-reference/aggregate-functions/reference/index.md @@ -7,119 +7,115 @@ toc_hidden: true # List of Aggregate Functions -Standard aggregate functions: - -- [count](../reference/count.md) -- [min](../reference/min.md) -- [max](../reference/max.md) -- [sum](../reference/sum.md) -- [avg](../reference/avg.md) -- [any](../reference/any.md) -- [stddevPop](../reference/stddevpop.md) -- [stddevPopStable](../reference/stddevpopstable.md) -- [stddevSamp](../reference/stddevsamp.md) -- [stddevSampStable](../reference/stddevsampstable.md) -- [varPop](../reference/varpop.md) -- [varSamp](../reference/varsamp.md) -- [corr](../reference/corr.md) -- [corr](../reference/corrstable.md) -- [corrMatrix](../reference/corrmatrix.md) -- [covarPop](../reference/covarpop.md) -- [covarStable](../reference/covarpopstable.md) -- [covarPopMatrix](../reference/covarpopmatrix.md) -- [covarSamp](../reference/covarsamp.md) -- [covarSampStable](../reference/covarsampstable.md) -- [covarSampMatrix](../reference/covarsampmatrix.md) -- [entropy](../reference/entropy.md) -- [exponentialMovingAverage](../reference/exponentialmovingaverage.md) -- [intervalLengthSum](../reference/intervalLengthSum.md) -- [kolmogorovSmirnovTest](../reference/kolmogorovsmirnovtest.md) -- [mannwhitneyutest](../reference/mannwhitneyutest.md) -- [median](../reference/median.md) -- [rankCorr](../reference/rankCorr.md) -- [sumKahan](../reference/sumkahan.md) -- [studentTTest](../reference/studentttest.md) -- [welchTTest](../reference/welchttest.md) - -ClickHouse-specific aggregate functions: +ClickHouse supports all standard SQL functions (sum, avg, min, max, count) and a wide range of aggregate functions for various applications: - [aggThrow](../reference/aggthrow.md) - [analysisOfVariance](../reference/analysis_of_variance.md) -- [any](../reference/any.md) - [anyHeavy](../reference/anyheavy.md) - [anyLast](../reference/anylast.md) -- [boundingRatio](../reference/boundrat.md) -- [first_value](../reference/first_value.md) -- [last_value](../reference/last_value.md) -- [argMin](../reference/argmin.md) +- [any](../reference/any.md) - [argMax](../reference/argmax.md) +- [argMin](../reference/argmin.md) - [avgWeighted](../reference/avgweighted.md) -- [topK](../reference/topk.md) -- [topKWeighted](../reference/topkweighted.md) -- [deltaSum](../reference/deltasum.md) +- [avg](../reference/avg.md) +- [boundingRatio](../reference/boundrat.md) +- [categoricalInformationValue](../reference/categoricalinformationvalue.md) +- [contingency](../reference/contingency.md) +- [corrMatrix](../reference/corrmatrix.md) +- [corr](../reference/corr.md) +- [corr](../reference/corrstable.md) +- [count](../reference/count.md) +- [covarPopMatrix](../reference/covarpopmatrix.md) +- [covarPop](../reference/covarpop.md) +- [covarSampMatrix](../reference/covarsampmatrix.md) +- [covarSampStable](../reference/covarsampstable.md) +- [covarSamp](../reference/covarsamp.md) +- [covarStable](../reference/covarpopstable.md) +- [cramersVBiasCorrected](../reference/cramersvbiascorrected.md) +- [cramersV](../reference/cramersv.md) - [deltaSumTimestamp](../reference/deltasumtimestamp.md) +- [deltaSum](../reference/deltasum.md) +- [entropy](../reference/entropy.md) +- [exponentialMovingAverage](../reference/exponentialmovingaverage.md) +- [first_value](../reference/first_value.md) - [flameGraph](../reference/flame_graph.md) -- [groupArray](../reference/grouparray.md) -- [groupArrayLast](../reference/grouparraylast.md) -- [groupUniqArray](../reference/groupuniqarray.md) - [groupArrayInsertAt](../reference/grouparrayinsertat.md) +- [groupArrayIntersect](../reference/grouparrayintersect.md) +- [groupArrayLast](../reference/grouparraylast.md) - [groupArrayMovingAvg](../reference/grouparraymovingavg.md) - [groupArrayMovingSum](../reference/grouparraymovingsum.md) - [groupArraySample](../reference/grouparraysample.md) - [groupArraySorted](../reference/grouparraysorted.md) -- [groupArrayIntersect](../reference/grouparrayintersect.md) +- [groupArray](../reference/grouparray.md) - [groupBitAnd](../reference/groupbitand.md) - [groupBitOr](../reference/groupbitor.md) - [groupBitXor](../reference/groupbitxor.md) -- [groupBitmap](../reference/groupbitmap.md) - [groupBitmapAnd](../reference/groupbitmapand.md) - [groupBitmapOr](../reference/groupbitmapor.md) - [groupBitmapXor](../reference/groupbitmapxor.md) -- [sumWithOverflow](../reference/sumwithoverflow.md) -- [sumMap](../reference/summap.md) -- [sumMapWithOverflow](../reference/summapwithoverflow.md) -- [sumMapFiltered](../parametric-functions.md/#summapfiltered) -- [sumMapFilteredWithOverflow](../parametric-functions.md/#summapfilteredwithoverflow) -- [minMap](../reference/minmap.md) -- [maxMap](../reference/maxmap.md) -- [skewSamp](../reference/skewsamp.md) -- [skewPop](../reference/skewpop.md) -- [kurtSamp](../reference/kurtsamp.md) +- [groupBitmap](../reference/groupbitmap.md) +- [groupUniqArray](../reference/groupuniqarray.md) +- [intervalLengthSum](../reference/intervalLengthSum.md) +- [kolmogorovSmirnovTest](../reference/kolmogorovsmirnovtest.md) - [kurtPop](../reference/kurtpop.md) -- [uniq](../reference/uniq.md) -- [uniqExact](../reference/uniqexact.md) -- [uniqCombined](../reference/uniqcombined.md) -- [uniqCombined64](../reference/uniqcombined64.md) -- [uniqHLL12](../reference/uniqhll12.md) -- [uniqTheta](../reference/uniqthetasketch.md) -- [quantile](../reference/quantile.md) -- [quantiles](../reference/quantiles.md) -- [quantileExact](../reference/quantileexact.md) -- [quantileExactLow](../reference/quantileexact.md#quantileexactlow) -- [quantileExactHigh](../reference/quantileexact.md#quantileexacthigh) -- [quantileExactWeighted](../reference/quantileexactweighted.md) -- [quantileTiming](../reference/quantiletiming.md) -- [quantileTimingWeighted](../reference/quantiletimingweighted.md) -- [quantileDeterministic](../reference/quantiledeterministic.md) -- [quantileTDigest](../reference/quantiletdigest.md) -- [quantileTDigestWeighted](../reference/quantiletdigestweighted.md) -- [quantileBFloat16](../reference/quantilebfloat16.md#quantilebfloat16) -- [quantileBFloat16Weighted](../reference/quantilebfloat16.md#quantilebfloat16weighted) -- [quantileDD](../reference/quantileddsketch.md#quantileddsketch) -- [simpleLinearRegression](../reference/simplelinearregression.md) -- [singleValueOrNull](../reference/singlevalueornull.md) -- [stochasticLinearRegression](../reference/stochasticlinearregression.md) -- [stochasticLogisticRegression](../reference/stochasticlogisticregression.md) -- [categoricalInformationValue](../reference/categoricalinformationvalue.md) -- [contingency](../reference/contingency.md) -- [cramersV](../reference/cramersv.md) -- [cramersVBiasCorrected](../reference/cramersvbiascorrected.md) -- [theilsU](../reference/theilsu.md) -- [maxIntersections](../reference/maxintersections.md) +- [kurtSamp](../reference/kurtsamp.md) +- [largestTriangleThreeBuckets](../reference/largestTriangleThreeBuckets.md) +- [last_value](../reference/last_value.md) +- [mannwhitneyutest](../reference/mannwhitneyutest.md) - [maxIntersectionsPosition](../reference/maxintersectionsposition.md) +- [maxIntersections](../reference/maxintersections.md) +- [maxMap](../reference/maxmap.md) +- [max](../reference/max.md) - [meanZTest](../reference/meanztest.md) +- [median](../reference/median.md) +- [minMap](../reference/minmap.md) +- [min](../reference/min.md) +- [quantileBFloat16Weighted](../reference/quantilebfloat16.md#quantilebfloat16weighted) +- [quantileBFloat16](../reference/quantilebfloat16.md#quantilebfloat16) +- [quantileDD](../reference/quantileddsketch.md#quantileddsketch) +- [quantileDeterministic](../reference/quantiledeterministic.md) +- [quantileExactHigh](../reference/quantileexact.md#quantileexacthigh) +- [quantileExactLow](../reference/quantileexact.md#quantileexactlow) +- [quantileExactWeighted](../reference/quantileexactweighted.md) +- [quantileExact](../reference/quantileexact.md) - [quantileGK](../reference/quantileGK.md) - [quantileInterpolatedWeighted](../reference/quantileinterpolatedweighted.md) +- [quantileTDigestWeighted](../reference/quantiletdigestweighted.md) +- [quantileTDigest](../reference/quantiletdigest.md) +- [quantileTimingWeighted](../reference/quantiletimingweighted.md) +- [quantileTiming](../reference/quantiletiming.md) +- [quantile](../reference/quantile.md) +- [quantiles](../reference/quantiles.md) +- [rankCorr](../reference/rankCorr.md) +- [simpleLinearRegression](../reference/simplelinearregression.md) +- [singleValueOrNull](../reference/singlevalueornull.md) +- [skewPop](../reference/skewpop.md) +- [skewSamp](../reference/skewsamp.md) - [sparkBar](../reference/sparkbar.md) +- [stddevPopStable](../reference/stddevpopstable.md) +- [stddevPop](../reference/stddevpop.md) +- [stddevSampStable](../reference/stddevsampstable.md) +- [stddevSamp](../reference/stddevsamp.md) +- [stochasticLinearRegression](../reference/stochasticlinearregression.md) +- [stochasticLogisticRegression](../reference/stochasticlogisticregression.md) +- [studentTTest](../reference/studentttest.md) - [sumCount](../reference/sumcount.md) -- [largestTriangleThreeBuckets](../reference/largestTriangleThreeBuckets.md) +- [sumKahan](../reference/sumkahan.md) +- [sumMapFilteredWithOverflow](../parametric-functions.md/#summapfilteredwithoverflow) +- [sumMapFiltered](../parametric-functions.md/#summapfiltered) +- [sumMapWithOverflow](../reference/summapwithoverflow.md) +- [sumMap](../reference/summap.md) +- [sumWithOverflow](../reference/sumwithoverflow.md) +- [sum](../reference/sum.md) +- [theilsU](../reference/theilsu.md) +- [topKWeighted](../reference/topkweighted.md) +- [topK](../reference/topk.md) +- [uniqCombined64](../reference/uniqcombined64.md) +- [uniqCombined](../reference/uniqcombined.md) +- [uniqExact](../reference/uniqexact.md) +- [uniqHLL12](../reference/uniqhll12.md) +- [uniqTheta](../reference/uniqthetasketch.md) +- [uniq](../reference/uniq.md) +- [varPop](../reference/varpop.md) +- [varSamp](../reference/varsamp.md) +- [welchTTest](../reference/welchttest.md) From 2b20c2d2f22f9a399f4f43f0920f6b0df978c1a9 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Mon, 11 Nov 2024 01:46:00 +0100 Subject: [PATCH 618/680] Fix a race --- src/Databases/DatabaseAtomic.cpp | 14 +++++++++----- src/Databases/DatabaseAtomic.h | 2 +- src/Databases/DatabaseOnDisk.cpp | 8 ++++++-- src/Databases/DatabaseOnDisk.h | 2 +- 4 files changed, 17 insertions(+), 9 deletions(-) diff --git a/src/Databases/DatabaseAtomic.cpp b/src/Databases/DatabaseAtomic.cpp index 88727d0389e..bd077ccd7b5 100644 --- a/src/Databases/DatabaseAtomic.cpp +++ b/src/Databases/DatabaseAtomic.cpp @@ -6,7 +6,6 @@ #include #include #include -#include #include #include #include @@ -19,6 +18,7 @@ #include #include + namespace fs = std::filesystem; namespace DB @@ -69,9 +69,13 @@ DatabaseAtomic::DatabaseAtomic(String name_, String metadata_path_, UUID uuid, C void DatabaseAtomic::createDirectories() { - if (database_atomic_directories_created.test_and_set()) - return; - DatabaseOnDisk::createDirectories(); + std::lock_guard lock(mutex); + createDirectoriesUnlocked(); +} + +void DatabaseAtomic::createDirectoriesUnlocked() +{ + DatabaseOnDisk::createDirectoriesUnlocked(); fs::create_directories(fs::path(getContext()->getPath()) / "metadata"); fs::create_directories(path_to_table_symlinks); tryCreateMetadataSymlink(); @@ -113,9 +117,9 @@ void DatabaseAtomic::drop(ContextPtr) void DatabaseAtomic::attachTable(ContextPtr /* context_ */, const String & name, const StoragePtr & table, const String & relative_table_path) { assert(relative_table_path != data_path && !relative_table_path.empty()); - createDirectories(); DetachedTables not_in_use; std::lock_guard lock(mutex); + createDirectoriesUnlocked(); not_in_use = cleanupDetachedTables(); auto table_id = table->getStorageID(); assertDetachedTableNotInUse(table_id.uuid); diff --git a/src/Databases/DatabaseAtomic.h b/src/Databases/DatabaseAtomic.h index 3d0b74e31a0..7e909128635 100644 --- a/src/Databases/DatabaseAtomic.h +++ b/src/Databases/DatabaseAtomic.h @@ -76,8 +76,8 @@ protected: using DetachedTables = std::unordered_map; [[nodiscard]] DetachedTables cleanupDetachedTables() TSA_REQUIRES(mutex); - std::atomic_flag database_atomic_directories_created = ATOMIC_FLAG_INIT; void createDirectories(); + void createDirectoriesUnlocked() TSA_REQUIRES(mutex); void tryCreateMetadataSymlink(); diff --git a/src/Databases/DatabaseOnDisk.cpp b/src/Databases/DatabaseOnDisk.cpp index 2f4280fe485..93ecf9cf11c 100644 --- a/src/Databases/DatabaseOnDisk.cpp +++ b/src/Databases/DatabaseOnDisk.cpp @@ -185,8 +185,12 @@ DatabaseOnDisk::DatabaseOnDisk( void DatabaseOnDisk::createDirectories() { - if (directories_created.test_and_set()) - return; + std::lock_guard lock(mutex); + createDirectoriesUnlocked(); +} + +void DatabaseOnDisk::createDirectoriesUnlocked() +{ fs::create_directories(std::filesystem::path(getContext()->getPath()) / data_path); fs::create_directories(metadata_path); } diff --git a/src/Databases/DatabaseOnDisk.h b/src/Databases/DatabaseOnDisk.h index 0c0ecf76a26..1e11d21cc87 100644 --- a/src/Databases/DatabaseOnDisk.h +++ b/src/Databases/DatabaseOnDisk.h @@ -99,8 +99,8 @@ protected: virtual void removeDetachedPermanentlyFlag(ContextPtr context, const String & table_name, const String & table_metadata_path, bool attach); virtual void setDetachedTableNotInUseForce(const UUID & /*uuid*/) {} - std::atomic_flag directories_created = ATOMIC_FLAG_INIT; void createDirectories(); + void createDirectoriesUnlocked() TSA_REQUIRES(mutex); const String metadata_path; const String data_path; From f57bf2ee15fe93377b858efec767627321a69887 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Mon, 11 Nov 2024 09:38:26 +0000 Subject: [PATCH 619/680] Fix trash in the docs, pt. II --- .../getting-started/example-datasets/tpch.md | 2 +- .../aggregate-functions/reference/index.md | 113 +----------------- .../data-types/aggregatefunction.md | 4 +- docs/en/sql-reference/data-types/index.md | 29 +---- docs/en/sql-reference/data-types/json.md | 2 +- .../data-types/simpleaggregatefunction.md | 4 +- docs/en/sql-reference/functions/geo/index.md | 68 +---------- .../sql-reference/statements/create/index.md | 14 +-- docs/en/sql-reference/statements/index.md | 25 +--- 9 files changed, 16 insertions(+), 245 deletions(-) diff --git a/docs/en/getting-started/example-datasets/tpch.md b/docs/en/getting-started/example-datasets/tpch.md index de2c425b402..3ea4bffec38 100644 --- a/docs/en/getting-started/example-datasets/tpch.md +++ b/docs/en/getting-started/example-datasets/tpch.md @@ -46,7 +46,7 @@ Detailed table sizes with scale factor 100: | orders | 150.000.000 | 6.15 GB | | lineitem | 600.00.00 | 26.69 GB | -(The table sizes in ClickHouse are taken from `system.tables.total_bytes` and based on below table definitions. +(Compressed sizes in ClickHouse are taken from `system.tables.total_bytes` and based on below table definitions.) Now create tables in ClickHouse. diff --git a/docs/en/sql-reference/aggregate-functions/reference/index.md b/docs/en/sql-reference/aggregate-functions/reference/index.md index d7b287f764b..ee8f0d5882e 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/index.md +++ b/docs/en/sql-reference/aggregate-functions/reference/index.md @@ -7,115 +7,4 @@ toc_hidden: true # List of Aggregate Functions -ClickHouse supports all standard SQL functions (sum, avg, min, max, count) and a wide range of aggregate functions for various applications: - -- [aggThrow](../reference/aggthrow.md) -- [analysisOfVariance](../reference/analysis_of_variance.md) -- [anyHeavy](../reference/anyheavy.md) -- [anyLast](../reference/anylast.md) -- [any](../reference/any.md) -- [argMax](../reference/argmax.md) -- [argMin](../reference/argmin.md) -- [avgWeighted](../reference/avgweighted.md) -- [avg](../reference/avg.md) -- [boundingRatio](../reference/boundrat.md) -- [categoricalInformationValue](../reference/categoricalinformationvalue.md) -- [contingency](../reference/contingency.md) -- [corrMatrix](../reference/corrmatrix.md) -- [corr](../reference/corr.md) -- [corr](../reference/corrstable.md) -- [count](../reference/count.md) -- [covarPopMatrix](../reference/covarpopmatrix.md) -- [covarPop](../reference/covarpop.md) -- [covarSampMatrix](../reference/covarsampmatrix.md) -- [covarSampStable](../reference/covarsampstable.md) -- [covarSamp](../reference/covarsamp.md) -- [covarStable](../reference/covarpopstable.md) -- [cramersVBiasCorrected](../reference/cramersvbiascorrected.md) -- [cramersV](../reference/cramersv.md) -- [deltaSumTimestamp](../reference/deltasumtimestamp.md) -- [deltaSum](../reference/deltasum.md) -- [entropy](../reference/entropy.md) -- [exponentialMovingAverage](../reference/exponentialmovingaverage.md) -- [first_value](../reference/first_value.md) -- [flameGraph](../reference/flame_graph.md) -- [groupArrayInsertAt](../reference/grouparrayinsertat.md) -- [groupArrayIntersect](../reference/grouparrayintersect.md) -- [groupArrayLast](../reference/grouparraylast.md) -- [groupArrayMovingAvg](../reference/grouparraymovingavg.md) -- [groupArrayMovingSum](../reference/grouparraymovingsum.md) -- [groupArraySample](../reference/grouparraysample.md) -- [groupArraySorted](../reference/grouparraysorted.md) -- [groupArray](../reference/grouparray.md) -- [groupBitAnd](../reference/groupbitand.md) -- [groupBitOr](../reference/groupbitor.md) -- [groupBitXor](../reference/groupbitxor.md) -- [groupBitmapAnd](../reference/groupbitmapand.md) -- [groupBitmapOr](../reference/groupbitmapor.md) -- [groupBitmapXor](../reference/groupbitmapxor.md) -- [groupBitmap](../reference/groupbitmap.md) -- [groupUniqArray](../reference/groupuniqarray.md) -- [intervalLengthSum](../reference/intervalLengthSum.md) -- [kolmogorovSmirnovTest](../reference/kolmogorovsmirnovtest.md) -- [kurtPop](../reference/kurtpop.md) -- [kurtSamp](../reference/kurtsamp.md) -- [largestTriangleThreeBuckets](../reference/largestTriangleThreeBuckets.md) -- [last_value](../reference/last_value.md) -- [mannwhitneyutest](../reference/mannwhitneyutest.md) -- [maxIntersectionsPosition](../reference/maxintersectionsposition.md) -- [maxIntersections](../reference/maxintersections.md) -- [maxMap](../reference/maxmap.md) -- [max](../reference/max.md) -- [meanZTest](../reference/meanztest.md) -- [median](../reference/median.md) -- [minMap](../reference/minmap.md) -- [min](../reference/min.md) -- [quantileBFloat16Weighted](../reference/quantilebfloat16.md#quantilebfloat16weighted) -- [quantileBFloat16](../reference/quantilebfloat16.md#quantilebfloat16) -- [quantileDD](../reference/quantileddsketch.md#quantileddsketch) -- [quantileDeterministic](../reference/quantiledeterministic.md) -- [quantileExactHigh](../reference/quantileexact.md#quantileexacthigh) -- [quantileExactLow](../reference/quantileexact.md#quantileexactlow) -- [quantileExactWeighted](../reference/quantileexactweighted.md) -- [quantileExact](../reference/quantileexact.md) -- [quantileGK](../reference/quantileGK.md) -- [quantileInterpolatedWeighted](../reference/quantileinterpolatedweighted.md) -- [quantileTDigestWeighted](../reference/quantiletdigestweighted.md) -- [quantileTDigest](../reference/quantiletdigest.md) -- [quantileTimingWeighted](../reference/quantiletimingweighted.md) -- [quantileTiming](../reference/quantiletiming.md) -- [quantile](../reference/quantile.md) -- [quantiles](../reference/quantiles.md) -- [rankCorr](../reference/rankCorr.md) -- [simpleLinearRegression](../reference/simplelinearregression.md) -- [singleValueOrNull](../reference/singlevalueornull.md) -- [skewPop](../reference/skewpop.md) -- [skewSamp](../reference/skewsamp.md) -- [sparkBar](../reference/sparkbar.md) -- [stddevPopStable](../reference/stddevpopstable.md) -- [stddevPop](../reference/stddevpop.md) -- [stddevSampStable](../reference/stddevsampstable.md) -- [stddevSamp](../reference/stddevsamp.md) -- [stochasticLinearRegression](../reference/stochasticlinearregression.md) -- [stochasticLogisticRegression](../reference/stochasticlogisticregression.md) -- [studentTTest](../reference/studentttest.md) -- [sumCount](../reference/sumcount.md) -- [sumKahan](../reference/sumkahan.md) -- [sumMapFilteredWithOverflow](../parametric-functions.md/#summapfilteredwithoverflow) -- [sumMapFiltered](../parametric-functions.md/#summapfiltered) -- [sumMapWithOverflow](../reference/summapwithoverflow.md) -- [sumMap](../reference/summap.md) -- [sumWithOverflow](../reference/sumwithoverflow.md) -- [sum](../reference/sum.md) -- [theilsU](../reference/theilsu.md) -- [topKWeighted](../reference/topkweighted.md) -- [topK](../reference/topk.md) -- [uniqCombined64](../reference/uniqcombined64.md) -- [uniqCombined](../reference/uniqcombined.md) -- [uniqExact](../reference/uniqexact.md) -- [uniqHLL12](../reference/uniqhll12.md) -- [uniqTheta](../reference/uniqthetasketch.md) -- [uniq](../reference/uniq.md) -- [varPop](../reference/varpop.md) -- [varSamp](../reference/varsamp.md) -- [welchTTest](../reference/welchttest.md) +ClickHouse supports all standard SQL aggregate functions ([sum](../reference/sum.md), [avg](../reference/avg.md), [min](../reference/min.md), [max](../reference/max.md), [count](../reference/count.md)), as well as a wide range of other aggregate functions. diff --git a/docs/en/sql-reference/data-types/aggregatefunction.md b/docs/en/sql-reference/data-types/aggregatefunction.md index 37f0d0e50ae..4cad27db68b 100644 --- a/docs/en/sql-reference/data-types/aggregatefunction.md +++ b/docs/en/sql-reference/data-types/aggregatefunction.md @@ -6,7 +6,9 @@ sidebar_label: AggregateFunction # AggregateFunction -Aggregate functions can have an implementation-defined intermediate state that can be serialized to an `AggregateFunction(...)` data type and stored in a table, usually, by means of [a materialized view](../../sql-reference/statements/create/view.md). The common way to produce an aggregate function state is by calling the aggregate function with the `-State` suffix. To get the final result of aggregation in the future, you must use the same aggregate function with the `-Merge`suffix. +Aggregate functions have an implementation-defined intermediate state that can be serialized to an `AggregateFunction(...)` data type and stored in a table, usually, by means of [a materialized view](../../sql-reference/statements/create/view.md). +The common way to produce an aggregate function state is by calling the aggregate function with the `-State` suffix. +To get the final result of aggregation in the future, you must use the same aggregate function with the `-Merge`suffix. `AggregateFunction(name, types_of_arguments...)` — parametric data type. diff --git a/docs/en/sql-reference/data-types/index.md b/docs/en/sql-reference/data-types/index.md index 2b89dd145e6..134678f71bb 100644 --- a/docs/en/sql-reference/data-types/index.md +++ b/docs/en/sql-reference/data-types/index.md @@ -6,29 +6,8 @@ sidebar_position: 1 # Data Types in ClickHouse -ClickHouse can store various kinds of data in table cells. This section describes the supported data types and special considerations for using and/or implementing them if any. +This section describes the data types supported by ClickHouse, for example [integers](int-uint.md), [floats](float.md) and [strings](string.md). -:::note -You can check whether a data type name is case-sensitive in the [system.data_type_families](../../operations/system-tables/data_type_families.md#system_tables-data_type_families) table. -::: - -ClickHouse data types include: - -- **Integer types**: [signed and unsigned integers](./int-uint.md) (`UInt8`, `UInt16`, `UInt32`, `UInt64`, `UInt128`, `UInt256`, `Int8`, `Int16`, `Int32`, `Int64`, `Int128`, `Int256`) -- **Floating-point numbers**: [floats](./float.md)(`Float32` and `Float64`) and [`Decimal` values](./decimal.md) -- **Boolean**: ClickHouse has a [`Boolean` type](./boolean.md) -- **Strings**: [`String`](./string.md) and [`FixedString`](./fixedstring.md) -- **Dates**: use [`Date`](./date.md) and [`Date32`](./date32.md) for days, and [`DateTime`](./datetime.md) and [`DateTime64`](./datetime64.md) for instances in time -- **Object**: the [`Object`](./json.md) stores a JSON document in a single column (deprecated) -- **JSON**: the [`JSON` object](./newjson.md) stores a JSON document in a single column -- **UUID**: a performant option for storing [`UUID` values](./uuid.md) -- **Low cardinality types**: use an [`Enum`](./enum.md) when you have a handful of unique values, or use [`LowCardinality`](./lowcardinality.md) when you have up to 10,000 unique values of a column -- **Arrays**: any column can be defined as an [`Array` of values](./array.md) -- **Maps**: use [`Map`](./map.md) for storing key/value pairs -- **Aggregation function types**: use [`SimpleAggregateFunction`](./simpleaggregatefunction.md) and [`AggregateFunction`](./aggregatefunction.md) for storing the intermediate status of aggregate function results -- **Nested data structures**: A [`Nested` data structure](./nested-data-structures/index.md) is like a table inside a cell -- **Tuples**: A [`Tuple` of elements](./tuple.md), each having an individual type. -- **Nullable**: [`Nullable`](./nullable.md) allows you to store a value as `NULL` when a value is "missing" (instead of the column settings its default value for the data type) -- **IP addresses**: use [`IPv4`](./ipv4.md) and [`IPv6`](./ipv6.md) to efficiently store IP addresses -- **Geo types**: for [geographical data](./geo.md), including `Point`, `Ring`, `Polygon` and `MultiPolygon` -- **Special data types**: including [`Expression`](./special-data-types/expression.md), [`Set`](./special-data-types/set.md), [`Nothing`](./special-data-types/nothing.md) and [`Interval`](./special-data-types/interval.md) +System table [system.data_type_families](../../operations/system-tables/data_type_families.md#system_tables-data_type_families) provides an +overview of all available data types. +It also shows whether a data type is an alias to another data type and its name is case-sensitive (e.g. `bool` vs. `BOOL`). diff --git a/docs/en/sql-reference/data-types/json.md b/docs/en/sql-reference/data-types/json.md index e48b308a620..ce69f15f0fa 100644 --- a/docs/en/sql-reference/data-types/json.md +++ b/docs/en/sql-reference/data-types/json.md @@ -7,7 +7,7 @@ keywords: [object, data type] # Object Data Type (deprecated) -**This feature is not production-ready and is now deprecated.** If you need to work with JSON documents, consider using [this guide](/docs/en/integrations/data-formats/json/overview) instead. A new implementation to support JSON object is in progress and can be tracked [here](https://github.com/ClickHouse/ClickHouse/issues/54864). +**This feature is not production-ready and deprecated.** If you need to work with JSON documents, consider using [this guide](/docs/en/integrations/data-formats/json/overview) instead. A new implementation to support JSON object is in progress and can be tracked [here](https://github.com/ClickHouse/ClickHouse/issues/54864).

diff --git a/docs/en/sql-reference/data-types/simpleaggregatefunction.md b/docs/en/sql-reference/data-types/simpleaggregatefunction.md index 4fb74ac30e4..8edd8b5b8ff 100644 --- a/docs/en/sql-reference/data-types/simpleaggregatefunction.md +++ b/docs/en/sql-reference/data-types/simpleaggregatefunction.md @@ -5,7 +5,9 @@ sidebar_label: SimpleAggregateFunction --- # SimpleAggregateFunction -`SimpleAggregateFunction(name, types_of_arguments...)` data type stores current value of the aggregate function, and does not store its full state as [`AggregateFunction`](../../sql-reference/data-types/aggregatefunction.md) does. This optimization can be applied to functions for which the following property holds: the result of applying a function `f` to a row set `S1 UNION ALL S2` can be obtained by applying `f` to parts of the row set separately, and then again applying `f` to the results: `f(S1 UNION ALL S2) = f(f(S1) UNION ALL f(S2))`. This property guarantees that partial aggregation results are enough to compute the combined one, so we do not have to store and process any extra data. +`SimpleAggregateFunction(name, types_of_arguments...)` data type stores current value (intermediate state) of the aggregate function, but not its full state as [`AggregateFunction`](../../sql-reference/data-types/aggregatefunction.md) does. +This optimization can be applied to functions for which the following property holds: the result of applying a function `f` to a row set `S1 UNION ALL S2` can be obtained by applying `f` to parts of the row set separately, and then again applying `f` to the results: `f(S1 UNION ALL S2) = f(f(S1) UNION ALL f(S2))`. +This property guarantees that partial aggregation results are enough to compute the combined one, so we do not have to store and process any extra data. The common way to produce an aggregate function value is by calling the aggregate function with the [-SimpleState](../../sql-reference/aggregate-functions/combinators.md#agg-functions-combinator-simplestate) suffix. diff --git a/docs/en/sql-reference/functions/geo/index.md b/docs/en/sql-reference/functions/geo/index.md index d46e60281e2..51b6868611a 100644 --- a/docs/en/sql-reference/functions/geo/index.md +++ b/docs/en/sql-reference/functions/geo/index.md @@ -5,70 +5,4 @@ sidebar_position: 62 title: "Geo Functions" --- - -## Geographical Coordinates Functions - -- [greatCircleDistance](./coordinates.md#greatcircledistance) -- [geoDistance](./coordinates.md#geodistance) -- [greatCircleAngle](./coordinates.md#greatcircleangle) -- [pointInEllipses](./coordinates.md#pointinellipses) -- [pointInPolygon](./coordinates.md#pointinpolygon) - -## Geohash Functions -- [geohashEncode](./geohash.md#geohashencode) -- [geohashDecode](./geohash.md#geohashdecode) -- [geohashesInBox](./geohash.md#geohashesinbox) - -## H3 Indexes Functions - -- [h3IsValid](./h3.md#h3isvalid) -- [h3GetResolution](./h3.md#h3getresolution) -- [h3EdgeAngle](./h3.md#h3edgeangle) -- [h3EdgeLengthM](./h3.md#h3edgelengthm) -- [h3EdgeLengthKm](./h3.md#h3edgelengthkm) -- [geoToH3](./h3.md#geotoh3) -- [h3ToGeo](./h3.md#h3togeo) -- [h3ToGeoBoundary](./h3.md#h3togeoboundary) -- [h3kRing](./h3.md#h3kring) -- [h3GetBaseCell](./h3.md#h3getbasecell) -- [h3HexAreaM2](./h3.md#h3hexaream2) -- [h3HexAreaKm2](./h3.md#h3hexareakm2) -- [h3IndexesAreNeighbors](./h3.md#h3indexesareneighbors) -- [h3ToChildren](./h3.md#h3tochildren) -- [h3ToParent](./h3.md#h3toparent) -- [h3ToString](./h3.md#h3tostring) -- [stringToH3](./h3.md#stringtoh3) -- [h3GetResolution](./h3.md#h3getresolution) -- [h3IsResClassIII](./h3.md#h3isresclassiii) -- [h3IsPentagon](./h3.md#h3ispentagon) -- [h3GetFaces](./h3.md#h3getfaces) -- [h3CellAreaM2](./h3.md#h3cellaream2) -- [h3CellAreaRads2](./h3.md#h3cellarearads2) -- [h3ToCenterChild](./h3.md#h3tocenterchild) -- [h3ExactEdgeLengthM](./h3.md#h3exactedgelengthm) -- [h3ExactEdgeLengthKm](./h3.md#h3exactedgelengthkm) -- [h3ExactEdgeLengthRads](./h3.md#h3exactedgelengthrads) -- [h3NumHexagons](./h3.md#h3numhexagons) -- [h3Line](./h3.md#h3line) -- [h3Distance](./h3.md#h3distance) -- [h3HexRing](./h3.md#h3hexring) -- [h3GetUnidirectionalEdge](./h3.md#h3getunidirectionaledge) -- [h3UnidirectionalEdgeIsValid](./h3.md#h3unidirectionaledgeisvalid) -- [h3GetOriginIndexFromUnidirectionalEdge](./h3.md#h3getoriginindexfromunidirectionaledge) -- [h3GetDestinationIndexFromUnidirectionalEdge](./h3.md#h3getdestinationindexfromunidirectionaledge) -- [h3GetIndexesFromUnidirectionalEdge](./h3.md#h3getindexesfromunidirectionaledge) -- [h3GetUnidirectionalEdgesFromHexagon](./h3.md#h3getunidirectionaledgesfromhexagon) -- [h3GetUnidirectionalEdgeBoundary](./h3.md#h3getunidirectionaledgeboundary) - -## S2 Index Functions - -- [geoToS2](./s2.md#geotos2) -- [s2ToGeo](./s2.md#s2togeo) -- [s2GetNeighbors](./s2.md#s2getneighbors) -- [s2CellsIntersect](./s2.md#s2cellsintersect) -- [s2CapContains](./s2.md#s2capcontains) -- [s2CapUnion](./s2.md#s2capunion) -- [s2RectAdd](./s2.md#s2rectadd) -- [s2RectContains](./s2.md#s2rectcontains) -- [s2RectUnion](./s2.md#s2rectunion) -- [s2RectIntersection](./s2.md#s2rectintersection) +Functions for working with geometric objects, for example [to calculate distances between points on a sphere](./coordinates.md), [compute geohashes](./geohash.md), and work with [h3 indexes](./h3.md). diff --git a/docs/en/sql-reference/statements/create/index.md b/docs/en/sql-reference/statements/create/index.md index fa39526a53e..5854d7cf9d2 100644 --- a/docs/en/sql-reference/statements/create/index.md +++ b/docs/en/sql-reference/statements/create/index.md @@ -6,16 +6,4 @@ sidebar_label: CREATE # CREATE Queries -Create queries make a new entity of one of the following kinds: - -- [DATABASE](/docs/en/sql-reference/statements/create/database.md) -- [TABLE](/docs/en/sql-reference/statements/create/table.md) -- [VIEW](/docs/en/sql-reference/statements/create/view.md) -- [DICTIONARY](/docs/en/sql-reference/statements/create/dictionary.md) -- [FUNCTION](/docs/en/sql-reference/statements/create/function.md) -- [USER](/docs/en/sql-reference/statements/create/user.md) -- [ROLE](/docs/en/sql-reference/statements/create/role.md) -- [ROW POLICY](/docs/en/sql-reference/statements/create/row-policy.md) -- [QUOTA](/docs/en/sql-reference/statements/create/quota.md) -- [SETTINGS PROFILE](/docs/en/sql-reference/statements/create/settings-profile.md) -- [NAMED COLLECTION](/docs/en/sql-reference/statements/create/named-collection.md) +CREATE queries create (for example) new [databases](/docs/en/sql-reference/statements/create/database.md), [tables](/docs/en/sql-reference/statements/create/table.md) and [views](/docs/en/sql-reference/statements/create/view.md). diff --git a/docs/en/sql-reference/statements/index.md b/docs/en/sql-reference/statements/index.md index 5aa61cf8d21..f288b30b27b 100644 --- a/docs/en/sql-reference/statements/index.md +++ b/docs/en/sql-reference/statements/index.md @@ -6,27 +6,4 @@ sidebar_label: List of statements # ClickHouse SQL Statements -Statements represent various kinds of action you can perform using SQL queries. Each kind of statement has it’s own syntax and usage details that are described separately: - -- [SELECT](/docs/en/sql-reference/statements/select/index.md) -- [INSERT INTO](/docs/en/sql-reference/statements/insert-into.md) -- [CREATE](/docs/en/sql-reference/statements/create/index.md) -- [ALTER](/docs/en/sql-reference/statements/alter/index.md) -- [SYSTEM](/docs/en/sql-reference/statements/system.md) -- [SHOW](/docs/en/sql-reference/statements/show.md) -- [GRANT](/docs/en/sql-reference/statements/grant.md) -- [REVOKE](/docs/en/sql-reference/statements/revoke.md) -- [ATTACH](/docs/en/sql-reference/statements/attach.md) -- [CHECK TABLE](/docs/en/sql-reference/statements/check-table.md) -- [DESCRIBE TABLE](/docs/en/sql-reference/statements/describe-table.md) -- [DETACH](/docs/en/sql-reference/statements/detach.md) -- [DROP](/docs/en/sql-reference/statements/drop.md) -- [EXISTS](/docs/en/sql-reference/statements/exists.md) -- [KILL](/docs/en/sql-reference/statements/kill.md) -- [OPTIMIZE](/docs/en/sql-reference/statements/optimize.md) -- [RENAME](/docs/en/sql-reference/statements/rename.md) -- [SET](/docs/en/sql-reference/statements/set.md) -- [SET ROLE](/docs/en/sql-reference/statements/set-role.md) -- [TRUNCATE](/docs/en/sql-reference/statements/truncate.md) -- [USE](/docs/en/sql-reference/statements/use.md) -- [EXPLAIN](/docs/en/sql-reference/statements/explain.md) +Users interact with ClickHouse using SQL statements. ClickHouse supports common SQL statements like [SELECT](select/index.md) and [CREATE](create/index.md), but it also provides specialized statements like [KILL](kill.md) and [OPTIMIZE](optimize.md). From 5aa9e64070cda74b65fa6cb639e2c83cd1abee67 Mon Sep 17 00:00:00 2001 From: Robert Schulze Date: Mon, 11 Nov 2024 10:11:23 +0000 Subject: [PATCH 620/680] Fix spelling --- utils/check-style/aspell-ignore/en/aspell-dict.txt | 1 + 1 file changed, 1 insertion(+) diff --git a/utils/check-style/aspell-ignore/en/aspell-dict.txt b/utils/check-style/aspell-ignore/en/aspell-dict.txt index a58b5e9ff58..a0d4d1d349e 100644 --- a/utils/check-style/aspell-ignore/en/aspell-dict.txt +++ b/utils/check-style/aspell-ignore/en/aspell-dict.txt @@ -1811,6 +1811,7 @@ geocode geohash geohashDecode geohashEncode +geohashes geohashesInBox geoip geospatial From 206bd174c37a7e6ea47eda9c228c2aa6a5f2fff3 Mon Sep 17 00:00:00 2001 From: Vitaly Baranov Date: Sat, 2 Nov 2024 19:30:03 +0100 Subject: [PATCH 621/680] Corrections after reworking backup/restore synchronization. --- src/Backups/BackupConcurrencyCheck.cpp | 16 +- src/Backups/BackupConcurrencyCheck.h | 11 +- src/Backups/BackupCoordinationCleaner.cpp | 36 +- src/Backups/BackupCoordinationCleaner.h | 17 +- src/Backups/BackupCoordinationLocal.cpp | 3 +- src/Backups/BackupCoordinationLocal.h | 13 +- src/Backups/BackupCoordinationOnCluster.cpp | 95 +- src/Backups/BackupCoordinationOnCluster.h | 20 +- src/Backups/BackupCoordinationStageSync.cpp | 895 ++++++++++++------- src/Backups/BackupCoordinationStageSync.h | 97 +- src/Backups/BackupsWorker.cpp | 108 ++- src/Backups/BackupsWorker.h | 2 - src/Backups/IBackupCoordination.h | 28 +- src/Backups/IRestoreCoordination.h | 28 +- src/Backups/RestoreCoordinationLocal.cpp | 4 +- src/Backups/RestoreCoordinationLocal.h | 14 +- src/Backups/RestoreCoordinationOnCluster.cpp | 95 +- src/Backups/RestoreCoordinationOnCluster.h | 20 +- 18 files changed, 887 insertions(+), 615 deletions(-) diff --git a/src/Backups/BackupConcurrencyCheck.cpp b/src/Backups/BackupConcurrencyCheck.cpp index 8b29ae41b53..a67d241845d 100644 --- a/src/Backups/BackupConcurrencyCheck.cpp +++ b/src/Backups/BackupConcurrencyCheck.cpp @@ -14,12 +14,12 @@ namespace ErrorCodes BackupConcurrencyCheck::BackupConcurrencyCheck( - const UUID & backup_or_restore_uuid_, bool is_restore_, bool on_cluster_, + const String & zookeeper_path_, bool allow_concurrency_, BackupConcurrencyCounters & counters_) - : is_restore(is_restore_), backup_or_restore_uuid(backup_or_restore_uuid_), on_cluster(on_cluster_), counters(counters_) + : is_restore(is_restore_), on_cluster(on_cluster_), zookeeper_path(zookeeper_path_), counters(counters_) { std::lock_guard lock{counters.mutex}; @@ -32,7 +32,7 @@ BackupConcurrencyCheck::BackupConcurrencyCheck( size_t num_on_cluster_restores = counters.on_cluster_restores.size(); if (on_cluster) { - if (!counters.on_cluster_restores.contains(backup_or_restore_uuid)) + if (!counters.on_cluster_restores.contains(zookeeper_path)) ++num_on_cluster_restores; } else @@ -47,7 +47,7 @@ BackupConcurrencyCheck::BackupConcurrencyCheck( size_t num_on_cluster_backups = counters.on_cluster_backups.size(); if (on_cluster) { - if (!counters.on_cluster_backups.contains(backup_or_restore_uuid)) + if (!counters.on_cluster_backups.contains(zookeeper_path)) ++num_on_cluster_backups; } else @@ -64,9 +64,9 @@ BackupConcurrencyCheck::BackupConcurrencyCheck( if (on_cluster) { if (is_restore) - ++counters.on_cluster_restores[backup_or_restore_uuid]; + ++counters.on_cluster_restores[zookeeper_path]; else - ++counters.on_cluster_backups[backup_or_restore_uuid]; + ++counters.on_cluster_backups[zookeeper_path]; } else { @@ -86,7 +86,7 @@ BackupConcurrencyCheck::~BackupConcurrencyCheck() { if (is_restore) { - auto it = counters.on_cluster_restores.find(backup_or_restore_uuid); + auto it = counters.on_cluster_restores.find(zookeeper_path); if (it != counters.on_cluster_restores.end()) { if (!--it->second) @@ -95,7 +95,7 @@ BackupConcurrencyCheck::~BackupConcurrencyCheck() } else { - auto it = counters.on_cluster_backups.find(backup_or_restore_uuid); + auto it = counters.on_cluster_backups.find(zookeeper_path); if (it != counters.on_cluster_backups.end()) { if (!--it->second) diff --git a/src/Backups/BackupConcurrencyCheck.h b/src/Backups/BackupConcurrencyCheck.h index 048a23a716a..a1baeff5464 100644 --- a/src/Backups/BackupConcurrencyCheck.h +++ b/src/Backups/BackupConcurrencyCheck.h @@ -1,7 +1,8 @@ #pragma once -#include +#include #include +#include #include #include @@ -19,9 +20,9 @@ public: /// Checks concurrency of a BACKUP operation or a RESTORE operation. /// Keep a constructed instance of BackupConcurrencyCheck until the operation is done. BackupConcurrencyCheck( - const UUID & backup_or_restore_uuid_, bool is_restore_, bool on_cluster_, + const String & zookeeper_path_, bool allow_concurrency_, BackupConcurrencyCounters & counters_); @@ -31,8 +32,8 @@ public: private: const bool is_restore; - const UUID backup_or_restore_uuid; const bool on_cluster; + const String zookeeper_path; BackupConcurrencyCounters & counters; }; @@ -47,8 +48,8 @@ private: friend class BackupConcurrencyCheck; size_t local_backups TSA_GUARDED_BY(mutex) = 0; size_t local_restores TSA_GUARDED_BY(mutex) = 0; - std::unordered_map on_cluster_backups TSA_GUARDED_BY(mutex); - std::unordered_map on_cluster_restores TSA_GUARDED_BY(mutex); + std::unordered_map on_cluster_backups TSA_GUARDED_BY(mutex); + std::unordered_map on_cluster_restores TSA_GUARDED_BY(mutex); std::mutex mutex; }; diff --git a/src/Backups/BackupCoordinationCleaner.cpp b/src/Backups/BackupCoordinationCleaner.cpp index 1f5068a94de..47095f27eb3 100644 --- a/src/Backups/BackupCoordinationCleaner.cpp +++ b/src/Backups/BackupCoordinationCleaner.cpp @@ -4,31 +4,29 @@ namespace DB { -BackupCoordinationCleaner::BackupCoordinationCleaner(const String & zookeeper_path_, const WithRetries & with_retries_, LoggerPtr log_) - : zookeeper_path(zookeeper_path_), with_retries(with_retries_), log(log_) +BackupCoordinationCleaner::BackupCoordinationCleaner(bool is_restore_, const String & zookeeper_path_, const WithRetries & with_retries_, LoggerPtr log_) + : is_restore(is_restore_), zookeeper_path(zookeeper_path_), with_retries(with_retries_), log(log_) { } -void BackupCoordinationCleaner::cleanup() +bool BackupCoordinationCleaner::cleanup(bool throw_if_error) { - tryRemoveAllNodes(/* throw_if_error = */ true, /* retries_kind = */ WithRetries::kNormal); + WithRetries::Kind retries_kind = throw_if_error ? WithRetries::kNormal : WithRetries::kErrorHandling; + return cleanupImpl(throw_if_error, retries_kind); } -bool BackupCoordinationCleaner::tryCleanupAfterError() noexcept -{ - return tryRemoveAllNodes(/* throw_if_error = */ false, /* retries_kind = */ WithRetries::kNormal); -} - -bool BackupCoordinationCleaner::tryRemoveAllNodes(bool throw_if_error, WithRetries::Kind retries_kind) +bool BackupCoordinationCleaner::cleanupImpl(bool throw_if_error, WithRetries::Kind retries_kind) { { std::lock_guard lock{mutex}; - if (cleanup_result.succeeded) - return true; - if (cleanup_result.exception) + if (succeeded) { - if (throw_if_error) - std::rethrow_exception(cleanup_result.exception); + LOG_TRACE(log, "Nodes from ZooKeeper are already removed"); + return true; + } + if (tried) + { + LOG_INFO(log, "Skipped removing nodes from ZooKeeper because because earlier we failed to do that"); return false; } } @@ -44,16 +42,18 @@ bool BackupCoordinationCleaner::tryRemoveAllNodes(bool throw_if_error, WithRetri }); std::lock_guard lock{mutex}; - cleanup_result.succeeded = true; + tried = true; + succeeded = true; return true; } catch (...) { - LOG_TRACE(log, "Caught exception while removing nodes from ZooKeeper for this restore: {}", + LOG_TRACE(log, "Caught exception while removing nodes from ZooKeeper for this {}: {}", + is_restore ? "restore" : "backup", getCurrentExceptionMessage(/* with_stacktrace= */ false, /* check_embedded_stacktrace= */ true)); std::lock_guard lock{mutex}; - cleanup_result.exception = std::current_exception(); + tried = true; if (throw_if_error) throw; diff --git a/src/Backups/BackupCoordinationCleaner.h b/src/Backups/BackupCoordinationCleaner.h index 43e095d9f33..c760a3611f9 100644 --- a/src/Backups/BackupCoordinationCleaner.h +++ b/src/Backups/BackupCoordinationCleaner.h @@ -12,14 +12,14 @@ namespace DB class BackupCoordinationCleaner { public: - BackupCoordinationCleaner(const String & zookeeper_path_, const WithRetries & with_retries_, LoggerPtr log_); + BackupCoordinationCleaner(bool is_restore_, const String & zookeeper_path_, const WithRetries & with_retries_, LoggerPtr log_); - void cleanup(); - bool tryCleanupAfterError() noexcept; + bool cleanup(bool throw_if_error); private: - bool tryRemoveAllNodes(bool throw_if_error, WithRetries::Kind retries_kind); + bool cleanupImpl(bool throw_if_error, WithRetries::Kind retries_kind); + const bool is_restore; const String zookeeper_path; /// A reference to a field of the parent object which is either BackupCoordinationOnCluster or RestoreCoordinationOnCluster. @@ -27,13 +27,8 @@ private: const LoggerPtr log; - struct CleanupResult - { - bool succeeded = false; - std::exception_ptr exception; - }; - CleanupResult cleanup_result TSA_GUARDED_BY(mutex); - + bool tried TSA_GUARDED_BY(mutex) = false; + bool succeeded TSA_GUARDED_BY(mutex) = false; std::mutex mutex; }; diff --git a/src/Backups/BackupCoordinationLocal.cpp b/src/Backups/BackupCoordinationLocal.cpp index 8bd6b4d327d..402e789eacb 100644 --- a/src/Backups/BackupCoordinationLocal.cpp +++ b/src/Backups/BackupCoordinationLocal.cpp @@ -11,12 +11,11 @@ namespace DB { BackupCoordinationLocal::BackupCoordinationLocal( - const UUID & backup_uuid_, bool is_plain_backup_, bool allow_concurrent_backup_, BackupConcurrencyCounters & concurrency_counters_) : log(getLogger("BackupCoordinationLocal")) - , concurrency_check(backup_uuid_, /* is_restore = */ false, /* on_cluster = */ false, allow_concurrent_backup_, concurrency_counters_) + , concurrency_check(/* is_restore = */ false, /* on_cluster = */ false, /* zookeeper_path = */ "", allow_concurrent_backup_, concurrency_counters_) , file_infos(is_plain_backup_) { } diff --git a/src/Backups/BackupCoordinationLocal.h b/src/Backups/BackupCoordinationLocal.h index 09991c0d301..e63fcde981a 100644 --- a/src/Backups/BackupCoordinationLocal.h +++ b/src/Backups/BackupCoordinationLocal.h @@ -23,20 +23,19 @@ class BackupCoordinationLocal : public IBackupCoordination { public: explicit BackupCoordinationLocal( - const UUID & backup_uuid_, bool is_plain_backup_, bool allow_concurrent_backup_, BackupConcurrencyCounters & concurrency_counters_); ~BackupCoordinationLocal() override; + void setBackupQueryIsSentToOtherHosts() override {} + bool isBackupQuerySentToOtherHosts() const override { return false; } Strings setStage(const String &, const String &, bool) override { return {}; } - void setBackupQueryWasSentToOtherHosts() override {} - bool trySetError(std::exception_ptr) override { return true; } - void finish() override {} - bool tryFinishAfterError() noexcept override { return true; } - void waitForOtherHostsToFinish() override {} - bool tryWaitForOtherHostsToFinishAfterError() noexcept override { return true; } + bool setError(std::exception_ptr, bool) override { return true; } + bool waitOtherHostsFinish(bool) const override { return true; } + bool finish(bool) override { return true; } + bool cleanup(bool) override { return true; } void addReplicatedPartNames(const String & table_zk_path, const String & table_name_for_logs, const String & replica_name, const std::vector & part_names_and_checksums) override; diff --git a/src/Backups/BackupCoordinationOnCluster.cpp b/src/Backups/BackupCoordinationOnCluster.cpp index dc34939f805..1b14f226eff 100644 --- a/src/Backups/BackupCoordinationOnCluster.cpp +++ b/src/Backups/BackupCoordinationOnCluster.cpp @@ -184,17 +184,21 @@ BackupCoordinationOnCluster::BackupCoordinationOnCluster( , plain_backup(is_plain_backup_) , log(getLogger("BackupCoordinationOnCluster")) , with_retries(log, get_zookeeper_, keeper_settings, process_list_element_, [root_zookeeper_path_](Coordination::ZooKeeperWithFaultInjection::Ptr zk) { zk->sync(root_zookeeper_path_); }) - , concurrency_check(backup_uuid_, /* is_restore = */ false, /* on_cluster = */ true, allow_concurrent_backup_, concurrency_counters_) - , stage_sync(/* is_restore = */ false, fs::path{zookeeper_path} / "stage", current_host, all_hosts, allow_concurrent_backup_, with_retries, schedule_, process_list_element_, log) - , cleaner(zookeeper_path, with_retries, log) + , cleaner(/* is_restore = */ false, zookeeper_path, with_retries, log) + , stage_sync(/* is_restore = */ false, fs::path{zookeeper_path} / "stage", current_host, all_hosts, allow_concurrent_backup_, concurrency_counters_, with_retries, schedule_, process_list_element_, log) { - createRootNodes(); + try + { + createRootNodes(); + } + catch (...) + { + stage_sync.setError(std::current_exception(), /* throw_if_error = */ false); + throw; + } } -BackupCoordinationOnCluster::~BackupCoordinationOnCluster() -{ - tryFinishImpl(); -} +BackupCoordinationOnCluster::~BackupCoordinationOnCluster() = default; void BackupCoordinationOnCluster::createRootNodes() { @@ -217,69 +221,52 @@ void BackupCoordinationOnCluster::createRootNodes() }); } +void BackupCoordinationOnCluster::setBackupQueryIsSentToOtherHosts() +{ + stage_sync.setQueryIsSentToOtherHosts(); +} + +bool BackupCoordinationOnCluster::isBackupQuerySentToOtherHosts() const +{ + return stage_sync.isQuerySentToOtherHosts(); +} + Strings BackupCoordinationOnCluster::setStage(const String & new_stage, const String & message, bool sync) { stage_sync.setStage(new_stage, message); - - if (!sync) - return {}; - - return stage_sync.waitForHostsToReachStage(new_stage, all_hosts_without_initiator); + if (sync) + return stage_sync.waitHostsReachStage(all_hosts_without_initiator, new_stage); + return {}; } -void BackupCoordinationOnCluster::setBackupQueryWasSentToOtherHosts() +bool BackupCoordinationOnCluster::setError(std::exception_ptr exception, bool throw_if_error) { - backup_query_was_sent_to_other_hosts = true; + return stage_sync.setError(exception, throw_if_error); } -bool BackupCoordinationOnCluster::trySetError(std::exception_ptr exception) +bool BackupCoordinationOnCluster::waitOtherHostsFinish(bool throw_if_error) const { - return stage_sync.trySetError(exception); + return stage_sync.waitOtherHostsFinish(throw_if_error); } -void BackupCoordinationOnCluster::finish() +bool BackupCoordinationOnCluster::finish(bool throw_if_error) { - bool other_hosts_also_finished = false; - stage_sync.finish(other_hosts_also_finished); - - if ((current_host == kInitiator) && (other_hosts_also_finished || !backup_query_was_sent_to_other_hosts)) - cleaner.cleanup(); + return stage_sync.finish(throw_if_error); } -bool BackupCoordinationOnCluster::tryFinishAfterError() noexcept +bool BackupCoordinationOnCluster::cleanup(bool throw_if_error) { - return tryFinishImpl(); -} - -bool BackupCoordinationOnCluster::tryFinishImpl() noexcept -{ - bool other_hosts_also_finished = false; - if (!stage_sync.tryFinishAfterError(other_hosts_also_finished)) - return false; - - if ((current_host == kInitiator) && (other_hosts_also_finished || !backup_query_was_sent_to_other_hosts)) + /// All the hosts must finish before we remove the coordination nodes. + bool expect_other_hosts_finished = stage_sync.isQuerySentToOtherHosts() || !stage_sync.isErrorSet(); + bool all_hosts_finished = stage_sync.finished() && (stage_sync.otherHostsFinished() || !expect_other_hosts_finished); + if (!all_hosts_finished) { - if (!cleaner.tryCleanupAfterError()) - return false; - } - - return true; -} - -void BackupCoordinationOnCluster::waitForOtherHostsToFinish() -{ - if ((current_host != kInitiator) || !backup_query_was_sent_to_other_hosts) - return; - stage_sync.waitForOtherHostsToFinish(); -} - -bool BackupCoordinationOnCluster::tryWaitForOtherHostsToFinishAfterError() noexcept -{ - if (current_host != kInitiator) + auto unfinished_hosts = expect_other_hosts_finished ? stage_sync.getUnfinishedHosts() : Strings{current_host}; + LOG_INFO(log, "Skipping removing nodes from ZooKeeper because hosts {} didn't finish", + BackupCoordinationStageSync::getHostsDesc(unfinished_hosts)); return false; - if (!backup_query_was_sent_to_other_hosts) - return true; - return stage_sync.tryWaitForOtherHostsToFinishAfterError(); + } + return cleaner.cleanup(throw_if_error); } ZooKeeperRetriesInfo BackupCoordinationOnCluster::getOnClusterInitializationKeeperRetriesInfo() const diff --git a/src/Backups/BackupCoordinationOnCluster.h b/src/Backups/BackupCoordinationOnCluster.h index 7369c2cc746..b439ab619d8 100644 --- a/src/Backups/BackupCoordinationOnCluster.h +++ b/src/Backups/BackupCoordinationOnCluster.h @@ -1,7 +1,6 @@ #pragma once #include -#include #include #include #include @@ -20,7 +19,7 @@ class BackupCoordinationOnCluster : public IBackupCoordination { public: /// Empty string as the current host is used to mark the initiator of a BACKUP ON CLUSTER query. - static const constexpr std::string_view kInitiator; + static const constexpr std::string_view kInitiator = BackupCoordinationStageSync::kInitiator; BackupCoordinationOnCluster( const UUID & backup_uuid_, @@ -37,13 +36,13 @@ public: ~BackupCoordinationOnCluster() override; + void setBackupQueryIsSentToOtherHosts() override; + bool isBackupQuerySentToOtherHosts() const override; Strings setStage(const String & new_stage, const String & message, bool sync) override; - void setBackupQueryWasSentToOtherHosts() override; - bool trySetError(std::exception_ptr exception) override; - void finish() override; - bool tryFinishAfterError() noexcept override; - void waitForOtherHostsToFinish() override; - bool tryWaitForOtherHostsToFinishAfterError() noexcept override; + bool setError(std::exception_ptr exception, bool throw_if_error) override; + bool waitOtherHostsFinish(bool throw_if_error) const override; + bool finish(bool throw_if_error) override; + bool cleanup(bool throw_if_error) override; void addReplicatedPartNames( const String & table_zk_path, @@ -110,11 +109,10 @@ private: const bool plain_backup; LoggerPtr const log; + /// The order is important: `stage_sync` must be initialized after `with_retries` and `cleaner`. const WithRetries with_retries; - BackupConcurrencyCheck concurrency_check; - BackupCoordinationStageSync stage_sync; BackupCoordinationCleaner cleaner; - std::atomic backup_query_was_sent_to_other_hosts = false; + BackupCoordinationStageSync stage_sync; mutable std::optional replicated_tables TSA_GUARDED_BY(replicated_tables_mutex); mutable std::optional replicated_access TSA_GUARDED_BY(replicated_access_mutex); diff --git a/src/Backups/BackupCoordinationStageSync.cpp b/src/Backups/BackupCoordinationStageSync.cpp index 9a05f9490c2..fcf09d7c315 100644 --- a/src/Backups/BackupCoordinationStageSync.cpp +++ b/src/Backups/BackupCoordinationStageSync.cpp @@ -42,9 +42,6 @@ namespace kCurrentVersion = 2, }; - - /// Empty string as the current host is used to mark the initiator of a BACKUP ON CLUSTER or RESTORE ON CLUSTER query. - const constexpr std::string_view kInitiator; } bool BackupCoordinationStageSync::HostInfo::operator ==(const HostInfo & other) const @@ -63,12 +60,32 @@ bool BackupCoordinationStageSync::State::operator ==(const State & other) const bool BackupCoordinationStageSync::State::operator !=(const State & other) const = default; +void BackupCoordinationStageSync::State::merge(const State & other) +{ + if (other.host_with_error && !host_with_error) + { + const String & host = *other.host_with_error; + host_with_error = host; + hosts.at(host).exception = other.hosts.at(host).exception; + } + + for (const auto & [host, other_host_info] : other.hosts) + { + auto & host_info = hosts.at(host); + host_info.stages.insert(other_host_info.stages.begin(), other_host_info.stages.end()); + if (other_host_info.finished) + host_info.finished = true; + } +} + + BackupCoordinationStageSync::BackupCoordinationStageSync( bool is_restore_, const String & zookeeper_path_, const String & current_host_, const Strings & all_hosts_, bool allow_concurrency_, + BackupConcurrencyCounters & concurrency_counters_, const WithRetries & with_retries_, ThreadPoolCallbackRunnerUnsafe schedule_, QueryStatusPtr process_list_element_, @@ -89,35 +106,29 @@ BackupCoordinationStageSync::BackupCoordinationStageSync( , max_attempts_after_bad_version(with_retries.getKeeperSettings().max_attempts_after_bad_version) , zookeeper_path(zookeeper_path_) , root_zookeeper_path(zookeeper_path.parent_path().parent_path()) - , operation_node_path(zookeeper_path.parent_path()) + , operation_zookeeper_path(zookeeper_path.parent_path()) , operation_node_name(zookeeper_path.parent_path().filename()) - , stage_node_path(zookeeper_path) , start_node_path(zookeeper_path / ("started|" + current_host)) , finish_node_path(zookeeper_path / ("finished|" + current_host)) , num_hosts_node_path(zookeeper_path / "num_hosts") + , error_node_path(zookeeper_path / "error") , alive_node_path(zookeeper_path / ("alive|" + current_host)) , alive_tracker_node_path(fs::path{root_zookeeper_path} / "alive_tracker") - , error_node_path(zookeeper_path / "error") , zk_nodes_changed(std::make_shared()) { - if ((zookeeper_path.filename() != "stage") || !operation_node_name.starts_with(is_restore ? "restore-" : "backup-") - || (root_zookeeper_path == operation_node_path)) - { - throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected path in ZooKeeper specified: {}", zookeeper_path); - } - initializeState(); createRootNodes(); try { + concurrency_check.emplace(is_restore, /* on_cluster = */ true, zookeeper_path, allow_concurrency, concurrency_counters_); createStartAndAliveNodes(); startWatchingThread(); } catch (...) { - trySetError(std::current_exception()); - tryFinishImpl(); + if (setError(std::current_exception(), /* throw_if_error = */ false)) + finish(/* throw_if_error = */ false); throw; } } @@ -125,7 +136,26 @@ BackupCoordinationStageSync::BackupCoordinationStageSync( BackupCoordinationStageSync::~BackupCoordinationStageSync() { - tryFinishImpl(); + /// Normally either finish() or setError() must be called. + if (!tried_to_finish) + { + if (state.host_with_error) + { + /// setError() was called and succeeded. + finish(/* throw_if_error = */ false); + } + else if (!tried_to_set_error) + { + /// Neither finish() nor setError() were called, it's a bug. + chassert(false, "~BackupCoordinationStageSync() is called without finish() or setError()"); + LOG_ERROR(log, "~BackupCoordinationStageSync() is called without finish() or setError()"); + } + } + + /// Normally the watching thread should be stopped already because the finish() function stops it. + /// However if an error happened then the watching thread can be still running, + /// so here in the destructor we have to ensure that it's stopped. + stopWatchingThread(); } @@ -137,6 +167,12 @@ void BackupCoordinationStageSync::initializeState() for (const String & host : all_hosts) state.hosts.emplace(host, HostInfo{.host = host, .last_connection_time = now, .last_connection_time_monotonic = monotonic_now}); + + if (!state.hosts.contains(current_host)) + throw Exception(ErrorCodes::LOGICAL_ERROR, "List of hosts must contain the current host"); + + if (!state.hosts.contains(String{kInitiator})) + throw Exception(ErrorCodes::LOGICAL_ERROR, "List of hosts must contain the initiator"); } @@ -179,6 +215,12 @@ String BackupCoordinationStageSync::getHostsDesc(const Strings & hosts) void BackupCoordinationStageSync::createRootNodes() { + if ((zookeeper_path.filename() != "stage") || !operation_node_name.starts_with(is_restore ? "restore-" : "backup-") + || (root_zookeeper_path == operation_zookeeper_path)) + { + throw Exception(ErrorCodes::LOGICAL_ERROR, "Unexpected path in ZooKeeper specified: {}", zookeeper_path); + } + auto holder = with_retries.createRetriesControlHolder("BackupStageSync::createRootNodes", WithRetries::kInitialization); holder.retries_ctl.retryLoop( [&, &zookeeper = holder.faulty_zookeeper]() @@ -252,27 +294,27 @@ void BackupCoordinationStageSync::createStartAndAliveNodes(Coordination::ZooKeep Coordination::Requests requests; requests.reserve(6); - size_t operation_node_path_pos = static_cast(-1); - if (!zookeeper->exists(operation_node_path)) + size_t operation_node_pos = static_cast(-1); + if (!zookeeper->exists(operation_zookeeper_path)) { - operation_node_path_pos = requests.size(); - requests.emplace_back(zkutil::makeCreateRequest(operation_node_path, "", zkutil::CreateMode::Persistent)); + operation_node_pos = requests.size(); + requests.emplace_back(zkutil::makeCreateRequest(operation_zookeeper_path, "", zkutil::CreateMode::Persistent)); } - size_t stage_node_path_pos = static_cast(-1); - if (!zookeeper->exists(stage_node_path)) + size_t zookeeper_path_pos = static_cast(-1); + if (!zookeeper->exists(zookeeper_path)) { - stage_node_path_pos = requests.size(); - requests.emplace_back(zkutil::makeCreateRequest(stage_node_path, "", zkutil::CreateMode::Persistent)); + zookeeper_path_pos = requests.size(); + requests.emplace_back(zkutil::makeCreateRequest(zookeeper_path, "", zkutil::CreateMode::Persistent)); } - size_t num_hosts_node_path_pos = requests.size(); + size_t num_hosts_node_pos = requests.size(); if (num_hosts) requests.emplace_back(zkutil::makeSetRequest(num_hosts_node_path, toString(*num_hosts + 1), num_hosts_version)); else requests.emplace_back(zkutil::makeCreateRequest(num_hosts_node_path, "1", zkutil::CreateMode::Persistent)); - size_t alive_tracker_node_path_pos = requests.size(); + size_t alive_tracker_node_pos = requests.size(); requests.emplace_back(zkutil::makeSetRequest(alive_tracker_node_path, "", alive_tracker_version)); requests.emplace_back(zkutil::makeCreateRequest(start_node_path, std::to_string(kCurrentVersion), zkutil::CreateMode::Persistent)); @@ -284,7 +326,7 @@ void BackupCoordinationStageSync::createStartAndAliveNodes(Coordination::ZooKeep if (code == Coordination::Error::ZOK) { LOG_INFO(log, "Created start node #{} in ZooKeeper for {} (coordination version: {})", - num_hosts.value_or(0) + 1, current_host_desc, kCurrentVersion); + num_hosts.value_or(0) + 1, current_host_desc, static_cast(kCurrentVersion)); return; } @@ -294,40 +336,34 @@ void BackupCoordinationStageSync::createStartAndAliveNodes(Coordination::ZooKeep LOG_TRACE(log, "{} (attempt #{}){}", message, attempt_no, will_try_again ? ", will try again" : ""); }; - if ((responses.size() > operation_node_path_pos) && - (responses[operation_node_path_pos]->error == Coordination::Error::ZNODEEXISTS)) + if ((operation_node_pos < responses.size()) && + (responses[operation_node_pos]->error == Coordination::Error::ZNODEEXISTS)) { - show_error_before_next_attempt(fmt::format("Node {} in ZooKeeper already exists", operation_node_path)); + show_error_before_next_attempt(fmt::format("Node {} already exists", operation_zookeeper_path)); /// needs another attempt } - else if ((responses.size() > stage_node_path_pos) && - (responses[stage_node_path_pos]->error == Coordination::Error::ZNODEEXISTS)) + else if ((zookeeper_path_pos < responses.size()) && + (responses[zookeeper_path_pos]->error == Coordination::Error::ZNODEEXISTS)) { - show_error_before_next_attempt(fmt::format("Node {} in ZooKeeper already exists", stage_node_path)); + show_error_before_next_attempt(fmt::format("Node {} already exists", zookeeper_path)); /// needs another attempt } - else if ((responses.size() > num_hosts_node_path_pos) && num_hosts && - (responses[num_hosts_node_path_pos]->error == Coordination::Error::ZBADVERSION)) + else if ((num_hosts_node_pos < responses.size()) && !num_hosts && + (responses[num_hosts_node_pos]->error == Coordination::Error::ZNODEEXISTS)) { - show_error_before_next_attempt("Other host changed the 'num_hosts' node in ZooKeeper"); + show_error_before_next_attempt(fmt::format("Node {} already exists", num_hosts_node_path)); + /// needs another attempt + } + else if ((num_hosts_node_pos < responses.size()) && num_hosts && + (responses[num_hosts_node_pos]->error == Coordination::Error::ZBADVERSION)) + { + show_error_before_next_attempt(fmt::format("The version of node {} changed", num_hosts_node_path)); num_hosts.reset(); /// needs to reread 'num_hosts' again } - else if ((responses.size() > num_hosts_node_path_pos) && num_hosts && - (responses[num_hosts_node_path_pos]->error == Coordination::Error::ZNONODE)) + else if ((alive_tracker_node_pos < responses.size()) && + (responses[alive_tracker_node_pos]->error == Coordination::Error::ZBADVERSION)) { - show_error_before_next_attempt("Other host removed the 'num_hosts' node in ZooKeeper"); - num_hosts.reset(); /// needs to reread 'num_hosts' again - } - else if ((responses.size() > num_hosts_node_path_pos) && !num_hosts && - (responses[num_hosts_node_path_pos]->error == Coordination::Error::ZNODEEXISTS)) - { - show_error_before_next_attempt("Other host created the 'num_hosts' node in ZooKeeper"); - /// needs another attempt - } - else if ((responses.size() > alive_tracker_node_path_pos) && - (responses[alive_tracker_node_path_pos]->error == Coordination::Error::ZBADVERSION)) - { - show_error_before_next_attempt("Concurrent backup or restore changed some 'alive' nodes in ZooKeeper"); + show_error_before_next_attempt(fmt::format("The version of node {} changed", alive_tracker_node_path)); check_concurrency = true; /// needs to recheck for concurrency again } else @@ -337,8 +373,7 @@ void BackupCoordinationStageSync::createStartAndAliveNodes(Coordination::ZooKeep } throw Exception(ErrorCodes::FAILED_TO_SYNC_BACKUP_OR_RESTORE, - "Couldn't create the 'start' node in ZooKeeper for {} after {} attempts", - current_host_desc, max_attempts_after_bad_version); + "Couldn't create node {} in ZooKeeper after {} attempts", start_node_path, max_attempts_after_bad_version); } @@ -387,36 +422,53 @@ void BackupCoordinationStageSync::startWatchingThread() void BackupCoordinationStageSync::stopWatchingThread() { - should_stop_watching_thread = true; + { + std::lock_guard lock{mutex}; + if (should_stop_watching_thread) + return; + should_stop_watching_thread = true; - /// Wake up waiting threads. - if (zk_nodes_changed) - zk_nodes_changed->set(); - state_changed.notify_all(); + /// Wake up waiting threads. + if (zk_nodes_changed) + zk_nodes_changed->set(); + state_changed.notify_all(); + } if (watching_thread_future.valid()) watching_thread_future.wait(); + + LOG_TRACE(log, "Stopped the watching thread"); } void BackupCoordinationStageSync::watchingThread() { - while (!should_stop_watching_thread) + auto should_stop = [&] + { + std::lock_guard lock{mutex}; + return should_stop_watching_thread; + }; + + while (!should_stop()) { try { /// Check if the current BACKUP or RESTORE command is already cancelled. checkIfQueryCancelled(); + } + catch (...) + { + tryLogCurrentException(log, "Caugth exception while watching"); + } - /// Reset the `connected` flag for each host, we'll set them to true again after we find the 'alive' nodes. - resetConnectedFlag(); - + try + { /// Recreate the 'alive' node if necessary and read a new state from ZooKeeper. auto holder = with_retries.createRetriesControlHolder("BackupStageSync::watchingThread"); auto & zookeeper = holder.faulty_zookeeper; with_retries.renewZooKeeper(zookeeper); - if (should_stop_watching_thread) + if (should_stop()) return; /// Recreate the 'alive' node if it was removed. @@ -427,7 +479,10 @@ void BackupCoordinationStageSync::watchingThread() } catch (...) { - tryLogCurrentException(log, "Caugth exception while watching"); + tryLogCurrentException(log, "Caught exception while watching"); + + /// Reset the `connected` flag for each host, we'll set them to true again after we find the 'alive' nodes. + resetConnectedFlag(); } try @@ -438,7 +493,7 @@ void BackupCoordinationStageSync::watchingThread() } catch (...) { - tryLogCurrentException(log, "Caugth exception while checking if the query should be cancelled"); + tryLogCurrentException(log, "Caught exception while watching"); } zk_nodes_changed->tryWait(sync_period_ms.count()); @@ -473,7 +528,7 @@ void BackupCoordinationStageSync::readCurrentState(Coordination::ZooKeeperWithFa zk_nodes_changed->reset(); /// Get zk nodes and subscribe on their changes. - Strings new_zk_nodes = zookeeper->getChildren(stage_node_path, nullptr, zk_nodes_changed); + Strings new_zk_nodes = zookeeper->getChildren(zookeeper_path, nullptr, zk_nodes_changed); std::sort(new_zk_nodes.begin(), new_zk_nodes.end()); /// Sorting is necessary because we compare the list of zk nodes with its previous versions. State new_state; @@ -492,6 +547,8 @@ void BackupCoordinationStageSync::readCurrentState(Coordination::ZooKeeperWithFa zk_nodes = new_zk_nodes; new_state = state; + for (auto & [_, host_info] : new_state.hosts) + host_info.connected = false; } auto get_host_info = [&](const String & host) -> HostInfo * @@ -514,7 +571,8 @@ void BackupCoordinationStageSync::readCurrentState(Coordination::ZooKeeperWithFa { String serialized_error = zookeeper->get(error_node_path); auto [exception, host] = parseErrorNode(serialized_error); - if (auto * host_info = get_host_info(host)) + auto * host_info = get_host_info(host); + if (exception && host_info) { host_info->exception = exception; new_state.host_with_error = host; @@ -576,6 +634,9 @@ void BackupCoordinationStageSync::readCurrentState(Coordination::ZooKeeperWithFa { std::lock_guard lock{mutex}; + /// We were reading `new_state` from ZooKeeper with `mutex` unlocked, so `state` could get more information during that reading, + /// we don't want to lose that information, that's why we use merge() here. + new_state.merge(state); was_state_changed = (new_state != state); state = std::move(new_state); } @@ -604,26 +665,10 @@ int BackupCoordinationStageSync::parseStartNode(const String & start_node_conten } -std::pair BackupCoordinationStageSync::parseErrorNode(const String & error_node_contents) -{ - ReadBufferFromOwnString buf{error_node_contents}; - String host; - readStringBinary(host, buf); - auto exception = std::make_exception_ptr(readException(buf, fmt::format("Got error from {}", getHostDesc(host)))); - return {exception, host}; -} - - void BackupCoordinationStageSync::checkIfQueryCancelled() { if (process_list_element->checkTimeLimitSoft()) return; /// Not cancelled. - - std::lock_guard lock{mutex}; - if (state.cancelled) - return; /// Already marked as cancelled. - - state.cancelled = true; state_changed.notify_all(); } @@ -634,13 +679,13 @@ void BackupCoordinationStageSync::cancelQueryIfError() { std::lock_guard lock{mutex}; - if (state.cancelled || !state.host_with_error) + if (!state.host_with_error) return; - state.cancelled = true; exception = state.hosts.at(*state.host_with_error).exception; } + chassert(exception); process_list_element->cancelQuery(false, exception); state_changed.notify_all(); } @@ -652,7 +697,7 @@ void BackupCoordinationStageSync::cancelQueryIfDisconnectedTooLong() { std::lock_guard lock{mutex}; - if (state.cancelled || state.host_with_error || ((failure_after_host_disconnected_for_seconds.count() == 0))) + if (state.host_with_error || ((failure_after_host_disconnected_for_seconds.count() == 0))) return; auto monotonic_now = std::chrono::steady_clock::now(); @@ -685,27 +730,92 @@ void BackupCoordinationStageSync::cancelQueryIfDisconnectedTooLong() } } } - - if (!exception) - return; - - state.cancelled = true; } + if (!exception) + return; + process_list_element->cancelQuery(false, exception); state_changed.notify_all(); } +void BackupCoordinationStageSync::setQueryIsSentToOtherHosts() +{ + std::lock_guard lock{mutex}; + query_is_sent_to_other_hosts = true; +} + +bool BackupCoordinationStageSync::isQuerySentToOtherHosts() const +{ + std::lock_guard lock{mutex}; + return query_is_sent_to_other_hosts; +} + + void BackupCoordinationStageSync::setStage(const String & stage, const String & stage_result) { LOG_INFO(log, "{} reached stage {}", current_host_desc, stage); + + { + std::lock_guard lock{mutex}; + if (state.hosts.at(current_host).stages.contains(stage)) + return; /// Already set. + } + + if ((getInitiatorVersion() == kVersionWithoutFinishNode) && (stage == BackupCoordinationStage::COMPLETED)) + { + LOG_TRACE(log, "Stopping the watching thread because the initiator uses outdated version {}", getInitiatorVersion()); + stopWatchingThread(); + } + auto holder = with_retries.createRetriesControlHolder("BackupStageSync::setStage"); holder.retries_ctl.retryLoop([&, &zookeeper = holder.faulty_zookeeper]() { with_retries.renewZooKeeper(zookeeper); - zookeeper->createIfNotExists(getStageNodePath(stage), stage_result); + createStageNode(stage, stage_result, zookeeper); }); + + /// If the initiator of the query has that old version then it doesn't expect us to create the 'finish' node and moreover + /// the initiator can start removing all the nodes immediately after all hosts report about reaching the "completed" status. + /// So to avoid weird errors in the logs we won't create the 'finish' node if the initiator of the query has that old version. + if ((getInitiatorVersion() == kVersionWithoutFinishNode) && (stage == BackupCoordinationStage::COMPLETED)) + { + LOG_INFO(log, "Skipped creating the 'finish' node because the initiator uses outdated version {}", getInitiatorVersion()); + std::lock_guard lock{mutex}; + tried_to_finish = true; + state.hosts.at(current_host).finished = true; + } +} + + +void BackupCoordinationStageSync::createStageNode(const String & stage, const String & stage_result, Coordination::ZooKeeperWithFaultInjection::Ptr zookeeper) +{ + String serialized_error; + if (zookeeper->tryGet(error_node_path, serialized_error)) + { + auto [exception, host] = parseErrorNode(serialized_error); + if (exception) + std::rethrow_exception(exception); + } + + auto code = zookeeper->tryCreate(getStageNodePath(stage), stage_result, zkutil::CreateMode::Persistent); + if (code == Coordination::Error::ZOK) + { + std::lock_guard lock{mutex}; + state.hosts.at(current_host).stages[stage] = stage_result; + return; + } + + if (code == Coordination::Error::ZNODEEXISTS) + { + String another_result = zookeeper->get(getStageNodePath(stage)); + std::lock_guard lock{mutex}; + state.hosts.at(current_host).stages[stage] = another_result; + return; + } + + throw zkutil::KeeperException::fromPath(code, getStageNodePath(stage)); } @@ -715,71 +825,7 @@ String BackupCoordinationStageSync::getStageNodePath(const String & stage) const } -bool BackupCoordinationStageSync::trySetError(std::exception_ptr exception) noexcept -{ - try - { - std::rethrow_exception(exception); - } - catch (const Exception & e) - { - return trySetError(e); - } - catch (...) - { - return trySetError(Exception(getCurrentExceptionMessageAndPattern(true, true), getCurrentExceptionCode())); - } -} - - -bool BackupCoordinationStageSync::trySetError(const Exception & exception) -{ - try - { - setError(exception); - return true; - } - catch (...) - { - return false; - } -} - - -void BackupCoordinationStageSync::setError(const Exception & exception) -{ - /// Most likely this exception has been already logged so here we're logging it without stacktrace. - String exception_message = getExceptionMessage(exception, /* with_stacktrace= */ false, /* check_embedded_stacktrace= */ true); - LOG_INFO(log, "Sending exception from {} to other hosts: {}", current_host_desc, exception_message); - - auto holder = with_retries.createRetriesControlHolder("BackupStageSync::setError", WithRetries::kErrorHandling); - - holder.retries_ctl.retryLoop([&, &zookeeper = holder.faulty_zookeeper]() - { - with_retries.renewZooKeeper(zookeeper); - - WriteBufferFromOwnString buf; - writeStringBinary(current_host, buf); - writeException(exception, buf, true); - auto code = zookeeper->tryCreate(error_node_path, buf.str(), zkutil::CreateMode::Persistent); - - if (code == Coordination::Error::ZOK) - { - LOG_TRACE(log, "Sent exception from {} to other hosts", current_host_desc); - } - else if (code == Coordination::Error::ZNODEEXISTS) - { - LOG_INFO(log, "An error has been already assigned for this {}", operation_name); - } - else - { - throw zkutil::KeeperException::fromPath(code, error_node_path); - } - }); -} - - -Strings BackupCoordinationStageSync::waitForHostsToReachStage(const String & stage_to_wait, const Strings & hosts, std::optional timeout) const +Strings BackupCoordinationStageSync::waitHostsReachStage(const Strings & hosts, const String & stage_to_wait) const { Strings results; results.resize(hosts.size()); @@ -787,44 +833,28 @@ Strings BackupCoordinationStageSync::waitForHostsToReachStage(const String & sta std::unique_lock lock{mutex}; /// TSA_NO_THREAD_SAFETY_ANALYSIS is here because Clang Thread Safety Analysis doesn't understand std::unique_lock. - auto check_if_hosts_ready = [&](bool time_is_out) TSA_NO_THREAD_SAFETY_ANALYSIS + auto check_if_hosts_reach_stage = [&]() TSA_NO_THREAD_SAFETY_ANALYSIS { - return checkIfHostsReachStage(hosts, stage_to_wait, time_is_out, timeout, results); + return checkIfHostsReachStage(hosts, stage_to_wait, results); }; - if (timeout) - { - if (!state_changed.wait_for(lock, *timeout, [&] { return check_if_hosts_ready(/* time_is_out = */ false); })) - check_if_hosts_ready(/* time_is_out = */ true); - } - else - { - state_changed.wait(lock, [&] { return check_if_hosts_ready(/* time_is_out = */ false); }); - } + state_changed.wait(lock, check_if_hosts_reach_stage); return results; } -bool BackupCoordinationStageSync::checkIfHostsReachStage( - const Strings & hosts, - const String & stage_to_wait, - bool time_is_out, - std::optional timeout, - Strings & results) const +bool BackupCoordinationStageSync::checkIfHostsReachStage(const Strings & hosts, const String & stage_to_wait, Strings & results) const { - if (should_stop_watching_thread) - throw Exception(ErrorCodes::LOGICAL_ERROR, "finish() was called while waiting for a stage"); - process_list_element->checkTimeLimit(); for (size_t i = 0; i != hosts.size(); ++i) { const String & host = hosts[i]; auto it = state.hosts.find(host); - if (it == state.hosts.end()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "waitForHostsToReachStage() was called for unexpected {}, all hosts are {}", getHostDesc(host), getHostsDesc(all_hosts)); + throw Exception(ErrorCodes::LOGICAL_ERROR, + "waitHostsReachStage() was called for unexpected {}, all hosts are {}", getHostDesc(host), getHostsDesc(all_hosts)); const HostInfo & host_info = it->second; auto stage_it = host_info.stages.find(stage_to_wait); @@ -835,10 +865,11 @@ bool BackupCoordinationStageSync::checkIfHostsReachStage( } if (host_info.finished) - { throw Exception(ErrorCodes::FAILED_TO_SYNC_BACKUP_OR_RESTORE, "{} finished without coming to stage {}", getHostDesc(host), stage_to_wait); - } + + if (should_stop_watching_thread) + throw Exception(ErrorCodes::LOGICAL_ERROR, "waitHostsReachStage() can't wait for stage {} after the watching thread stopped", stage_to_wait); String host_status; if (!host_info.started) @@ -846,85 +877,73 @@ bool BackupCoordinationStageSync::checkIfHostsReachStage( else if (!host_info.connected) host_status = fmt::format(": the host is currently disconnected, last connection was at {}", host_info.last_connection_time); - if (!time_is_out) - { - LOG_TRACE(log, "Waiting for {} to reach stage {}{}", getHostDesc(host), stage_to_wait, host_status); - return false; - } - else - { - throw Exception(ErrorCodes::FAILED_TO_SYNC_BACKUP_OR_RESTORE, - "Waited longer than timeout {} for {} to reach stage {}{}", - *timeout, getHostDesc(host), stage_to_wait, host_status); - } + LOG_TRACE(log, "Waiting for {} to reach stage {}{}", getHostDesc(host), stage_to_wait, host_status); + return false; /// wait for next change of `state_changed` } LOG_INFO(log, "Hosts {} reached stage {}", getHostsDesc(hosts), stage_to_wait); - return true; + return true; /// stop waiting } -void BackupCoordinationStageSync::finish(bool & other_hosts_also_finished) +bool BackupCoordinationStageSync::finish(bool throw_if_error) { - tryFinishImpl(other_hosts_also_finished, /* throw_if_error = */ true, /* retries_kind = */ WithRetries::kNormal); + WithRetries::Kind retries_kind = WithRetries::kNormal; + if (throw_if_error) + retries_kind = WithRetries::kErrorHandling; + + return finishImpl(throw_if_error, retries_kind); } -bool BackupCoordinationStageSync::tryFinishAfterError(bool & other_hosts_also_finished) noexcept +bool BackupCoordinationStageSync::finishImpl(bool throw_if_error, WithRetries::Kind retries_kind) { - return tryFinishImpl(other_hosts_also_finished, /* throw_if_error = */ false, /* retries_kind = */ WithRetries::kErrorHandling); -} - - -bool BackupCoordinationStageSync::tryFinishImpl() -{ - bool other_hosts_also_finished; - return tryFinishAfterError(other_hosts_also_finished); -} - - -bool BackupCoordinationStageSync::tryFinishImpl(bool & other_hosts_also_finished, bool throw_if_error, WithRetries::Kind retries_kind) -{ - auto get_value_other_hosts_also_finished = [&] TSA_REQUIRES(mutex) - { - other_hosts_also_finished = true; - for (const auto & [host, host_info] : state.hosts) - { - if ((host != current_host) && !host_info.finished) - other_hosts_also_finished = false; - } - }; - { std::lock_guard lock{mutex}; - if (finish_result.succeeded) + + if (finishedNoLock()) { - get_value_other_hosts_also_finished(); + LOG_INFO(log, "The finish node for {} already exists", current_host_desc); return true; } - if (finish_result.exception) + + if (tried_to_finish) { - if (throw_if_error) - std::rethrow_exception(finish_result.exception); + /// We don't repeat creating the finish node, no matter if it was successful or not. + LOG_INFO(log, "Skipped creating the finish node for {} because earlier we failed to do that", current_host_desc); return false; } + + bool failed_to_set_error = tried_to_set_error && !state.host_with_error; + if (failed_to_set_error) + { + /// Tried to create the 'error' node, but failed. + /// Then it's better not to create the 'finish' node in this case because otherwise other hosts might think we've succeeded. + LOG_INFO(log, "Skipping creating the finish node for {} because there was an error which we were unable to send to other hosts", current_host_desc); + return false; + } + + if (current_host == kInitiator) + { + /// Normally the initiator should wait for other hosts to finish before creating its own finish node. + /// We show warning if some of the other hosts didn't finish. + bool expect_other_hosts_finished = query_is_sent_to_other_hosts || !state.host_with_error; + bool other_hosts_finished = otherHostsFinishedNoLock() || !expect_other_hosts_finished; + if (!other_hosts_finished) + LOG_WARNING(log, "Hosts {} didn't finish before the initiator", getHostsDesc(getUnfinishedOtherHostsNoLock())); + } } + stopWatchingThread(); + try { - stopWatchingThread(); - auto holder = with_retries.createRetriesControlHolder("BackupStageSync::finish", retries_kind); holder.retries_ctl.retryLoop([&, &zookeeper = holder.faulty_zookeeper]() { with_retries.renewZooKeeper(zookeeper); - createFinishNodeAndRemoveAliveNode(zookeeper); + createFinishNodeAndRemoveAliveNode(zookeeper, throw_if_error); }); - - std::lock_guard lock{mutex}; - finish_result.succeeded = true; - get_value_other_hosts_also_finished(); - return true; } catch (...) { @@ -933,63 +952,87 @@ bool BackupCoordinationStageSync::tryFinishImpl(bool & other_hosts_also_finished getCurrentExceptionMessage(/* with_stacktrace= */ false, /* check_embedded_stacktrace= */ true)); std::lock_guard lock{mutex}; - finish_result.exception = std::current_exception(); + tried_to_finish = true; + if (throw_if_error) throw; return false; } + + { + std::lock_guard lock{mutex}; + tried_to_finish = true; + state.hosts.at(current_host).finished = true; + } + + return true; } -void BackupCoordinationStageSync::createFinishNodeAndRemoveAliveNode(Coordination::ZooKeeperWithFaultInjection::Ptr zookeeper) +void BackupCoordinationStageSync::createFinishNodeAndRemoveAliveNode(Coordination::ZooKeeperWithFaultInjection::Ptr zookeeper, bool throw_if_error) { - if (zookeeper->exists(finish_node_path)) - return; - - /// If the initiator of the query has that old version then it doesn't expect us to create the 'finish' node and moreover - /// the initiator can start removing all the nodes immediately after all hosts report about reaching the "completed" status. - /// So to avoid weird errors in the logs we won't create the 'finish' node if the initiator of the query has that old version. - if ((getInitiatorVersion() == kVersionWithoutFinishNode) && (current_host != kInitiator)) - { - LOG_INFO(log, "Skipped creating the 'finish' node because the initiator uses outdated version {}", getInitiatorVersion()); - return; - } - std::optional num_hosts; int num_hosts_version = -1; for (size_t attempt_no = 1; attempt_no <= max_attempts_after_bad_version; ++attempt_no) { + /// The 'num_hosts' node may not exist if createStartAndAliveNodes() failed in the constructor. if (!num_hosts) { + String num_hosts_str; Coordination::Stat stat; - num_hosts = parseFromString(zookeeper->get(num_hosts_node_path, &stat)); - num_hosts_version = stat.version; + if (zookeeper->tryGet(num_hosts_node_path, num_hosts_str, &stat)) + { + num_hosts = parseFromString(num_hosts_str); + num_hosts_version = stat.version; + } } + String serialized_error; + if (throw_if_error && zookeeper->tryGet(error_node_path, serialized_error)) + { + auto [exception, host] = parseErrorNode(serialized_error); + if (exception) + std::rethrow_exception(exception); + } + + if (zookeeper->exists(finish_node_path)) + return; + + bool start_node_exists = zookeeper->exists(start_node_path); + Coordination::Requests requests; requests.reserve(3); requests.emplace_back(zkutil::makeCreateRequest(finish_node_path, "", zkutil::CreateMode::Persistent)); - size_t num_hosts_node_path_pos = requests.size(); - requests.emplace_back(zkutil::makeSetRequest(num_hosts_node_path, toString(*num_hosts - 1), num_hosts_version)); - - size_t alive_node_path_pos = static_cast(-1); + size_t alive_node_pos = static_cast(-1); if (zookeeper->exists(alive_node_path)) { - alive_node_path_pos = requests.size(); + alive_node_pos = requests.size(); requests.emplace_back(zkutil::makeRemoveRequest(alive_node_path, -1)); } + size_t num_hosts_node_pos = static_cast(-1); + if (num_hosts) + { + num_hosts_node_pos = requests.size(); + requests.emplace_back(zkutil::makeSetRequest(num_hosts_node_path, toString(start_node_exists ? (*num_hosts - 1) : *num_hosts), num_hosts_version)); + } + Coordination::Responses responses; auto code = zookeeper->tryMulti(requests, responses); if (code == Coordination::Error::ZOK) { - --*num_hosts; - String hosts_left_desc = ((*num_hosts == 0) ? "no hosts left" : fmt::format("{} hosts left", *num_hosts)); - LOG_INFO(log, "Created the 'finish' node in ZooKeeper for {}, {}", current_host_desc, hosts_left_desc); + String hosts_left_desc; + if (num_hosts) + { + if (start_node_exists) + --*num_hosts; + hosts_left_desc = (*num_hosts == 0) ? ", no hosts left" : fmt::format(", {} hosts left", *num_hosts); + } + LOG_INFO(log, "Created the 'finish' node in ZooKeeper for {}{}", current_host_desc, hosts_left_desc); return; } @@ -999,18 +1042,18 @@ void BackupCoordinationStageSync::createFinishNodeAndRemoveAliveNode(Coordinatio LOG_TRACE(log, "{} (attempt #{}){}", message, attempt_no, will_try_again ? ", will try again" : ""); }; - if ((responses.size() > num_hosts_node_path_pos) && - (responses[num_hosts_node_path_pos]->error == Coordination::Error::ZBADVERSION)) + if ((alive_node_pos < responses.size()) && + (responses[alive_node_pos]->error == Coordination::Error::ZNONODE)) { - show_error_before_next_attempt("Other host changed the 'num_hosts' node in ZooKeeper"); - num_hosts.reset(); /// needs to reread 'num_hosts' again - } - else if ((responses.size() > alive_node_path_pos) && - (responses[alive_node_path_pos]->error == Coordination::Error::ZNONODE)) - { - show_error_before_next_attempt(fmt::format("Node {} in ZooKeeper doesn't exist", alive_node_path_pos)); + show_error_before_next_attempt(fmt::format("Node {} doesn't exist", alive_node_path)); /// needs another attempt } + else if ((num_hosts_node_pos < responses.size()) && + (responses[num_hosts_node_pos]->error == Coordination::Error::ZBADVERSION)) + { + show_error_before_next_attempt(fmt::format("The version of node {} changed", num_hosts_node_path)); + num_hosts.reset(); /// needs to reread 'num_hosts' again + } else { zkutil::KeeperMultiException::check(code, requests, responses); @@ -1026,60 +1069,73 @@ void BackupCoordinationStageSync::createFinishNodeAndRemoveAliveNode(Coordinatio int BackupCoordinationStageSync::getInitiatorVersion() const { std::lock_guard lock{mutex}; - auto it = state.hosts.find(String{kInitiator}); - if (it == state.hosts.end()) - throw Exception(ErrorCodes::LOGICAL_ERROR, "There is no initiator of this {} query, it's a bug", operation_name); - const HostInfo & host_info = it->second; - return host_info.version; + return state.hosts.at(String{kInitiator}).version; } -void BackupCoordinationStageSync::waitForOtherHostsToFinish() const -{ - tryWaitForOtherHostsToFinishImpl(/* reason = */ "", /* throw_if_error = */ true, /* timeout = */ {}); -} - - -bool BackupCoordinationStageSync::tryWaitForOtherHostsToFinishAfterError() const noexcept +bool BackupCoordinationStageSync::waitOtherHostsFinish(bool throw_if_error) const { std::optional timeout; - if (finish_timeout_after_error.count() != 0) - timeout = finish_timeout_after_error; + String reason; - String reason = fmt::format("{} needs other hosts to finish before cleanup", current_host_desc); - return tryWaitForOtherHostsToFinishImpl(reason, /* throw_if_error = */ false, timeout); + if (!throw_if_error) + { + if (finish_timeout_after_error.count() != 0) + timeout = finish_timeout_after_error; + reason = "after error before cleanup"; + } + + return waitOtherHostsFinishImpl(reason, timeout, throw_if_error); } -bool BackupCoordinationStageSync::tryWaitForOtherHostsToFinishImpl(const String & reason, bool throw_if_error, std::optional timeout) const +bool BackupCoordinationStageSync::waitOtherHostsFinishImpl(const String & reason, std::optional timeout, bool throw_if_error) const { std::unique_lock lock{mutex}; /// TSA_NO_THREAD_SAFETY_ANALYSIS is here because Clang Thread Safety Analysis doesn't understand std::unique_lock. - auto check_if_other_hosts_finish = [&](bool time_is_out) TSA_NO_THREAD_SAFETY_ANALYSIS + auto other_hosts_finished = [&]() TSA_NO_THREAD_SAFETY_ANALYSIS { return otherHostsFinishedNoLock(); }; + + if (other_hosts_finished()) { - return checkIfOtherHostsFinish(reason, throw_if_error, time_is_out, timeout); + LOG_TRACE(log, "Other hosts have already finished"); + return true; + } + + bool failed_to_set_error = TSA_SUPPRESS_WARNING_FOR_READ(tried_to_set_error) && !TSA_SUPPRESS_WARNING_FOR_READ(state).host_with_error; + if (failed_to_set_error) + { + /// Tried to create the 'error' node, but failed. + /// Then it's better not to wait for other hosts to finish in this case because other hosts don't know they should finish. + LOG_INFO(log, "Skipping waiting for other hosts to finish because there was an error which we were unable to send to other hosts"); + return false; + } + + bool result = false; + + /// TSA_NO_THREAD_SAFETY_ANALYSIS is here because Clang Thread Safety Analysis doesn't understand std::unique_lock. + auto check_if_hosts_finish = [&](bool time_is_out) TSA_NO_THREAD_SAFETY_ANALYSIS + { + return checkIfOtherHostsFinish(reason, timeout, time_is_out, result, throw_if_error); }; if (timeout) { - if (state_changed.wait_for(lock, *timeout, [&] { return check_if_other_hosts_finish(/* time_is_out = */ false); })) - return true; - return check_if_other_hosts_finish(/* time_is_out = */ true); + if (!state_changed.wait_for(lock, *timeout, [&] { return check_if_hosts_finish(/* time_is_out = */ false); })) + check_if_hosts_finish(/* time_is_out = */ true); } else { - state_changed.wait(lock, [&] { return check_if_other_hosts_finish(/* time_is_out = */ false); }); - return true; + state_changed.wait(lock, [&] { return check_if_hosts_finish(/* time_is_out = */ false); }); } + + return result; } -bool BackupCoordinationStageSync::checkIfOtherHostsFinish(const String & reason, bool throw_if_error, bool time_is_out, std::optional timeout) const +bool BackupCoordinationStageSync::checkIfOtherHostsFinish( + const String & reason, std::optional timeout, bool time_is_out, bool & result, bool throw_if_error) const { - if (should_stop_watching_thread) - throw Exception(ErrorCodes::LOGICAL_ERROR, "finish() was called while waiting for other hosts to finish"); - if (throw_if_error) process_list_element->checkTimeLimit(); @@ -1088,38 +1144,261 @@ bool BackupCoordinationStageSync::checkIfOtherHostsFinish(const String & reason, if ((host == current_host) || host_info.finished) continue; + String reason_text = reason.empty() ? "" : (" " + reason); + String host_status; if (!host_info.started) host_status = fmt::format(": the host hasn't started working on this {} yet", operation_name); else if (!host_info.connected) host_status = fmt::format(": the host is currently disconnected, last connection was at {}", host_info.last_connection_time); - if (!time_is_out) + if (time_is_out) { - String reason_text = reason.empty() ? "" : (" because " + reason); - LOG_TRACE(log, "Waiting for {} to finish{}{}", getHostDesc(host), reason_text, host_status); - return false; - } - else - { - String reason_text = reason.empty() ? "" : fmt::format(" (reason of waiting: {})", reason); - if (!throw_if_error) - { - LOG_INFO(log, "Waited longer than timeout {} for {} to finish{}{}", - *timeout, getHostDesc(host), host_status, reason_text); - return false; - } - else + if (throw_if_error) { throw Exception(ErrorCodes::FAILED_TO_SYNC_BACKUP_OR_RESTORE, "Waited longer than timeout {} for {} to finish{}{}", - *timeout, getHostDesc(host), host_status, reason_text); + *timeout, getHostDesc(host), reason_text, host_status); } + LOG_INFO(log, "Waited longer than timeout {} for {} to finish{}{}", + *timeout, getHostDesc(host), reason_text, host_status); + result = false; + return true; /// stop waiting } + + if (should_stop_watching_thread) + { + LOG_ERROR(log, "waitOtherHostFinish({}) can't wait for other hosts to finish after the watching thread stopped", throw_if_error); + chassert(false, "waitOtherHostFinish() can't wait for other hosts to finish after the watching thread stopped"); + if (throw_if_error) + throw Exception(ErrorCodes::LOGICAL_ERROR, "waitOtherHostsFinish() can't wait for other hosts to finish after the watching thread stopped"); + result = false; + return true; /// stop waiting + } + + LOG_TRACE(log, "Waiting for {} to finish{}{}", getHostDesc(host), reason_text, host_status); + return false; /// wait for next change of `state_changed` } LOG_TRACE(log, "Other hosts finished working on this {}", operation_name); + result = true; + return true; /// stop waiting +} + + +bool BackupCoordinationStageSync::finished() const +{ + std::lock_guard lock{mutex}; + return finishedNoLock(); +} + + +bool BackupCoordinationStageSync::finishedNoLock() const +{ + return state.hosts.at(current_host).finished; +} + + +bool BackupCoordinationStageSync::otherHostsFinished() const +{ + std::lock_guard lock{mutex}; + return otherHostsFinishedNoLock(); +} + + +bool BackupCoordinationStageSync::otherHostsFinishedNoLock() const +{ + for (const auto & [host, host_info] : state.hosts) + { + if (!host_info.finished && (host != current_host)) + return false; + } return true; } + +bool BackupCoordinationStageSync::allHostsFinishedNoLock() const +{ + return finishedNoLock() && otherHostsFinishedNoLock(); +} + + +Strings BackupCoordinationStageSync::getUnfinishedHosts() const +{ + std::lock_guard lock{mutex}; + return getUnfinishedHostsNoLock(); +} + + +Strings BackupCoordinationStageSync::getUnfinishedHostsNoLock() const +{ + if (allHostsFinishedNoLock()) + return {}; + + Strings res; + res.reserve(all_hosts.size()); + for (const auto & [host, host_info] : state.hosts) + { + if (!host_info.finished) + res.emplace_back(host); + } + return res; +} + + +Strings BackupCoordinationStageSync::getUnfinishedOtherHosts() const +{ + std::lock_guard lock{mutex}; + return getUnfinishedOtherHostsNoLock(); +} + + +Strings BackupCoordinationStageSync::getUnfinishedOtherHostsNoLock() const +{ + if (otherHostsFinishedNoLock()) + return {}; + + Strings res; + res.reserve(all_hosts.size() - 1); + for (const auto & [host, host_info] : state.hosts) + { + if (!host_info.finished && (host != current_host)) + res.emplace_back(host); + } + return res; +} + + +bool BackupCoordinationStageSync::setError(std::exception_ptr exception, bool throw_if_error) +{ + try + { + std::rethrow_exception(exception); + } + catch (const Exception & e) + { + return setError(e, throw_if_error); + } + catch (...) + { + return setError(Exception{getCurrentExceptionMessageAndPattern(true, true), getCurrentExceptionCode()}, throw_if_error); + } +} + + +bool BackupCoordinationStageSync::setError(const Exception & exception, bool throw_if_error) +{ + try + { + /// Most likely this exception has been already logged so here we're logging it without stacktrace. + String exception_message = getExceptionMessage(exception, /* with_stacktrace= */ false, /* check_embedded_stacktrace= */ true); + LOG_INFO(log, "Sending exception from {} to other hosts: {}", current_host_desc, exception_message); + + { + std::lock_guard lock{mutex}; + if (state.host_with_error) + { + LOG_INFO(log, "The error node already exists"); + return true; + } + + if (tried_to_set_error) + { + LOG_INFO(log, "Skipped creating the error node because earlier we failed to do that"); + return false; + } + } + + auto holder = with_retries.createRetriesControlHolder("BackupStageSync::setError", WithRetries::kErrorHandling); + holder.retries_ctl.retryLoop([&, &zookeeper = holder.faulty_zookeeper]() + { + with_retries.renewZooKeeper(zookeeper); + createErrorNode(exception, zookeeper); + }); + + { + std::lock_guard lock{mutex}; + tried_to_set_error = true; + return true; + } + } + catch (...) + { + LOG_TRACE(log, "Caught exception while removing nodes from ZooKeeper for this {}: {}", + is_restore ? "restore" : "backup", + getCurrentExceptionMessage(/* with_stacktrace= */ false, /* check_embedded_stacktrace= */ true)); + + std::lock_guard lock{mutex}; + tried_to_set_error = true; + + if (throw_if_error) + throw; + return false; + } +} + + +void BackupCoordinationStageSync::createErrorNode(const Exception & exception, Coordination::ZooKeeperWithFaultInjection::Ptr zookeeper) +{ + String serialized_error; + { + WriteBufferFromOwnString buf; + writeStringBinary(current_host, buf); + writeException(exception, buf, true); + serialized_error = buf.str(); + } + + auto code = zookeeper->tryCreate(error_node_path, serialized_error, zkutil::CreateMode::Persistent); + + if (code == Coordination::Error::ZOK) + { + std::lock_guard lock{mutex}; + if (!state.host_with_error) + { + state.host_with_error = current_host; + state.hosts.at(current_host).exception = parseErrorNode(serialized_error).first; + } + LOG_TRACE(log, "Sent exception from {} to other hosts", current_host_desc); + return; + } + + if (code == Coordination::Error::ZNODEEXISTS) + { + String another_error = zookeeper->get(error_node_path); + auto [another_exception, host] = parseErrorNode(another_error); + if (another_exception) + { + std::lock_guard lock{mutex}; + if (!state.host_with_error) + { + state.host_with_error = host; + state.hosts.at(host).exception = another_exception; + } + LOG_INFO(log, "Another error is already assigned for this {}", operation_name); + return; + } + } + + throw zkutil::KeeperException::fromPath(code, error_node_path); +} + + +std::pair BackupCoordinationStageSync::parseErrorNode(const String & error_node_contents) const +{ + ReadBufferFromOwnString buf{error_node_contents}; + String host; + readStringBinary(host, buf); + if (std::find(all_hosts.begin(), all_hosts.end(), host) == all_hosts.end()) + return {}; + auto exception = std::make_exception_ptr(readException(buf, fmt::format("Got error from {}", getHostDesc(host)))); + return {exception, host}; +} + + +bool BackupCoordinationStageSync::isErrorSet() const +{ + std::lock_guard lock{mutex}; + return state.host_with_error.has_value(); +} + } diff --git a/src/Backups/BackupCoordinationStageSync.h b/src/Backups/BackupCoordinationStageSync.h index dc0d3c3c83d..11d3d1cf6f4 100644 --- a/src/Backups/BackupCoordinationStageSync.h +++ b/src/Backups/BackupCoordinationStageSync.h @@ -1,7 +1,9 @@ #pragma once +#include #include + namespace DB { @@ -9,12 +11,16 @@ namespace DB class BackupCoordinationStageSync { public: + /// Empty string as the current host is used to mark the initiator of a BACKUP ON CLUSTER or RESTORE ON CLUSTER query. + static const constexpr std::string_view kInitiator; + BackupCoordinationStageSync( bool is_restore_, /// true if this is a RESTORE ON CLUSTER command, false if this is a BACKUP ON CLUSTER command const String & zookeeper_path_, /// path to the "stage" folder in ZooKeeper const String & current_host_, /// the current host, or an empty string if it's the initiator of the BACKUP/RESTORE ON CLUSTER command const Strings & all_hosts_, /// all the hosts (including the initiator and the current host) performing the BACKUP/RESTORE ON CLUSTER command bool allow_concurrency_, /// whether it's allowed to have concurrent backups or restores. + BackupConcurrencyCounters & concurrency_counters_, const WithRetries & with_retries_, ThreadPoolCallbackRunnerUnsafe schedule_, QueryStatusPtr process_list_element_, @@ -22,30 +28,37 @@ public: ~BackupCoordinationStageSync(); + /// Sets that the BACKUP or RESTORE query was sent to other hosts. + void setQueryIsSentToOtherHosts(); + bool isQuerySentToOtherHosts() const; + /// Sets the stage of the current host and signal other hosts if there were other hosts waiting for that. void setStage(const String & stage, const String & stage_result = {}); - /// Waits until all the specified hosts come to the specified stage. - /// The function returns the results which specified hosts set when they came to the required stage. - /// If it doesn't happen before the timeout then the function will stop waiting and throw an exception. - Strings waitForHostsToReachStage(const String & stage_to_wait, const Strings & hosts, std::optional timeout = {}) const; - - /// Waits until all the other hosts finish their work. - /// Stops waiting and throws an exception if another host encounters an error or if some host gets cancelled. - void waitForOtherHostsToFinish() const; - - /// Lets other host know that the current host has finished its work. - void finish(bool & other_hosts_also_finished); + /// Waits until specified hosts come to the specified stage. + /// The function returns the results which the specified hosts set when they came to the required stage. + Strings waitHostsReachStage(const Strings & hosts, const String & stage_to_wait) const; /// Lets other hosts know that the current host has encountered an error. - bool trySetError(std::exception_ptr exception) noexcept; + /// The function returns true if it successfully created the error node or if the error node was found already exist. + bool setError(std::exception_ptr exception, bool throw_if_error); + bool isErrorSet() const; - /// Waits until all the other hosts finish their work (as a part of error-handling process). - /// Doesn't stops waiting if some host encounters an error or gets cancelled. - bool tryWaitForOtherHostsToFinishAfterError() const noexcept; + /// Waits until the hosts other than the current host finish their work. Must be called before finish(). + /// Stops waiting and throws an exception if another host encounters an error or if some host gets cancelled. + bool waitOtherHostsFinish(bool throw_if_error) const; + bool otherHostsFinished() const; - /// Lets other host know that the current host has finished its work (as a part of error-handling process). - bool tryFinishAfterError(bool & other_hosts_also_finished) noexcept; + /// Lets other hosts know that the current host has finished its work. + bool finish(bool throw_if_error); + bool finished() const; + + /// Returns true if all the hosts have finished. + bool allHostsFinished() const { return finished() && otherHostsFinished(); } + + /// Returns a list of the hosts which haven't finished yet. + Strings getUnfinishedHosts() const; + Strings getUnfinishedOtherHosts() const; /// Returns a printable name of a specific host. For empty host the function returns "initiator". static String getHostDesc(const String & host); @@ -78,14 +91,17 @@ private: /// Reads the current state from ZooKeeper without throwing exceptions. void readCurrentState(Coordination::ZooKeeperWithFaultInjection::Ptr zookeeper); + + /// Creates a stage node to let other hosts know we've reached the specified stage. + void createStageNode(const String & stage, const String & stage_result, Coordination::ZooKeeperWithFaultInjection::Ptr zookeeper); String getStageNodePath(const String & stage) const; /// Lets other hosts know that the current host has encountered an error. - bool trySetError(const Exception & exception); - void setError(const Exception & exception); + bool setError(const Exception & exception, bool throw_if_error); + void createErrorNode(const Exception & exception, Coordination::ZooKeeperWithFaultInjection::Ptr zookeeper); /// Deserializes an error stored in the error node. - static std::pair parseErrorNode(const String & error_node_contents); + std::pair parseErrorNode(const String & error_node_contents) const; /// Reset the `connected` flag for each host. void resetConnectedFlag(); @@ -102,19 +118,27 @@ private: void cancelQueryIfDisconnectedTooLong(); /// Used by waitForHostsToReachStage() to check if everything is ready to return. - bool checkIfHostsReachStage(const Strings & hosts, const String & stage_to_wait, bool time_is_out, std::optional timeout, Strings & results) const TSA_REQUIRES(mutex); + bool checkIfHostsReachStage(const Strings & hosts, const String & stage_to_wait, Strings & results) const TSA_REQUIRES(mutex); /// Creates the 'finish' node. - bool tryFinishImpl(); - bool tryFinishImpl(bool & other_hosts_also_finished, bool throw_if_error, WithRetries::Kind retries_kind); - void createFinishNodeAndRemoveAliveNode(Coordination::ZooKeeperWithFaultInjection::Ptr zookeeper); + bool finishImpl(bool throw_if_error, WithRetries::Kind retries_kind); + void createFinishNodeAndRemoveAliveNode(Coordination::ZooKeeperWithFaultInjection::Ptr zookeeper, bool throw_if_error); /// Returns the version used by the initiator. int getInitiatorVersion() const; /// Waits until all the other hosts finish their work. - bool tryWaitForOtherHostsToFinishImpl(const String & reason, bool throw_if_error, std::optional timeout) const; - bool checkIfOtherHostsFinish(const String & reason, bool throw_if_error, bool time_is_out, std::optional timeout) const TSA_REQUIRES(mutex); + bool waitOtherHostsFinishImpl(const String & reason, std::optional timeout, bool throw_if_error) const; + bool checkIfOtherHostsFinish(const String & reason, std::optional timeout, bool time_is_out, bool & result, bool throw_if_error) const TSA_REQUIRES(mutex); + + /// Returns true if all the hosts have finished. + bool allHostsFinishedNoLock() const TSA_REQUIRES(mutex); + bool finishedNoLock() const TSA_REQUIRES(mutex); + bool otherHostsFinishedNoLock() const TSA_REQUIRES(mutex); + + /// Returns a list of the hosts which haven't finished yet. + Strings getUnfinishedHostsNoLock() const TSA_REQUIRES(mutex); + Strings getUnfinishedOtherHostsNoLock() const TSA_REQUIRES(mutex); const bool is_restore; const String operation_name; @@ -138,15 +162,16 @@ private: /// Paths in ZooKeeper. const std::filesystem::path zookeeper_path; const String root_zookeeper_path; - const String operation_node_path; + const String operation_zookeeper_path; const String operation_node_name; - const String stage_node_path; const String start_node_path; const String finish_node_path; const String num_hosts_node_path; + const String error_node_path; const String alive_node_path; const String alive_tracker_node_path; - const String error_node_path; + + std::optional concurrency_check; std::shared_ptr zk_nodes_changed; @@ -176,25 +201,21 @@ private: { std::map hosts; /// std::map because we need to compare states std::optional host_with_error; - bool cancelled = false; bool operator ==(const State & other) const; bool operator !=(const State & other) const; + void merge(const State & other); }; State state TSA_GUARDED_BY(mutex); mutable std::condition_variable state_changed; std::future watching_thread_future; - std::atomic should_stop_watching_thread = false; + bool should_stop_watching_thread TSA_GUARDED_BY(mutex) = false; - struct FinishResult - { - bool succeeded = false; - std::exception_ptr exception; - bool other_hosts_also_finished = false; - }; - FinishResult finish_result TSA_GUARDED_BY(mutex); + bool query_is_sent_to_other_hosts TSA_GUARDED_BY(mutex) = false; + bool tried_to_finish TSA_GUARDED_BY(mutex) = false; + bool tried_to_set_error TSA_GUARDED_BY(mutex) = false; mutable std::mutex mutex; }; diff --git a/src/Backups/BackupsWorker.cpp b/src/Backups/BackupsWorker.cpp index 8480dc5d64d..88ebf8eef32 100644 --- a/src/Backups/BackupsWorker.cpp +++ b/src/Backups/BackupsWorker.cpp @@ -329,6 +329,7 @@ std::pair BackupsWorker::start(const ASTPtr & backup_ struct BackupsWorker::BackupStarter { BackupsWorker & backups_worker; + LoggerPtr log; std::shared_ptr backup_query; ContextPtr query_context; /// We have to keep `query_context` until the end of the operation because a pointer to it is stored inside the ThreadGroup we're using. ContextMutablePtr backup_context; @@ -345,6 +346,7 @@ struct BackupsWorker::BackupStarter BackupStarter(BackupsWorker & backups_worker_, const ASTPtr & query_, const ContextPtr & context_) : backups_worker(backups_worker_) + , log(backups_worker.log) , backup_query(std::static_pointer_cast(query_->clone())) , query_context(context_) , backup_context(Context::createCopy(query_context)) @@ -399,9 +401,20 @@ struct BackupsWorker::BackupStarter chassert(!backup); backup = backups_worker.openBackupForWriting(backup_info, backup_settings, backup_coordination, backup_context); - backups_worker.doBackup( - backup, backup_query, backup_id, backup_name_for_logging, backup_settings, backup_coordination, backup_context, - on_cluster, cluster); + backups_worker.doBackup(backup, backup_query, backup_id, backup_settings, backup_coordination, backup_context, + on_cluster, cluster); + + backup_coordination->finish(/* throw_if_error = */ true); + backup.reset(); + + /// The backup coordination is not needed anymore. + if (!is_internal_backup) + backup_coordination->cleanup(/* throw_if_error = */ true); + backup_coordination.reset(); + + /// NOTE: setStatus is called after setNumFilesAndSize in order to have actual information in a backup log record + LOG_INFO(log, "{} {} was created successfully", (is_internal_backup ? "Internal backup" : "Backup"), backup_name_for_logging); + backups_worker.setStatus(backup_id, BackupStatus::BACKUP_CREATED); } void onException() @@ -416,16 +429,29 @@ struct BackupsWorker::BackupStarter if (backup && !backup->setIsCorrupted()) should_remove_files_in_backup = false; - if (backup_coordination && backup_coordination->trySetError(std::current_exception())) + bool all_hosts_finished = false; + + if (backup_coordination && backup_coordination->setError(std::current_exception(), /* throw_if_error = */ false)) { - bool other_hosts_finished = backup_coordination->tryWaitForOtherHostsToFinishAfterError(); + bool other_hosts_finished = !is_internal_backup + && (!backup_coordination->isBackupQuerySentToOtherHosts() || backup_coordination->waitOtherHostsFinish(/* throw_if_error = */ false)); - if (should_remove_files_in_backup && other_hosts_finished) - backup->tryRemoveAllFiles(); - - backup_coordination->tryFinishAfterError(); + all_hosts_finished = backup_coordination->finish(/* throw_if_error = */ false) && other_hosts_finished; } + if (!all_hosts_finished) + should_remove_files_in_backup = false; + + if (backup && should_remove_files_in_backup) + backup->tryRemoveAllFiles(); + + backup.reset(); + + if (backup_coordination && all_hosts_finished) + backup_coordination->cleanup(/* throw_if_error = */ false); + + backup_coordination.reset(); + backups_worker.setStatusSafe(backup_id, getBackupStatusFromCurrentException()); } }; @@ -497,7 +523,6 @@ void BackupsWorker::doBackup( BackupMutablePtr backup, const std::shared_ptr & backup_query, const OperationID & backup_id, - const String & backup_name_for_logging, const BackupSettings & backup_settings, std::shared_ptr backup_coordination, ContextMutablePtr context, @@ -521,10 +546,10 @@ void BackupsWorker::doBackup( backup_settings.copySettingsToQuery(*backup_query); sendQueryToOtherHosts(*backup_query, cluster, backup_settings.shard_num, backup_settings.replica_num, context, required_access, backup_coordination->getOnClusterInitializationKeeperRetriesInfo()); - backup_coordination->setBackupQueryWasSentToOtherHosts(); + backup_coordination->setBackupQueryIsSentToOtherHosts(); /// Wait until all the hosts have written their backup entries. - backup_coordination->waitForOtherHostsToFinish(); + backup_coordination->waitOtherHostsFinish(/* throw_if_error = */ true); } else { @@ -569,18 +594,8 @@ void BackupsWorker::doBackup( compressed_size = backup->getCompressedSize(); } - /// Close the backup. - backup.reset(); - - /// The backup coordination is not needed anymore. - backup_coordination->finish(); - /// NOTE: we need to update metadata again after backup->finalizeWriting(), because backup metadata is written there. setNumFilesAndSize(backup_id, num_files, total_size, num_entries, uncompressed_size, compressed_size, 0, 0); - - /// NOTE: setStatus is called after setNumFilesAndSize in order to have actual information in a backup log record - LOG_INFO(log, "{} {} was created successfully", (is_internal_backup ? "Internal backup" : "Backup"), backup_name_for_logging); - setStatus(backup_id, BackupStatus::BACKUP_CREATED); } @@ -687,6 +702,7 @@ void BackupsWorker::writeBackupEntries( struct BackupsWorker::RestoreStarter { BackupsWorker & backups_worker; + LoggerPtr log; std::shared_ptr restore_query; ContextPtr query_context; /// We have to keep `query_context` until the end of the operation because a pointer to it is stored inside the ThreadGroup we're using. ContextMutablePtr restore_context; @@ -702,6 +718,7 @@ struct BackupsWorker::RestoreStarter RestoreStarter(BackupsWorker & backups_worker_, const ASTPtr & query_, const ContextPtr & context_) : backups_worker(backups_worker_) + , log(backups_worker.log) , restore_query(std::static_pointer_cast(query_->clone())) , query_context(context_) , restore_context(Context::createCopy(query_context)) @@ -753,16 +770,17 @@ struct BackupsWorker::RestoreStarter } restore_coordination = backups_worker.makeRestoreCoordination(on_cluster, restore_settings, restore_context); - backups_worker.doRestore( - restore_query, - restore_id, - backup_name_for_logging, - backup_info, - restore_settings, - restore_coordination, - restore_context, - on_cluster, - cluster); + backups_worker.doRestore(restore_query, restore_id, backup_info, restore_settings, restore_coordination, restore_context, + on_cluster, cluster); + + /// The restore coordination is not needed anymore. + restore_coordination->finish(/* throw_if_error = */ true); + if (!is_internal_restore) + restore_coordination->cleanup(/* throw_if_error = */ true); + restore_coordination.reset(); + + LOG_INFO(log, "Restored from {} {} successfully", (is_internal_restore ? "internal backup" : "backup"), backup_name_for_logging); + backups_worker.setStatus(restore_id, BackupStatus::RESTORED); } void onException() @@ -770,12 +788,16 @@ struct BackupsWorker::RestoreStarter /// Something bad happened, some data were not restored. tryLogCurrentException(backups_worker.log, fmt::format("Failed to restore from {} {}", (is_internal_restore ? "internal backup" : "backup"), backup_name_for_logging)); - if (restore_coordination && restore_coordination->trySetError(std::current_exception())) + if (restore_coordination && restore_coordination->setError(std::current_exception(), /* throw_if_error = */ false)) { - restore_coordination->tryWaitForOtherHostsToFinishAfterError(); - restore_coordination->tryFinishAfterError(); + bool other_hosts_finished = !is_internal_restore + && (!restore_coordination->isRestoreQuerySentToOtherHosts() || restore_coordination->waitOtherHostsFinish(/* throw_if_error = */ false)); + if (restore_coordination->finish(/* throw_if_error = */ false) && other_hosts_finished) + restore_coordination->cleanup(/* throw_if_error = */ false); } + restore_coordination.reset(); + backups_worker.setStatusSafe(restore_id, getRestoreStatusFromCurrentException()); } }; @@ -838,7 +860,6 @@ BackupPtr BackupsWorker::openBackupForReading(const BackupInfo & backup_info, co void BackupsWorker::doRestore( const std::shared_ptr & restore_query, const OperationID & restore_id, - const String & backup_name_for_logging, const BackupInfo & backup_info, RestoreSettings restore_settings, std::shared_ptr restore_coordination, @@ -882,10 +903,10 @@ void BackupsWorker::doRestore( restore_settings.copySettingsToQuery(*restore_query); sendQueryToOtherHosts(*restore_query, cluster, restore_settings.shard_num, restore_settings.replica_num, context, {}, restore_coordination->getOnClusterInitializationKeeperRetriesInfo()); - restore_coordination->setRestoreQueryWasSentToOtherHosts(); + restore_coordination->setRestoreQueryIsSentToOtherHosts(); /// Wait until all the hosts have done with their restoring work. - restore_coordination->waitForOtherHostsToFinish(); + restore_coordination->waitOtherHostsFinish(/* throw_if_error = */ true); } else { @@ -905,12 +926,6 @@ void BackupsWorker::doRestore( backup, context, getThreadPool(ThreadPoolId::RESTORE), after_task_callback}; restorer.run(RestorerFromBackup::RESTORE); } - - /// The restore coordination is not needed anymore. - restore_coordination->finish(); - - LOG_INFO(log, "Restored from {} {} successfully", (is_internal_restore ? "internal backup" : "backup"), backup_name_for_logging); - setStatus(restore_id, BackupStatus::RESTORED); } @@ -943,7 +958,7 @@ BackupsWorker::makeBackupCoordination(bool on_cluster, const BackupSettings & ba if (!on_cluster) { return std::make_shared( - *backup_settings.backup_uuid, !backup_settings.deduplicate_files, allow_concurrent_backups, *concurrency_counters); + !backup_settings.deduplicate_files, allow_concurrent_backups, *concurrency_counters); } bool is_internal_backup = backup_settings.internal; @@ -981,8 +996,7 @@ BackupsWorker::makeRestoreCoordination(bool on_cluster, const RestoreSettings & { if (!on_cluster) { - return std::make_shared( - *restore_settings.restore_uuid, allow_concurrent_restores, *concurrency_counters); + return std::make_shared(allow_concurrent_restores, *concurrency_counters); } bool is_internal_restore = restore_settings.internal; diff --git a/src/Backups/BackupsWorker.h b/src/Backups/BackupsWorker.h index 37f91e269a9..2e5ca84f3f6 100644 --- a/src/Backups/BackupsWorker.h +++ b/src/Backups/BackupsWorker.h @@ -81,7 +81,6 @@ private: BackupMutablePtr backup, const std::shared_ptr & backup_query, const BackupOperationID & backup_id, - const String & backup_name_for_logging, const BackupSettings & backup_settings, std::shared_ptr backup_coordination, ContextMutablePtr context, @@ -102,7 +101,6 @@ private: void doRestore( const std::shared_ptr & restore_query, const BackupOperationID & restore_id, - const String & backup_name_for_logging, const BackupInfo & backup_info, RestoreSettings restore_settings, std::shared_ptr restore_coordination, diff --git a/src/Backups/IBackupCoordination.h b/src/Backups/IBackupCoordination.h index c0eb90de89b..8bd874b9d0d 100644 --- a/src/Backups/IBackupCoordination.h +++ b/src/Backups/IBackupCoordination.h @@ -20,29 +20,27 @@ class IBackupCoordination public: virtual ~IBackupCoordination() = default; + /// Sets that the backup query was sent to other hosts. + /// Function waitOtherHostsFinish() will check that to find out if it should really wait or not. + virtual void setBackupQueryIsSentToOtherHosts() = 0; + virtual bool isBackupQuerySentToOtherHosts() const = 0; + /// Sets the current stage and waits for other hosts to come to this stage too. virtual Strings setStage(const String & new_stage, const String & message, bool sync) = 0; - /// Sets that the backup query was sent to other hosts. - /// Function waitForOtherHostsToFinish() will check that to find out if it should really wait or not. - virtual void setBackupQueryWasSentToOtherHosts() = 0; - /// Lets other hosts know that the current host has encountered an error. - virtual bool trySetError(std::exception_ptr exception) = 0; - - /// Lets other hosts know that the current host has finished its work. - virtual void finish() = 0; - - /// Lets other hosts know that the current host has finished its work (as a part of error-handling process). - virtual bool tryFinishAfterError() noexcept = 0; + /// Returns true if the information is successfully passed so other hosts can read it. + virtual bool setError(std::exception_ptr exception, bool throw_if_error) = 0; /// Waits until all the other hosts finish their work. /// Stops waiting and throws an exception if another host encounters an error or if some host gets cancelled. - virtual void waitForOtherHostsToFinish() = 0; + virtual bool waitOtherHostsFinish(bool throw_if_error) const = 0; - /// Waits until all the other hosts finish their work (as a part of error-handling process). - /// Doesn't stops waiting if some host encounters an error or gets cancelled. - virtual bool tryWaitForOtherHostsToFinishAfterError() noexcept = 0; + /// Lets other hosts know that the current host has finished its work. + virtual bool finish(bool throw_if_error) = 0; + + /// Removes temporary nodes in ZooKeeper. + virtual bool cleanup(bool throw_if_error) = 0; struct PartNameAndChecksum { diff --git a/src/Backups/IRestoreCoordination.h b/src/Backups/IRestoreCoordination.h index daabf1745f3..cc7bfd24202 100644 --- a/src/Backups/IRestoreCoordination.h +++ b/src/Backups/IRestoreCoordination.h @@ -18,29 +18,27 @@ class IRestoreCoordination public: virtual ~IRestoreCoordination() = default; + /// Sets that the restore query was sent to other hosts. + /// Function waitOtherHostsFinish() will check that to find out if it should really wait or not. + virtual void setRestoreQueryIsSentToOtherHosts() = 0; + virtual bool isRestoreQuerySentToOtherHosts() const = 0; + /// Sets the current stage and waits for other hosts to come to this stage too. virtual Strings setStage(const String & new_stage, const String & message, bool sync) = 0; - /// Sets that the restore query was sent to other hosts. - /// Function waitForOtherHostsToFinish() will check that to find out if it should really wait or not. - virtual void setRestoreQueryWasSentToOtherHosts() = 0; - /// Lets other hosts know that the current host has encountered an error. - virtual bool trySetError(std::exception_ptr exception) = 0; - - /// Lets other hosts know that the current host has finished its work. - virtual void finish() = 0; - - /// Lets other hosts know that the current host has finished its work (as a part of error-handling process). - virtual bool tryFinishAfterError() noexcept = 0; + /// Returns true if the information is successfully passed so other hosts can read it. + virtual bool setError(std::exception_ptr exception, bool throw_if_error) = 0; /// Waits until all the other hosts finish their work. /// Stops waiting and throws an exception if another host encounters an error or if some host gets cancelled. - virtual void waitForOtherHostsToFinish() = 0; + virtual bool waitOtherHostsFinish(bool throw_if_error) const = 0; - /// Waits until all the other hosts finish their work (as a part of error-handling process). - /// Doesn't stops waiting if some host encounters an error or gets cancelled. - virtual bool tryWaitForOtherHostsToFinishAfterError() noexcept = 0; + /// Lets other hosts know that the current host has finished its work. + virtual bool finish(bool throw_if_error) = 0; + + /// Removes temporary nodes in ZooKeeper. + virtual bool cleanup(bool throw_if_error) = 0; /// Starts creating a table in a replicated database. Returns false if there is another host which is already creating this table. virtual bool acquireCreatingTableInReplicatedDatabase(const String & database_zk_path, const String & table_name) = 0; diff --git a/src/Backups/RestoreCoordinationLocal.cpp b/src/Backups/RestoreCoordinationLocal.cpp index 569f58f1909..a9eee1fb159 100644 --- a/src/Backups/RestoreCoordinationLocal.cpp +++ b/src/Backups/RestoreCoordinationLocal.cpp @@ -10,9 +10,9 @@ namespace DB { RestoreCoordinationLocal::RestoreCoordinationLocal( - const UUID & restore_uuid, bool allow_concurrent_restore_, BackupConcurrencyCounters & concurrency_counters_) + bool allow_concurrent_restore_, BackupConcurrencyCounters & concurrency_counters_) : log(getLogger("RestoreCoordinationLocal")) - , concurrency_check(restore_uuid, /* is_restore = */ true, /* on_cluster = */ false, allow_concurrent_restore_, concurrency_counters_) + , concurrency_check(/* is_restore = */ true, /* on_cluster = */ false, /* zookeeper_path = */ "", allow_concurrent_restore_, concurrency_counters_) { } diff --git a/src/Backups/RestoreCoordinationLocal.h b/src/Backups/RestoreCoordinationLocal.h index 6be357c4b7e..6e3262a8a2e 100644 --- a/src/Backups/RestoreCoordinationLocal.h +++ b/src/Backups/RestoreCoordinationLocal.h @@ -17,16 +17,16 @@ class ASTCreateQuery; class RestoreCoordinationLocal : public IRestoreCoordination { public: - RestoreCoordinationLocal(const UUID & restore_uuid_, bool allow_concurrent_restore_, BackupConcurrencyCounters & concurrency_counters_); + RestoreCoordinationLocal(bool allow_concurrent_restore_, BackupConcurrencyCounters & concurrency_counters_); ~RestoreCoordinationLocal() override; + void setRestoreQueryIsSentToOtherHosts() override {} + bool isRestoreQuerySentToOtherHosts() const override { return false; } Strings setStage(const String &, const String &, bool) override { return {}; } - void setRestoreQueryWasSentToOtherHosts() override {} - bool trySetError(std::exception_ptr) override { return true; } - void finish() override {} - bool tryFinishAfterError() noexcept override { return true; } - void waitForOtherHostsToFinish() override {} - bool tryWaitForOtherHostsToFinishAfterError() noexcept override { return true; } + bool setError(std::exception_ptr, bool) override { return true; } + bool waitOtherHostsFinish(bool) const override { return true; } + bool finish(bool) override { return true; } + bool cleanup(bool) override { return true; } /// Starts creating a table in a replicated database. Returns false if there is another host which is already creating this table. bool acquireCreatingTableInReplicatedDatabase(const String & database_zk_path, const String & table_name) override; diff --git a/src/Backups/RestoreCoordinationOnCluster.cpp b/src/Backups/RestoreCoordinationOnCluster.cpp index 2029ad8b072..fad7341c044 100644 --- a/src/Backups/RestoreCoordinationOnCluster.cpp +++ b/src/Backups/RestoreCoordinationOnCluster.cpp @@ -35,17 +35,21 @@ RestoreCoordinationOnCluster::RestoreCoordinationOnCluster( , current_host_index(BackupCoordinationOnCluster::findCurrentHostIndex(current_host, all_hosts)) , log(getLogger("RestoreCoordinationOnCluster")) , with_retries(log, get_zookeeper_, keeper_settings, process_list_element_, [root_zookeeper_path_](Coordination::ZooKeeperWithFaultInjection::Ptr zk) { zk->sync(root_zookeeper_path_); }) - , concurrency_check(restore_uuid_, /* is_restore = */ true, /* on_cluster = */ true, allow_concurrent_restore_, concurrency_counters_) - , stage_sync(/* is_restore = */ true, fs::path{zookeeper_path} / "stage", current_host, all_hosts, allow_concurrent_restore_, with_retries, schedule_, process_list_element_, log) - , cleaner(zookeeper_path, with_retries, log) + , cleaner(/* is_restore = */ true, zookeeper_path, with_retries, log) + , stage_sync(/* is_restore = */ true, fs::path{zookeeper_path} / "stage", current_host, all_hosts, allow_concurrent_restore_, concurrency_counters_, with_retries, schedule_, process_list_element_, log) { - createRootNodes(); + try + { + createRootNodes(); + } + catch (...) + { + stage_sync.setError(std::current_exception(), /* throw_if_error = */ false); + throw; + } } -RestoreCoordinationOnCluster::~RestoreCoordinationOnCluster() -{ - tryFinishImpl(); -} +RestoreCoordinationOnCluster::~RestoreCoordinationOnCluster() = default; void RestoreCoordinationOnCluster::createRootNodes() { @@ -66,69 +70,52 @@ void RestoreCoordinationOnCluster::createRootNodes() }); } +void RestoreCoordinationOnCluster::setRestoreQueryIsSentToOtherHosts() +{ + stage_sync.setQueryIsSentToOtherHosts(); +} + +bool RestoreCoordinationOnCluster::isRestoreQuerySentToOtherHosts() const +{ + return stage_sync.isQuerySentToOtherHosts(); +} + Strings RestoreCoordinationOnCluster::setStage(const String & new_stage, const String & message, bool sync) { stage_sync.setStage(new_stage, message); - - if (!sync) - return {}; - - return stage_sync.waitForHostsToReachStage(new_stage, all_hosts_without_initiator); + if (sync) + return stage_sync.waitHostsReachStage(all_hosts_without_initiator, new_stage); + return {}; } -void RestoreCoordinationOnCluster::setRestoreQueryWasSentToOtherHosts() +bool RestoreCoordinationOnCluster::setError(std::exception_ptr exception, bool throw_if_error) { - restore_query_was_sent_to_other_hosts = true; + return stage_sync.setError(exception, throw_if_error); } -bool RestoreCoordinationOnCluster::trySetError(std::exception_ptr exception) +bool RestoreCoordinationOnCluster::waitOtherHostsFinish(bool throw_if_error) const { - return stage_sync.trySetError(exception); + return stage_sync.waitOtherHostsFinish(throw_if_error); } -void RestoreCoordinationOnCluster::finish() +bool RestoreCoordinationOnCluster::finish(bool throw_if_error) { - bool other_hosts_also_finished = false; - stage_sync.finish(other_hosts_also_finished); - - if ((current_host == kInitiator) && (other_hosts_also_finished || !restore_query_was_sent_to_other_hosts)) - cleaner.cleanup(); + return stage_sync.finish(throw_if_error); } -bool RestoreCoordinationOnCluster::tryFinishAfterError() noexcept +bool RestoreCoordinationOnCluster::cleanup(bool throw_if_error) { - return tryFinishImpl(); -} - -bool RestoreCoordinationOnCluster::tryFinishImpl() noexcept -{ - bool other_hosts_also_finished = false; - if (!stage_sync.tryFinishAfterError(other_hosts_also_finished)) - return false; - - if ((current_host == kInitiator) && (other_hosts_also_finished || !restore_query_was_sent_to_other_hosts)) + /// All the hosts must finish before we remove the coordination nodes. + bool expect_other_hosts_finished = stage_sync.isQuerySentToOtherHosts() || !stage_sync.isErrorSet(); + bool all_hosts_finished = stage_sync.finished() && (stage_sync.otherHostsFinished() || !expect_other_hosts_finished); + if (!all_hosts_finished) { - if (!cleaner.tryCleanupAfterError()) - return false; - } - - return true; -} - -void RestoreCoordinationOnCluster::waitForOtherHostsToFinish() -{ - if ((current_host != kInitiator) || !restore_query_was_sent_to_other_hosts) - return; - stage_sync.waitForOtherHostsToFinish(); -} - -bool RestoreCoordinationOnCluster::tryWaitForOtherHostsToFinishAfterError() noexcept -{ - if (current_host != kInitiator) + auto unfinished_hosts = expect_other_hosts_finished ? stage_sync.getUnfinishedHosts() : Strings{current_host}; + LOG_INFO(log, "Skipping removing nodes from ZooKeeper because hosts {} didn't finish", + BackupCoordinationStageSync::getHostsDesc(unfinished_hosts)); return false; - if (!restore_query_was_sent_to_other_hosts) - return true; - return stage_sync.tryWaitForOtherHostsToFinishAfterError(); + } + return cleaner.cleanup(throw_if_error); } ZooKeeperRetriesInfo RestoreCoordinationOnCluster::getOnClusterInitializationKeeperRetriesInfo() const diff --git a/src/Backups/RestoreCoordinationOnCluster.h b/src/Backups/RestoreCoordinationOnCluster.h index 87a8dd3ce83..99929cbdac3 100644 --- a/src/Backups/RestoreCoordinationOnCluster.h +++ b/src/Backups/RestoreCoordinationOnCluster.h @@ -1,7 +1,6 @@ #pragma once #include -#include #include #include #include @@ -15,7 +14,7 @@ class RestoreCoordinationOnCluster : public IRestoreCoordination { public: /// Empty string as the current host is used to mark the initiator of a RESTORE ON CLUSTER query. - static const constexpr std::string_view kInitiator; + static const constexpr std::string_view kInitiator = BackupCoordinationStageSync::kInitiator; RestoreCoordinationOnCluster( const UUID & restore_uuid_, @@ -31,13 +30,13 @@ public: ~RestoreCoordinationOnCluster() override; + void setRestoreQueryIsSentToOtherHosts() override; + bool isRestoreQuerySentToOtherHosts() const override; Strings setStage(const String & new_stage, const String & message, bool sync) override; - void setRestoreQueryWasSentToOtherHosts() override; - bool trySetError(std::exception_ptr exception) override; - void finish() override; - bool tryFinishAfterError() noexcept override; - void waitForOtherHostsToFinish() override; - bool tryWaitForOtherHostsToFinishAfterError() noexcept override; + bool setError(std::exception_ptr exception, bool throw_if_error) override; + bool waitOtherHostsFinish(bool throw_if_error) const override; + bool finish(bool throw_if_error) override; + bool cleanup(bool throw_if_error) override; /// Starts creating a table in a replicated database. Returns false if there is another host which is already creating this table. bool acquireCreatingTableInReplicatedDatabase(const String & database_zk_path, const String & table_name) override; @@ -78,11 +77,10 @@ private: const size_t current_host_index; LoggerPtr const log; + /// The order is important: `stage_sync` must be initialized after `with_retries` and `cleaner`. const WithRetries with_retries; - BackupConcurrencyCheck concurrency_check; - BackupCoordinationStageSync stage_sync; BackupCoordinationCleaner cleaner; - std::atomic restore_query_was_sent_to_other_hosts = false; + BackupCoordinationStageSync stage_sync; }; } From 19bcc5550bad0444d652d760edddbe15fe0611da Mon Sep 17 00:00:00 2001 From: Vitaly Baranov Date: Mon, 11 Nov 2024 01:36:20 +0100 Subject: [PATCH 622/680] Fix tests. --- .../test_cancel_backup.py | 29 +++++++------------ 1 file changed, 11 insertions(+), 18 deletions(-) diff --git a/tests/integration/test_backup_restore_on_cluster/test_cancel_backup.py b/tests/integration/test_backup_restore_on_cluster/test_cancel_backup.py index f63dc2aef3d..4ad53acc735 100644 --- a/tests/integration/test_backup_restore_on_cluster/test_cancel_backup.py +++ b/tests/integration/test_backup_restore_on_cluster/test_cancel_backup.py @@ -251,23 +251,16 @@ def kill_query( if is_initial_query is not None else "" ) + old_time = time.monotonic() node.query( f"KILL QUERY WHERE (query_kind='{query_kind}') AND (query LIKE '%{id}%'){filter_for_is_initial_query} SYNC" ) - node.query("SYSTEM FLUSH LOGS") - duration = ( - int( - node.query( - f"SELECT query_duration_ms FROM system.query_log WHERE query_kind='KillQuery' AND query LIKE '%{id}%' AND type='QueryFinish'" - ) - ) - / 1000 - ) + waited = time.monotonic() - old_time print( - f"{get_node_name(node)}: Cancelled {operation_name} {id} after {duration} seconds" + f"{get_node_name(node)}: Cancelled {operation_name} {id} after {waited} seconds" ) if timeout is not None: - assert duration < timeout + assert waited < timeout # Stops all ZooKeeper servers. @@ -305,7 +298,7 @@ def sleep(seconds): class NoTrashChecker: def __init__(self): self.expect_backups = [] - self.expect_unfinished_backups = [] + self.allow_unfinished_backups = [] self.expect_errors = [] self.allow_errors = [] self.check_zookeeper = True @@ -373,7 +366,7 @@ class NoTrashChecker: if unfinished_backups: print(f"Found unfinished backups: {unfinished_backups}") assert new_backups == set(self.expect_backups) - assert unfinished_backups == set(self.expect_unfinished_backups) + assert unfinished_backups.difference(self.allow_unfinished_backups) == set() all_errors = set() start_time = time.strftime( @@ -641,7 +634,7 @@ def test_long_disconnection_stops_backup(): assert get_status(initiator, backup_id=backup_id) == "CREATING_BACKUP" assert get_num_system_processes(initiator, backup_id=backup_id) >= 1 - no_trash_checker.expect_unfinished_backups = [backup_id] + no_trash_checker.allow_unfinished_backups = [backup_id] no_trash_checker.allow_errors = [ "FAILED_TO_SYNC_BACKUP_OR_RESTORE", "KEEPER_EXCEPTION", @@ -674,7 +667,7 @@ def test_long_disconnection_stops_backup(): # A backup is expected to fail, but it isn't expected to fail too soon. print(f"Backup failed after {time_to_fail} seconds disconnection") assert time_to_fail > 3 - assert time_to_fail < 30 + assert time_to_fail < 35 # A backup must NOT be stopped if Zookeeper is disconnected shorter than `failure_after_host_disconnected_for_seconds`. @@ -695,7 +688,7 @@ def test_short_disconnection_doesnt_stop_backup(): backup_id = random_id() initiator.query( f"BACKUP TABLE tbl ON CLUSTER 'cluster' TO {get_backup_name(backup_id)} SETTINGS id='{backup_id}' ASYNC", - settings={"backup_restore_failure_after_host_disconnected_for_seconds": 6}, + settings={"backup_restore_failure_after_host_disconnected_for_seconds": 10}, ) assert get_status(initiator, backup_id=backup_id) == "CREATING_BACKUP" @@ -703,13 +696,13 @@ def test_short_disconnection_doesnt_stop_backup(): # Dropping connection for less than `failure_after_host_disconnected_for_seconds` with PartitionManager() as pm: - random_sleep(3) + random_sleep(4) node_to_drop_zk_connection = random_node() print( f"Dropping connection between {get_node_name(node_to_drop_zk_connection)} and ZooKeeper" ) pm.drop_instance_zk_connections(node_to_drop_zk_connection) - random_sleep(3) + random_sleep(4) print( f"Restoring connection between {get_node_name(node_to_drop_zk_connection)} and ZooKeeper" ) From c4946cf1594c6083c01a39954495aea1be01f574 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Mi=D1=81hael=20Stetsyuk?= <59827607+mstetsyuk@users.noreply.github.com> Date: Mon, 4 Nov 2024 10:41:26 +0000 Subject: [PATCH 623/680] style fix --- src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp | 1 - 1 file changed, 1 deletion(-) diff --git a/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp b/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp index c73c9f6d048..addaeb65350 100644 --- a/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp +++ b/src/Storages/MergeTree/ReplicatedMergeTreeRestartingThread.cpp @@ -29,7 +29,6 @@ namespace MergeTreeSetting namespace ErrorCodes { extern const int REPLICA_IS_ALREADY_ACTIVE; - extern const int REPLICA_STATUS_CHANGED; extern const int LOGICAL_ERROR; extern const int SUPPORT_IS_DISABLED; } From 05dfc6dbdba48964cfd147a3635613966da78f0a Mon Sep 17 00:00:00 2001 From: kssenii Date: Mon, 11 Nov 2024 11:53:24 +0100 Subject: [PATCH 624/680] Update settings changes history --- src/Core/SettingsChangesHistory.cpp | 1 + 1 file changed, 1 insertion(+) diff --git a/src/Core/SettingsChangesHistory.cpp b/src/Core/SettingsChangesHistory.cpp index c6223bef2b2..7eb8455a169 100644 --- a/src/Core/SettingsChangesHistory.cpp +++ b/src/Core/SettingsChangesHistory.cpp @@ -76,6 +76,7 @@ static std::initializer_list Date: Mon, 11 Nov 2024 13:26:31 +0200 Subject: [PATCH 625/680] Fix typo Fix log message for more clean understanding --- docker/server/entrypoint.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docker/server/entrypoint.sh b/docker/server/entrypoint.sh index 2f87008f2e5..947244dd97f 100755 --- a/docker/server/entrypoint.sh +++ b/docker/server/entrypoint.sh @@ -162,7 +162,7 @@ if [ -n "${RUN_INITDB_SCRIPTS}" ]; then tries=${CLICKHOUSE_INIT_TIMEOUT:-1000} while ! wget --spider --no-check-certificate -T 1 -q "$URL" 2>/dev/null; do if [ "$tries" -le "0" ]; then - echo >&2 'ClickHouse init process failed.' + echo >&2 'ClickHouse init process timeout.' exit 1 fi tries=$(( tries-1 )) From 33f9e8bc2e5540386e5ccf7fec591eaa1bf5cc24 Mon Sep 17 00:00:00 2001 From: nauu Date: Mon, 11 Nov 2024 20:25:57 +0800 Subject: [PATCH 626/680] fix error --- src/IO/S3/URI.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/IO/S3/URI.cpp b/src/IO/S3/URI.cpp index ad746ff3326..aefe3ff338c 100644 --- a/src/IO/S3/URI.cpp +++ b/src/IO/S3/URI.cpp @@ -117,7 +117,7 @@ URI::URI(const std::string & uri_, bool allow_archive_path_syntax) is_virtual_hosted_style = true; if (name == "oss-data-acc") { - bucket = bucket.substr(0, bucket.find(".")); + bucket = bucket.substr(0, bucket.find('.')); endpoint = uri.getScheme() + "://" + uri.getHost().substr(bucket.length() + 1); } else From 5f0f2628b15988fda3467b50e21be9a149fc9eb7 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ra=C3=BAl=20Mar=C3=ADn?= Date: Mon, 11 Nov 2024 13:46:50 +0100 Subject: [PATCH 627/680] Avoid failures on fault injection --- tests/docker_scripts/attach_gdb.lib | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/docker_scripts/attach_gdb.lib b/tests/docker_scripts/attach_gdb.lib index 4170a19176c..f8a08b5e39d 100644 --- a/tests/docker_scripts/attach_gdb.lib +++ b/tests/docker_scripts/attach_gdb.lib @@ -5,7 +5,8 @@ source /repo/tests/docker_scripts/utils.lib function attach_gdb_to_clickhouse() { - IS_ASAN=$(clickhouse-client --query "SELECT count() FROM system.build_options WHERE name = 'CXX_FLAGS' AND position('sanitize=address' IN value)") + # Use retries to avoid failures due to fault injections + IS_ASAN=$(run_with_retry 5 clickhouse-client --query "SELECT count() FROM system.build_options WHERE name = 'CXX_FLAGS' AND position('sanitize=address' IN value)") if [[ "$IS_ASAN" = "1" ]]; then echo "ASAN build detected. Not using gdb since it disables LeakSanitizer detections" From 17f7097d5b66129a3f72f98114cd28575ed839dc Mon Sep 17 00:00:00 2001 From: avogar Date: Mon, 11 Nov 2024 13:28:52 +0000 Subject: [PATCH 628/680] Fix CAST from LowCardinality(Nullable) to Dynamic --- src/Functions/FunctionsConversion.cpp | 2 +- ...3261_low_cardinality_nullable_to_dynamic_cast.reference | 2 ++ .../03261_low_cardinality_nullable_to_dynamic_cast.sql | 7 +++++++ 3 files changed, 10 insertions(+), 1 deletion(-) create mode 100644 tests/queries/0_stateless/03261_low_cardinality_nullable_to_dynamic_cast.reference create mode 100644 tests/queries/0_stateless/03261_low_cardinality_nullable_to_dynamic_cast.sql diff --git a/src/Functions/FunctionsConversion.cpp b/src/Functions/FunctionsConversion.cpp index 0f6311c9716..5f1583f6e71 100644 --- a/src/Functions/FunctionsConversion.cpp +++ b/src/Functions/FunctionsConversion.cpp @@ -4390,7 +4390,7 @@ private: variant_column = IColumn::mutate(column); /// Otherwise we should filter column. else - variant_column = column->filter(filter, variant_size_hint)->assumeMutable(); + variant_column = IColumn::mutate(column->filter(filter, variant_size_hint)); assert_cast(*variant_column).nestedRemoveNullable(); return createVariantFromDescriptorsAndOneNonEmptyVariant(variant_types, std::move(discriminators), std::move(variant_column), variant_discr); diff --git a/tests/queries/0_stateless/03261_low_cardinality_nullable_to_dynamic_cast.reference b/tests/queries/0_stateless/03261_low_cardinality_nullable_to_dynamic_cast.reference new file mode 100644 index 00000000000..96e34d5a44c --- /dev/null +++ b/tests/queries/0_stateless/03261_low_cardinality_nullable_to_dynamic_cast.reference @@ -0,0 +1,2 @@ +\N +\N diff --git a/tests/queries/0_stateless/03261_low_cardinality_nullable_to_dynamic_cast.sql b/tests/queries/0_stateless/03261_low_cardinality_nullable_to_dynamic_cast.sql new file mode 100644 index 00000000000..fdb497a62bf --- /dev/null +++ b/tests/queries/0_stateless/03261_low_cardinality_nullable_to_dynamic_cast.sql @@ -0,0 +1,7 @@ +SET allow_suspicious_low_cardinality_types = 1, allow_experimental_dynamic_type = 1; +DROP TABLE IF EXISTS t0; +CREATE TABLE t0 (c0 LowCardinality(Nullable(Int))) ENGINE = Memory(); +INSERT INTO TABLE t0 (c0) VALUES (NULL); +SELECT c0::Dynamic FROM t0; +SELECT c0 FROM t0; +DROP TABLE t0; From 288756bc9aede92c6d005af34be94973a5d78203 Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Mon, 11 Nov 2024 13:32:01 +0000 Subject: [PATCH 629/680] Fix for stateful functions. --- .../QueryPlan/BuildQueryPipelineSettings.cpp | 13 ++++++++++++- .../QueryPlan/BuildQueryPipelineSettings.h | 2 ++ src/Processors/QueryPlan/FilterStep.cpp | 12 ++++++++++-- .../queries/0_stateless/03199_merge_filters_bug.sql | 2 +- 4 files changed, 25 insertions(+), 4 deletions(-) diff --git a/src/Processors/QueryPlan/BuildQueryPipelineSettings.cpp b/src/Processors/QueryPlan/BuildQueryPipelineSettings.cpp index fb3ed7f80fc..ce02ef8b9ba 100644 --- a/src/Processors/QueryPlan/BuildQueryPipelineSettings.cpp +++ b/src/Processors/QueryPlan/BuildQueryPipelineSettings.cpp @@ -6,12 +6,23 @@ namespace DB { +namespace Setting +{ + extern const SettingsBool query_plan_merge_filters; +} + BuildQueryPipelineSettings BuildQueryPipelineSettings::fromContext(ContextPtr from) { + const auto & query_settings = from->getSettingsRef(); BuildQueryPipelineSettings settings; - settings.actions_settings = ExpressionActionsSettings::fromSettings(from->getSettingsRef(), CompileExpressions::yes); + settings.actions_settings = ExpressionActionsSettings::fromSettings(query_settings, CompileExpressions::yes); settings.process_list_element = from->getProcessListElement(); settings.progress_callback = from->getProgressCallback(); + + /// Setting query_plan_merge_filters is enabled by default. + /// But it can brake short-circuit without splitting fiter step into smaller steps. + /// So, enable and disable this optimizations together. + settings.enable_multiple_filters_transforms_for_and_chain = query_settings[Setting::query_plan_merge_filters]; return settings; } diff --git a/src/Processors/QueryPlan/BuildQueryPipelineSettings.h b/src/Processors/QueryPlan/BuildQueryPipelineSettings.h index d99f9a7d1f1..6219e37db58 100644 --- a/src/Processors/QueryPlan/BuildQueryPipelineSettings.h +++ b/src/Processors/QueryPlan/BuildQueryPipelineSettings.h @@ -17,6 +17,8 @@ using TemporaryFileLookupPtr = std::shared_ptr; struct BuildQueryPipelineSettings { + bool enable_multiple_filters_transforms_for_and_chain = true; + ExpressionActionsSettings actions_settings; QueryStatusPtr process_list_element; ProgressCallback progress_callback = nullptr; diff --git a/src/Processors/QueryPlan/FilterStep.cpp b/src/Processors/QueryPlan/FilterStep.cpp index 3d56a2352dc..a6b157cdd1d 100644 --- a/src/Processors/QueryPlan/FilterStep.cpp +++ b/src/Processors/QueryPlan/FilterStep.cpp @@ -139,7 +139,11 @@ FilterStep::FilterStep( void FilterStep::transformPipeline(QueryPipelineBuilder & pipeline, const BuildQueryPipelineSettings & settings) { - auto and_atoms = splitAndChainIntoMultipleFilters(actions_dag, filter_column_name); + std::vector and_atoms; + + if (settings.enable_multiple_filters_transforms_for_and_chain && !actions_dag.hasStatefulFunctions()) + and_atoms = splitAndChainIntoMultipleFilters(actions_dag, filter_column_name); + for (auto & and_atom : and_atoms) { auto expression = std::make_shared(std::move(and_atom.dag), settings.getActionsSettings()); @@ -178,7 +182,11 @@ void FilterStep::describeActions(FormatSettings & settings) const String prefix(settings.offset, settings.indent_char); auto cloned_dag = actions_dag.clone(); - auto and_atoms = splitAndChainIntoMultipleFilters(cloned_dag, filter_column_name); + + std::vector and_atoms; + if (!actions_dag.hasStatefulFunctions()) + and_atoms = splitAndChainIntoMultipleFilters(cloned_dag, filter_column_name); + for (auto & and_atom : and_atoms) { auto expression = std::make_shared(std::move(and_atom.dag)); diff --git a/tests/queries/0_stateless/03199_merge_filters_bug.sql b/tests/queries/0_stateless/03199_merge_filters_bug.sql index ed2ec2ea217..bb2a4255a3d 100644 --- a/tests/queries/0_stateless/03199_merge_filters_bug.sql +++ b/tests/queries/0_stateless/03199_merge_filters_bug.sql @@ -49,7 +49,7 @@ tmp1 AS fs1 FROM t2 LEFT JOIN tmp1 USING (fs1) - WHERE (fs1 IN ('test')) SETTINGS enable_multiple_prewhere_read_steps = 0; + WHERE (fs1 IN ('test')) SETTINGS enable_multiple_prewhere_read_steps = 0, query_plan_merge_filters=0; optimize table t1 final; From 8c2e541392e552343431a6b9b411ee55f37e8fe8 Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Mon, 11 Nov 2024 14:27:48 +0000 Subject: [PATCH 630/680] Avoid using manes in multistage prewhere optimization. --- .../MergeTreeSplitPrewhereIntoReadSteps.cpp | 110 ++++++++++-------- 1 file changed, 60 insertions(+), 50 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp b/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp index 9c82817e8cb..73fe2600946 100644 --- a/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp +++ b/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp @@ -4,6 +4,7 @@ #include #include #include +#include #include @@ -57,9 +58,9 @@ struct DAGNodeRef const ActionsDAG::Node * node; }; -/// Result name -> DAGNodeRef -using OriginalToNewNodeMap = std::unordered_map; -using NodeNameToLastUsedStepMap = std::unordered_map; +/// Result -> DAGNodeRef +using OriginalToNewNodeMap = std::unordered_map; +using NodeNameToLastUsedStepMap = std::unordered_map; /// Clones the part of original DAG responsible for computing the original_dag_node and adds it to the new DAG. const ActionsDAG::Node & addClonedDAGToDAG( @@ -69,12 +70,12 @@ const ActionsDAG::Node & addClonedDAGToDAG( OriginalToNewNodeMap & node_remap, NodeNameToLastUsedStepMap & node_to_step_map) { - const String & node_name = original_dag_node->result_name; + //const String & node_name = original_dag_node->result_name; /// Look for the node in the map of already known nodes - if (node_remap.contains(node_name)) + if (node_remap.contains(original_dag_node)) { /// If the node is already in the new DAG, return it - const auto & node_ref = node_remap.at(node_name); + const auto & node_ref = node_remap.at(original_dag_node); if (node_ref.dag == new_dag.get()) return *node_ref.node; @@ -83,11 +84,11 @@ const ActionsDAG::Node & addClonedDAGToDAG( { node_ref.dag->addOrReplaceInOutputs(*node_ref.node); const auto & new_node = new_dag->addInput(node_ref.node->result_name, node_ref.node->result_type); - node_remap[node_name] = {new_dag.get(), &new_node}; /// TODO: here we update the node reference. Is it always correct? + node_remap[original_dag_node] = {new_dag.get(), &new_node}; /// TODO: here we update the node reference. Is it always correct? /// Remember the index of the last step which reuses this node. /// We cannot remove this node from the outputs before that step. - node_to_step_map[node_name] = step; + node_to_step_map[original_dag_node] = step; return new_node; } } @@ -96,7 +97,7 @@ const ActionsDAG::Node & addClonedDAGToDAG( if (original_dag_node->type == ActionsDAG::ActionType::INPUT) { const auto & new_node = new_dag->addInput(original_dag_node->result_name, original_dag_node->result_type); - node_remap[node_name] = {new_dag.get(), &new_node}; + node_remap[original_dag_node] = {new_dag.get(), &new_node}; return new_node; } @@ -105,7 +106,7 @@ const ActionsDAG::Node & addClonedDAGToDAG( { const auto & new_node = new_dag->addColumn( ColumnWithTypeAndName(original_dag_node->column, original_dag_node->result_type, original_dag_node->result_name)); - node_remap[node_name] = {new_dag.get(), &new_node}; + node_remap[original_dag_node] = {new_dag.get(), &new_node}; return new_node; } @@ -113,7 +114,7 @@ const ActionsDAG::Node & addClonedDAGToDAG( { const auto & alias_child = addClonedDAGToDAG(step, original_dag_node->children[0], new_dag, node_remap, node_to_step_map); const auto & new_node = new_dag->addAlias(alias_child, original_dag_node->result_name); - node_remap[node_name] = {new_dag.get(), &new_node}; + node_remap[original_dag_node] = {new_dag.get(), &new_node}; return new_node; } @@ -128,7 +129,7 @@ const ActionsDAG::Node & addClonedDAGToDAG( } const auto & new_node = new_dag->addFunction(original_dag_node->function_base, new_children, original_dag_node->result_name); - node_remap[node_name] = {new_dag.get(), &new_node}; + node_remap[original_dag_node] = {new_dag.get(), &new_node}; return new_node; } @@ -138,11 +139,11 @@ const ActionsDAG::Node & addClonedDAGToDAG( const ActionsDAG::Node & addFunction( const ActionsDAGPtr & new_dag, const FunctionOverloadResolverPtr & function, - ActionsDAG::NodeRawConstPtrs children, - OriginalToNewNodeMap & node_remap) + ActionsDAG::NodeRawConstPtrs children) + //OriginalToNewNodeMap & node_remap) { const auto & new_node = new_dag->addFunction(function, children, ""); - node_remap[new_node.result_name] = {new_dag.get(), &new_node}; + //node_remap[new_node.result_name] = {new_dag.get(), &new_node}; return new_node; } @@ -152,14 +153,14 @@ const ActionsDAG::Node & addFunction( const ActionsDAG::Node & addCast( const ActionsDAGPtr & dag, const ActionsDAG::Node & node_to_cast, - const DataTypePtr & to_type, - OriginalToNewNodeMap & node_remap) + const DataTypePtr & to_type) + //[[maybe_unused]] OriginalToNewNodeMap & node_remap) { if (!node_to_cast.result_type->equals(*to_type)) return node_to_cast; const auto & new_node = dag->addCast(node_to_cast, to_type, {}); - node_remap[new_node.result_name] = {dag.get(), &new_node}; + //node_remap[new_node.result_name] = {dag.get(), &new_node}; return new_node; } @@ -169,8 +170,8 @@ const ActionsDAG::Node & addCast( /// 2. makes sure that the result contains only 0 or 1 values even if the source column contains non-boolean values. const ActionsDAG::Node & addAndTrue( const ActionsDAGPtr & dag, - const ActionsDAG::Node & filter_node_to_normalize, - OriginalToNewNodeMap & node_remap) + const ActionsDAG::Node & filter_node_to_normalize) + //OriginalToNewNodeMap & node_remap) { Field const_true_value(true); @@ -181,7 +182,7 @@ const ActionsDAG::Node & addAndTrue( const auto * const_true_node = &dag->addColumn(std::move(const_true_column)); ActionsDAG::NodeRawConstPtrs children = {&filter_node_to_normalize, const_true_node}; FunctionOverloadResolverPtr func_builder_and = std::make_unique(std::make_shared()); - return addFunction(dag, func_builder_and, children, node_remap); + return addFunction(dag, func_builder_and, children); //, node_remap); } } @@ -243,7 +244,11 @@ bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionAction struct Step { ActionsDAGPtr actions; - String column_name; + /// Original condition, in case if we have only one condition, and it was not casted + const ActionsDAG::Node * original_node; + /// Result condition node + const ActionsDAG::Node * result_node; + //String column_name; }; std::vector steps; @@ -254,7 +259,9 @@ bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionAction { const auto & condition_group = condition_groups[step_index]; ActionsDAGPtr step_dag = std::make_unique(); - String result_name; + const ActionsDAG::Node * original_node = nullptr; + const ActionsDAG::Node * result_node; + //String result_name; std::vector new_condition_nodes; for (const auto * node : condition_group) @@ -267,48 +274,47 @@ bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionAction { /// Add AND function to combine the conditions FunctionOverloadResolverPtr func_builder_and = std::make_unique(std::make_shared()); - const auto & and_function_node = addFunction(step_dag, func_builder_and, new_condition_nodes, node_remap); - step_dag->addOrReplaceInOutputs(and_function_node); - result_name = and_function_node.result_name; + const auto & and_function_node = addFunction(step_dag, func_builder_and, new_condition_nodes); //, node_remap); + //step_dag->addOrReplaceInOutputs(and_function_node); + result_node = &and_function_node; } else { - const auto & result_node = *new_condition_nodes.front(); + result_node = new_condition_nodes.front(); /// Check if explicit cast is needed for the condition to serve as a filter. - const auto result_type_name = result_node.result_type->getName(); - if (result_type_name == "UInt8" || - result_type_name == "Nullable(UInt8)" || - result_type_name == "LowCardinality(UInt8)" || - result_type_name == "LowCardinality(Nullable(UInt8))") + //const auto result_type_name = result_node->result_type->getName(); + if (isUInt8(removeNullable(removeLowCardinality(result_node->result_type)))) { /// No need to cast - step_dag->addOrReplaceInOutputs(result_node); - result_name = result_node.result_name; + //step_dag->addOrReplaceInOutputs(result_node); + //result_name = result_node.result_name; } else { /// Build "condition AND True" expression to "cast" the condition to UInt8 or Nullable(UInt8) depending on its type. - const auto & cast_node = addAndTrue(step_dag, result_node, node_remap); - step_dag->addOrReplaceInOutputs(cast_node); - result_name = cast_node.result_name; + result_node = &addAndTrue(step_dag, *result_node); //, node_remap); + //step_dag->addOrReplaceInOutputs(cast_node); + //result_name = &cast_node.result_name; } } - steps.push_back({std::move(step_dag), result_name}); + step_dag->getOutputs().insert(step_dag->getOutputs().begin(), result_node); + steps.push_back({std::move(step_dag), original_node, result_node}); } /// 6. Find all outputs of the original DAG auto original_outputs = prewhere_info->prewhere_actions.getOutputs(); + steps.back().actions->getOutputs().clear(); /// 7. Find all outputs that were computed in the already built DAGs, mark these nodes as outputs in the steps where they were computed /// 8. Add computation of the remaining outputs to the last step with the procedure similar to 4 - NameSet all_output_names; + std::unordered_set all_outputs; for (const auto * output : original_outputs) { - all_output_names.insert(output->result_name); - if (node_remap.contains(output->result_name)) + all_outputs.insert(output); + if (node_remap.contains(output)) //->result_name)) { - const auto & new_node_info = node_remap[output->result_name]; - new_node_info.dag->addOrReplaceInOutputs(*new_node_info.node); + const auto & new_node_info = node_remap[output]; + new_node_info.dag->getOutputs().push_back(new_node_info.node); } else if (output->result_name == prewhere_info->prewhere_column_name) { @@ -319,20 +325,23 @@ bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionAction /// 1. AND the last condition with constant True. This is needed to make sure that in the last step filter has UInt8 type /// but contains values other than 0 and 1 (e.g. if it is (number%5) it contains 2,3,4) /// 2. CAST the result to the exact type of the PREWHERE column from the original DAG - const auto & last_step_result_node_info = node_remap[steps.back().column_name]; + //const auto & last_step_result_node_info = node_remap[steps.back().column_name]; auto & last_step_dag = steps.back().actions; + auto & last_step_result_node = steps.back().result_node; /// Build AND(last_step_result_node, true) - const auto & and_node = addAndTrue(last_step_dag, *last_step_result_node_info.node, node_remap); + const auto & and_node = addAndTrue(last_step_dag, *last_step_result_node); //, node_remap); /// Build CAST(and_node, type of PREWHERE column) - const auto & cast_node = addCast(last_step_dag, and_node, output->result_type, node_remap); + const auto & cast_node = addCast(last_step_dag, and_node, output->result_type); //, node_remap); /// Add alias for the result with the name of the PREWHERE column const auto & prewhere_result_node = last_step_dag->addAlias(cast_node, output->result_name); - last_step_dag->addOrReplaceInOutputs(prewhere_result_node); + //last_step_dag->addOrReplaceInOutputs(prewhere_result_node); + last_step_dag->getOutputs().push_back(&prewhere_result_node); + steps.back().result_node = &prewhere_result_node; } else { const auto & node_in_new_dag = addClonedDAGToDAG(steps.size() - 1, output, steps.back().actions, node_remap, node_to_step); - steps.back().actions->addOrReplaceInOutputs(node_in_new_dag); + steps.back().actions->getOutputs().push_back(&node_in_new_dag); } } @@ -345,10 +354,10 @@ bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionAction { .type = PrewhereExprStep::Filter, .actions = std::make_shared(std::move(*step.actions), actions_settings), - .filter_column_name = step.column_name, + .filter_column_name = step.result_node->result_name, /// Don't remove if it's in the list of original outputs .remove_filter_column = - !all_output_names.contains(step.column_name) && node_to_step[step.column_name] <= step_index, + step.original_node && !all_outputs.contains(step.original_node) && node_to_step[step.original_node] <= step_index, .need_filter = false, .perform_alter_conversions = true, }; @@ -356,6 +365,7 @@ bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionAction prewhere.steps.push_back(std::make_shared(std::move(new_step))); } + prewhere.steps.back()->remove_filter_column = prewhere_info->remove_prewhere_column; prewhere.steps.back()->need_filter = prewhere_info->need_filter; } From bcab2d51aa47f66d88ecc1c17463e2754260826d Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Mon, 11 Nov 2024 15:58:06 +0100 Subject: [PATCH 631/680] Use get_parameter_from_ssm in ci_buddy --- tests/ci/ci_buddy.py | 11 ++--------- 1 file changed, 2 insertions(+), 9 deletions(-) diff --git a/tests/ci/ci_buddy.py b/tests/ci/ci_buddy.py index 164af72f4be..07b748180cd 100644 --- a/tests/ci/ci_buddy.py +++ b/tests/ci/ci_buddy.py @@ -3,14 +3,13 @@ import json import os from typing import Dict, List, Union -import boto3 import requests from botocore.exceptions import ClientError from ci_config import CI from ci_utils import WithIter from commit_status_helper import get_commit_filtered_statuses, get_repo -from get_robot_token import get_best_robot_token +from get_robot_token import get_best_robot_token, get_parameter_from_ssm from github_helper import GitHub from pr_info import PRInfo @@ -89,15 +88,9 @@ class CIBuddy: def _get_webhooks(): name = "ci_buddy_web_hooks" - session = boto3.Session(region_name="us-east-1") # Replace with your region - ssm_client = session.client("ssm") json_string = None try: - response = ssm_client.get_parameter( - Name=name, - WithDecryption=True, # Set to True if the parameter is a SecureString - ) - json_string = response["Parameter"]["Value"] + json_string = get_parameter_from_ssm(name, decrypt=True) except ClientError as e: print(f"An error occurred: {e}") From 5c5016218b77cf77323c126bc064d90601beadef Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Mon, 11 Nov 2024 15:05:53 +0000 Subject: [PATCH 632/680] Fixing style. --- src/Processors/QueryPlan/BuildQueryPipelineSettings.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Processors/QueryPlan/BuildQueryPipelineSettings.cpp b/src/Processors/QueryPlan/BuildQueryPipelineSettings.cpp index ce02ef8b9ba..1832cc2ad42 100644 --- a/src/Processors/QueryPlan/BuildQueryPipelineSettings.cpp +++ b/src/Processors/QueryPlan/BuildQueryPipelineSettings.cpp @@ -20,7 +20,7 @@ BuildQueryPipelineSettings BuildQueryPipelineSettings::fromContext(ContextPtr fr settings.progress_callback = from->getProgressCallback(); /// Setting query_plan_merge_filters is enabled by default. - /// But it can brake short-circuit without splitting fiter step into smaller steps. + /// But it can brake short-circuit without splitting filter step into smaller steps. /// So, enable and disable this optimizations together. settings.enable_multiple_filters_transforms_for_and_chain = query_settings[Setting::query_plan_merge_filters]; return settings; From b7d80728190f1e56de3739186543afb575cf2063 Mon Sep 17 00:00:00 2001 From: Vitaly Baranov Date: Mon, 11 Nov 2024 16:06:17 +0100 Subject: [PATCH 633/680] Add waiting for prometheus instances to start before running test "test_prometheus_protocols". --- tests/integration/helpers/cluster.py | 25 ++++++++++++++++++- .../test_prometheus_protocols/test.py | 4 +-- 2 files changed, 26 insertions(+), 3 deletions(-) diff --git a/tests/integration/helpers/cluster.py b/tests/integration/helpers/cluster.py index b24593602ec..a0c2e1d1a70 100644 --- a/tests/integration/helpers/cluster.py +++ b/tests/integration/helpers/cluster.py @@ -744,11 +744,13 @@ class ClickHouseCluster: # available when with_prometheus == True self.with_prometheus = False self.prometheus_writer_host = "prometheus_writer" + self.prometheus_writer_ip = None self.prometheus_writer_port = 9090 self.prometheus_writer_logs_dir = p.abspath( p.join(self.instances_dir, "prometheus_writer/logs") ) self.prometheus_reader_host = "prometheus_reader" + self.prometheus_reader_ip = None self.prometheus_reader_port = 9091 self.prometheus_reader_logs_dir = p.abspath( p.join(self.instances_dir, "prometheus_reader/logs") @@ -2728,6 +2730,16 @@ class ClickHouseCluster: raise Exception("Can't wait LDAP to start") + def wait_prometheus_to_start(self): + self.prometheus_reader_ip = self.get_instance_ip(self.prometheus_reader_host) + self.prometheus_writer_ip = self.get_instance_ip(self.prometheus_writer_host) + self.wait_for_url( + f"http://{self.prometheus_reader_ip}:{self.prometheus_reader_port}/api/v1/query?query=time()" + ) + self.wait_for_url( + f"http://{self.prometheus_writer_ip}:{self.prometheus_writer_port}/api/v1/query?query=time()" + ) + def start(self): pytest_xdist_logging_to_separate_files.setup() logging.info("Running tests in {}".format(self.base_path)) @@ -3083,12 +3095,23 @@ class ClickHouseCluster: f"http://{self.jdbc_bridge_ip}:{self.jdbc_bridge_port}/ping" ) - if self.with_prometheus: + if self.with_prometheus and self.base_prometheus_cmd: os.makedirs(self.prometheus_writer_logs_dir) os.chmod(self.prometheus_writer_logs_dir, stat.S_IRWXU | stat.S_IRWXO) os.makedirs(self.prometheus_reader_logs_dir) os.chmod(self.prometheus_reader_logs_dir, stat.S_IRWXU | stat.S_IRWXO) + prometheus_start_cmd = self.base_prometheus_cmd + common_opts + + logging.info( + "Trying to create Prometheus instances by command %s", + " ".join(map(str, prometheus_start_cmd)), + ) + run_and_check(prometheus_start_cmd) + self.up_called = True + logging.info("Trying to connect to Prometheus...") + self.wait_prometheus_to_start() + clickhouse_start_cmd = self.base_cmd + ["up", "-d", "--no-recreate"] logging.debug( ( diff --git a/tests/integration/test_prometheus_protocols/test.py b/tests/integration/test_prometheus_protocols/test.py index e368c841c4e..49bc7817f02 100644 --- a/tests/integration/test_prometheus_protocols/test.py +++ b/tests/integration/test_prometheus_protocols/test.py @@ -20,7 +20,7 @@ node = cluster.add_instance( def execute_query_on_prometheus_writer(query, timestamp): return execute_query_impl( - cluster.get_instance_ip(cluster.prometheus_writer_host), + cluster.prometheus_writer_ip, cluster.prometheus_writer_port, "/api/v1/query", query, @@ -30,7 +30,7 @@ def execute_query_on_prometheus_writer(query, timestamp): def execute_query_on_prometheus_reader(query, timestamp): return execute_query_impl( - cluster.get_instance_ip(cluster.prometheus_reader_host), + cluster.prometheus_reader_ip, cluster.prometheus_reader_port, "/api/v1/query", query, From 0bdf4402fea83e1cb96b0040323c3247f5dcb0b5 Mon Sep 17 00:00:00 2001 From: "Mikhail f. Shiryaev" Date: Mon, 11 Nov 2024 15:58:38 +0100 Subject: [PATCH 634/680] Post critical errors from cherry_pick.py --- tests/ci/cherry_pick.py | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/tests/ci/cherry_pick.py b/tests/ci/cherry_pick.py index 9bdc184f661..ca32d5bc24c 100644 --- a/tests/ci/cherry_pick.py +++ b/tests/ci/cherry_pick.py @@ -34,8 +34,9 @@ from typing import List, Optional import __main__ +from ci_buddy import CIBuddy from ci_config import Labels -from env_helper import TEMP_PATH +from env_helper import IS_CI, TEMP_PATH from get_robot_token import get_best_robot_token from git_helper import GIT_PREFIX, git_runner, is_shallow from github_helper import GitHub, PullRequest, PullRequests, Repository @@ -653,6 +654,14 @@ def main(): bp.process_backports() if bp.error is not None: logging.error("Finished successfully, but errors occurred!") + if IS_CI: + ci_buddy = CIBuddy() + ci_buddy.post_job_error( + f"The cherry-pick finished with errors: {bp.error}", + with_instance_info=True, + with_wf_link=True, + critical=True, + ) raise bp.error From 40c4183ae70c720aaca797b165e3cf71aa4d8133 Mon Sep 17 00:00:00 2001 From: Sema Checherinda Date: Mon, 11 Nov 2024 17:26:28 +0100 Subject: [PATCH 635/680] fix tidy build --- src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.h | 10 +++++----- src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp | 4 ++-- src/Disks/ObjectStorages/Local/LocalObjectStorage.h | 8 ++++---- 3 files changed, 11 insertions(+), 11 deletions(-) diff --git a/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.h b/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.h index 317399b4753..7d6c914c398 100644 --- a/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.h +++ b/src/Disks/ObjectStorages/HDFS/HDFSObjectStorage.h @@ -77,11 +77,6 @@ public: size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE, const WriteSettings & write_settings = {}) override; - /// Remove file. Throws exception if file doesn't exists or it's a directory. - void removeObject(const StoredObject & object); - - void removeObjects(const StoredObjects & objects); - void removeObjectIfExists(const StoredObject & object) override; void removeObjectsIfExist(const StoredObjects & objects) override; @@ -117,6 +112,11 @@ private: void initializeHDFSFS() const; std::string extractObjectKeyFromURL(const StoredObject & object) const; + /// Remove file. Throws exception if file doesn't exists or it's a directory. + void removeObject(const StoredObject & object); + + void removeObjects(const StoredObjects & objects); + const Poco::Util::AbstractConfiguration & config; mutable HDFSBuilderWrapper hdfs_builder; diff --git a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp index 5f1b6aedc72..f24501dc60e 100644 --- a/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp +++ b/src/Disks/ObjectStorages/Local/LocalObjectStorage.cpp @@ -81,7 +81,7 @@ std::unique_ptr LocalObjectStorage::writeObject( /// NO return std::make_unique(object.remote_path, buf_size); } -void LocalObjectStorage::removeObject(const StoredObject & object) +void LocalObjectStorage::removeObject(const StoredObject & object) const { /// For local object storage files are actually removed when "metadata" is removed. if (!exists(object)) @@ -91,7 +91,7 @@ void LocalObjectStorage::removeObject(const StoredObject & object) ErrnoException::throwFromPath(ErrorCodes::CANNOT_UNLINK, object.remote_path, "Cannot unlink file {}", object.remote_path); } -void LocalObjectStorage::removeObjects(const StoredObjects & objects) +void LocalObjectStorage::removeObjects(const StoredObjects & objects) const { for (const auto & object : objects) removeObject(object); diff --git a/src/Disks/ObjectStorages/Local/LocalObjectStorage.h b/src/Disks/ObjectStorages/Local/LocalObjectStorage.h index ffc151bda04..5b3c3951364 100644 --- a/src/Disks/ObjectStorages/Local/LocalObjectStorage.h +++ b/src/Disks/ObjectStorages/Local/LocalObjectStorage.h @@ -42,10 +42,6 @@ public: size_t buf_size = DBMS_DEFAULT_BUFFER_SIZE, const WriteSettings & write_settings = {}) override; - void removeObject(const StoredObject & object); - - void removeObjects(const StoredObjects & objects); - void removeObjectIfExists(const StoredObject & object) override; void removeObjectsIfExist(const StoredObjects & objects) override; @@ -82,6 +78,10 @@ public: ReadSettings patchSettings(const ReadSettings & read_settings) const override; private: + void removeObject(const StoredObject & object) const; + + void removeObjects(const StoredObjects & objects) const; + String key_prefix; LoggerPtr log; std::string description; From 6f00b490679f9e26159105f095660f6b23ea34c2 Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Mon, 11 Nov 2024 16:41:23 +0000 Subject: [PATCH 636/680] Fixing more tests. --- .../MergeTreeSplitPrewhereIntoReadSteps.cpp | 2 +- ...filter_push_down_equivalent_sets.reference | 68 +++++++++++-------- .../0_stateless/03199_merge_filters_bug.sql | 34 +++++++++- 3 files changed, 74 insertions(+), 30 deletions(-) diff --git a/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp b/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp index 73fe2600946..2af9974c870 100644 --- a/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp +++ b/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp @@ -358,7 +358,7 @@ bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionAction /// Don't remove if it's in the list of original outputs .remove_filter_column = step.original_node && !all_outputs.contains(step.original_node) && node_to_step[step.original_node] <= step_index, - .need_filter = false, + .need_filter = true, .perform_alter_conversions = true, }; diff --git a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference index 80f4e309505..d0a3e7b02ae 100644 --- a/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference +++ b/tests/queries/0_stateless/03036_join_filter_push_down_equivalent_sets.reference @@ -163,17 +163,21 @@ Positions: 4 2 0 1 Filter (( + (JOIN actions + Change column names to column identifiers))) Header: __table1.id UInt64 __table1.value String - Filter column: and(equals(__table1.id, 5_UInt8), equals(__table1.id, 6_UInt8)) (removed) + AND column: equals(__table1.id, 5_UInt8) Actions: INPUT : 0 -> id UInt64 : 0 - INPUT : 1 -> value String : 1 + COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 1 + FUNCTION equals(id : 0, 5_UInt8 :: 1) -> equals(__table1.id, 5_UInt8) UInt8 : 2 + Positions: 2 0 2 + Filter column: and(equals(__table1.id, 5_UInt8), equals(__table1.id, 6_UInt8)) (removed) + Actions: INPUT : 2 -> value String : 0 + INPUT : 1 -> id UInt64 : 1 COLUMN Const(UInt8) -> 6_UInt8 UInt8 : 2 - COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 3 - ALIAS id : 0 -> __table1.id UInt64 : 4 - ALIAS value :: 1 -> __table1.value String : 5 - FUNCTION equals(id : 0, 6_UInt8 :: 2) -> equals(__table1.id, 6_UInt8) UInt8 : 1 - FUNCTION equals(id :: 0, 5_UInt8 :: 3) -> equals(__table1.id, 5_UInt8) UInt8 : 2 - FUNCTION and(equals(__table1.id, 5_UInt8) :: 2, equals(__table1.id, 6_UInt8) :: 1) -> and(equals(__table1.id, 5_UInt8), equals(__table1.id, 6_UInt8)) UInt8 : 3 - Positions: 3 4 5 + INPUT : 0 -> equals(__table1.id, 5_UInt8) UInt8 : 3 + ALIAS value :: 0 -> __table1.value String : 4 + ALIAS id : 1 -> __table1.id UInt64 : 0 + FUNCTION equals(id :: 1, 6_UInt8 :: 2) -> equals(__table1.id, 6_UInt8) UInt8 : 5 + FUNCTION and(equals(__table1.id, 5_UInt8) :: 3, equals(__table1.id, 6_UInt8) :: 5) -> and(equals(__table1.id, 5_UInt8), equals(__table1.id, 6_UInt8)) UInt8 : 2 + Positions: 2 0 4 ReadFromMergeTree (default.test_table_1) Header: id UInt64 value String @@ -183,17 +187,21 @@ Positions: 4 2 0 1 Filter (( + (JOIN actions + Change column names to column identifiers))) Header: __table2.id UInt64 __table2.value String - Filter column: and(equals(__table2.id, 6_UInt8), equals(__table2.id, 5_UInt8)) (removed) + AND column: equals(__table2.id, 6_UInt8) Actions: INPUT : 0 -> id UInt64 : 0 - INPUT : 1 -> value String : 1 + COLUMN Const(UInt8) -> 6_UInt8 UInt8 : 1 + FUNCTION equals(id : 0, 6_UInt8 :: 1) -> equals(__table2.id, 6_UInt8) UInt8 : 2 + Positions: 2 0 2 + Filter column: and(equals(__table2.id, 6_UInt8), equals(__table2.id, 5_UInt8)) (removed) + Actions: INPUT : 2 -> value String : 0 + INPUT : 1 -> id UInt64 : 1 COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 2 - COLUMN Const(UInt8) -> 6_UInt8 UInt8 : 3 - ALIAS id : 0 -> __table2.id UInt64 : 4 - ALIAS value :: 1 -> __table2.value String : 5 - FUNCTION equals(id : 0, 5_UInt8 :: 2) -> equals(__table2.id, 5_UInt8) UInt8 : 1 - FUNCTION equals(id :: 0, 6_UInt8 :: 3) -> equals(__table2.id, 6_UInt8) UInt8 : 2 - FUNCTION and(equals(__table2.id, 6_UInt8) :: 2, equals(__table2.id, 5_UInt8) :: 1) -> and(equals(__table2.id, 6_UInt8), equals(__table2.id, 5_UInt8)) UInt8 : 3 - Positions: 3 4 5 + INPUT : 0 -> equals(__table2.id, 6_UInt8) UInt8 : 3 + ALIAS value :: 0 -> __table2.value String : 4 + ALIAS id : 1 -> __table2.id UInt64 : 0 + FUNCTION equals(id :: 1, 5_UInt8 :: 2) -> equals(__table2.id, 5_UInt8) UInt8 : 5 + FUNCTION and(equals(__table2.id, 6_UInt8) :: 3, equals(__table2.id, 5_UInt8) :: 5) -> and(equals(__table2.id, 6_UInt8), equals(__table2.id, 5_UInt8)) UInt8 : 2 + Positions: 2 0 4 ReadFromMergeTree (default.test_table_2) Header: id UInt64 value String @@ -656,17 +664,21 @@ Positions: 4 2 0 1 __table1.value String __table2.value String __table2.id UInt64 - Filter column: and(equals(__table1.id, 5_UInt8), equals(__table2.id, 6_UInt8)) (removed) + AND column: equals(__table1.id, 5_UInt8) Actions: INPUT : 0 -> __table1.id UInt64 : 0 - INPUT :: 1 -> __table1.value String : 1 - INPUT :: 2 -> __table2.value String : 2 - INPUT : 3 -> __table2.id UInt64 : 3 - COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 4 - COLUMN Const(UInt8) -> 6_UInt8 UInt8 : 5 - FUNCTION equals(__table1.id : 0, 5_UInt8 :: 4) -> equals(__table1.id, 5_UInt8) UInt8 : 6 - FUNCTION equals(__table2.id : 3, 6_UInt8 :: 5) -> equals(__table2.id, 6_UInt8) UInt8 : 4 - FUNCTION and(equals(__table1.id, 5_UInt8) :: 6, equals(__table2.id, 6_UInt8) :: 4) -> and(equals(__table1.id, 5_UInt8), equals(__table2.id, 6_UInt8)) UInt8 : 5 - Positions: 5 0 1 2 3 + COLUMN Const(UInt8) -> 5_UInt8 UInt8 : 1 + FUNCTION equals(__table1.id : 0, 5_UInt8 :: 1) -> equals(__table1.id, 5_UInt8) UInt8 : 2 + Positions: 2 0 2 + Filter column: and(equals(__table1.id, 5_UInt8), equals(__table2.id, 6_UInt8)) (removed) + Actions: INPUT :: 1 -> __table1.id UInt64 : 0 + INPUT :: 2 -> __table1.value String : 1 + INPUT :: 3 -> __table2.value String : 2 + INPUT : 4 -> __table2.id UInt64 : 3 + COLUMN Const(UInt8) -> 6_UInt8 UInt8 : 4 + INPUT : 0 -> equals(__table1.id, 5_UInt8) UInt8 : 5 + FUNCTION equals(__table2.id : 3, 6_UInt8 :: 4) -> equals(__table2.id, 6_UInt8) UInt8 : 6 + FUNCTION and(equals(__table1.id, 5_UInt8) :: 5, equals(__table2.id, 6_UInt8) :: 6) -> and(equals(__table1.id, 5_UInt8), equals(__table2.id, 6_UInt8)) UInt8 : 4 + Positions: 4 0 1 2 3 Join (JOIN FillRightFirst) Header: __table1.id UInt64 __table1.value String diff --git a/tests/queries/0_stateless/03199_merge_filters_bug.sql b/tests/queries/0_stateless/03199_merge_filters_bug.sql index bb2a4255a3d..2023e0f1d73 100644 --- a/tests/queries/0_stateless/03199_merge_filters_bug.sql +++ b/tests/queries/0_stateless/03199_merge_filters_bug.sql @@ -51,6 +51,22 @@ tmp1 AS LEFT JOIN tmp1 USING (fs1) WHERE (fs1 IN ('test')) SETTINGS enable_multiple_prewhere_read_steps = 0, query_plan_merge_filters=0; +WITH +tmp1 AS +( + SELECT + CAST(s1, 'FixedString(10)') AS fs1, + s2 AS sector, + s3 + FROM t1 + WHERE (s3 != 'test') +) + SELECT + fs1 + FROM t2 + LEFT JOIN tmp1 USING (fs1) + WHERE (fs1 IN ('test')) SETTINGS enable_multiple_prewhere_read_steps = 1, query_plan_merge_filters=1; + optimize table t1 final; WITH @@ -67,4 +83,20 @@ tmp1 AS fs1 FROM t2 LEFT JOIN tmp1 USING (fs1) - WHERE (fs1 IN ('test')); + WHERE (fs1 IN ('test')) SETTINGS enable_multiple_prewhere_read_steps = 0, query_plan_merge_filters=0; + +WITH +tmp1 AS +( + SELECT + CAST(s1, 'FixedString(10)') AS fs1, + s2 AS sector, + s3 + FROM t1 + WHERE (s3 != 'test') +) + SELECT + fs1 + FROM t2 + LEFT JOIN tmp1 USING (fs1) + WHERE (fs1 IN ('test')) SETTINGS enable_multiple_prewhere_read_steps = 1, query_plan_merge_filters=1; From a0cc03b175b035e9c52e782811d990a619acc272 Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Mon, 11 Nov 2024 17:50:11 +0000 Subject: [PATCH 637/680] Cleanup. --- src/Processors/QueryPlan/FilterStep.cpp | 17 ++++++++ .../QueryPlan/ReadFromMergeTree.cpp | 2 + .../MergeTree/MergeTreeBlockReadUtils.cpp | 2 +- src/Storages/MergeTree/MergeTreeIOSettings.h | 2 + .../MergeTree/MergeTreeSelectProcessor.cpp | 8 ++-- .../MergeTree/MergeTreeSelectProcessor.h | 3 +- .../MergeTreeSplitPrewhereIntoReadSteps.cpp | 41 +++++++++++-------- 7 files changed, 51 insertions(+), 24 deletions(-) diff --git a/src/Processors/QueryPlan/FilterStep.cpp b/src/Processors/QueryPlan/FilterStep.cpp index a6b157cdd1d..5bf55f67208 100644 --- a/src/Processors/QueryPlan/FilterStep.cpp +++ b/src/Processors/QueryPlan/FilterStep.cpp @@ -64,6 +64,7 @@ static ActionsAndName splitSingleAndFilter(ActionsDAG & dag, const ActionsDAG::N return ActionsAndName{std::move(split_result.first), std::move(name)}; } +/// Try to split the left most AND atom to a separate DAG. static std::optional trySplitSingleAndFilter(ActionsDAG & dag, const std::string & filter_name) { const auto * filter = &dag.findInOutputs(filter_name); @@ -83,6 +84,7 @@ static std::optional trySplitSingleAndFilter(ActionsDAG & dag, c if (node->type == ActionsDAG::ActionType::FUNCTION && node->function_base->getName() == "and") { + /// The order is important. We should take the left-most atom, so put conditions on stack in reverse order. for (const auto * child : node->children | std::ranges::views::reverse) nodes.push(child); @@ -141,6 +143,8 @@ void FilterStep::transformPipeline(QueryPipelineBuilder & pipeline, const BuildQ { std::vector and_atoms; + /// Spliting AND filter condition to steps under the setting, which is enabled with merge_filters optimization. + /// This is needed to support short-circuit properly. if (settings.enable_multiple_filters_transforms_for_and_chain && !actions_dag.hasStatefulFunctions()) and_atoms = splitAndChainIntoMultipleFilters(actions_dag, filter_column_name); @@ -206,6 +210,19 @@ void FilterStep::describeActions(FormatSettings & settings) const void FilterStep::describeActions(JSONBuilder::JSONMap & map) const { + auto cloned_dag = actions_dag.clone(); + + std::vector and_atoms; + if (!actions_dag.hasStatefulFunctions()) + and_atoms = splitAndChainIntoMultipleFilters(cloned_dag, filter_column_name); + + for (auto & and_atom : and_atoms) + { + auto expression = std::make_shared(std::move(and_atom.dag)); + map.add("AND column", and_atom.name); + map.add("Expression", expression->toTree()); + } + map.add("Filter Column", filter_column_name); map.add("Removes Filter", remove_filter_column); diff --git a/src/Processors/QueryPlan/ReadFromMergeTree.cpp b/src/Processors/QueryPlan/ReadFromMergeTree.cpp index 3186df6a6b3..d144187821a 100644 --- a/src/Processors/QueryPlan/ReadFromMergeTree.cpp +++ b/src/Processors/QueryPlan/ReadFromMergeTree.cpp @@ -175,6 +175,7 @@ namespace Setting extern const SettingsBool use_skip_indexes; extern const SettingsBool use_skip_indexes_if_final; extern const SettingsBool use_uncompressed_cache; + extern const SettingsBool query_plan_merge_filters; extern const SettingsUInt64 merge_tree_min_read_task_size; } @@ -206,6 +207,7 @@ static MergeTreeReaderSettings getMergeTreeReaderSettings( .use_asynchronous_read_from_pool = settings[Setting::allow_asynchronous_read_from_io_pool_for_merge_tree] && (settings[Setting::max_streams_to_max_threads_ratio] > 1 || settings[Setting::max_streams_for_merge_tree_reading] > 1), .enable_multiple_prewhere_read_steps = settings[Setting::enable_multiple_prewhere_read_steps], + .force_shirt_circuit_execution = settings[Setting::query_plan_merge_filters] }; } diff --git a/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp b/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp index 7ba358d2d35..03a0aed80bf 100644 --- a/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp +++ b/src/Storages/MergeTree/MergeTreeBlockReadUtils.cpp @@ -330,7 +330,7 @@ MergeTreeReadTaskColumns getReadTaskColumns( auto prewhere_actions = MergeTreeSelectProcessor::getPrewhereActions( prewhere_info, actions_settings, - reader_settings.enable_multiple_prewhere_read_steps); + reader_settings.enable_multiple_prewhere_read_steps, reader_settings.force_shirt_circuit_execution); for (const auto & step : prewhere_actions.steps) add_step(*step); diff --git a/src/Storages/MergeTree/MergeTreeIOSettings.h b/src/Storages/MergeTree/MergeTreeIOSettings.h index 4d1d2533729..ecd4ad34961 100644 --- a/src/Storages/MergeTree/MergeTreeIOSettings.h +++ b/src/Storages/MergeTree/MergeTreeIOSettings.h @@ -45,6 +45,8 @@ struct MergeTreeReaderSettings bool use_asynchronous_read_from_pool = false; /// If PREWHERE has multiple conditions combined with AND, execute them in separate read/filtering steps. bool enable_multiple_prewhere_read_steps = false; + /// In case of multiple prewhere steps, execute filtering earlier to support short-circuit properly. + bool force_shirt_circuit_execution = false; /// If true, try to lower size of read buffer according to granule size and compressed block size. bool adjust_read_buffer_size = true; /// If true, it's allowed to read the whole part without reading marks. diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp index 5efd33ce09a..8beff55e698 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.cpp @@ -91,7 +91,7 @@ MergeTreeSelectProcessor::MergeTreeSelectProcessor( , algorithm(std::move(algorithm_)) , prewhere_info(prewhere_info_) , actions_settings(actions_settings_) - , prewhere_actions(getPrewhereActions(prewhere_info, actions_settings, reader_settings_.enable_multiple_prewhere_read_steps)) + , prewhere_actions(getPrewhereActions(prewhere_info, actions_settings, reader_settings_.enable_multiple_prewhere_read_steps, reader_settings_.force_shirt_circuit_execution)) , reader_settings(reader_settings_) , result_header(transformHeader(pool->getHeader(), prewhere_info)) { @@ -124,9 +124,9 @@ String MergeTreeSelectProcessor::getName() const return fmt::format("MergeTreeSelect(pool: {}, algorithm: {})", pool->getName(), algorithm->getName()); } -bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionActionsSettings & actions_settings, PrewhereExprInfo & prewhere); +bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionActionsSettings & actions_settings, PrewhereExprInfo & prewhere, bool force_shirt_circuit_execution); -PrewhereExprInfo MergeTreeSelectProcessor::getPrewhereActions(PrewhereInfoPtr prewhere_info, const ExpressionActionsSettings & actions_settings, bool enable_multiple_prewhere_read_steps) +PrewhereExprInfo MergeTreeSelectProcessor::getPrewhereActions(PrewhereInfoPtr prewhere_info, const ExpressionActionsSettings & actions_settings, bool enable_multiple_prewhere_read_steps, bool force_shirt_circuit_execution) { PrewhereExprInfo prewhere_actions; if (prewhere_info) @@ -147,7 +147,7 @@ PrewhereExprInfo MergeTreeSelectProcessor::getPrewhereActions(PrewhereInfoPtr pr } if (!enable_multiple_prewhere_read_steps || - !tryBuildPrewhereSteps(prewhere_info, actions_settings, prewhere_actions)) + !tryBuildPrewhereSteps(prewhere_info, actions_settings, prewhere_actions, force_shirt_circuit_execution)) { PrewhereExprStep prewhere_step { diff --git a/src/Storages/MergeTree/MergeTreeSelectProcessor.h b/src/Storages/MergeTree/MergeTreeSelectProcessor.h index 33069a78e33..afd88116e15 100644 --- a/src/Storages/MergeTree/MergeTreeSelectProcessor.h +++ b/src/Storages/MergeTree/MergeTreeSelectProcessor.h @@ -73,7 +73,8 @@ public: static PrewhereExprInfo getPrewhereActions( PrewhereInfoPtr prewhere_info, const ExpressionActionsSettings & actions_settings, - bool enable_multiple_prewhere_read_steps); + bool enable_multiple_prewhere_read_steps, + bool force_shirt_circuit_execution); void addPartLevelToChunk(bool add_part_level_) { add_part_level = add_part_level_; } diff --git a/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp b/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp index 2af9974c870..c35e356bf18 100644 --- a/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp +++ b/src/Storages/MergeTree/MergeTreeSplitPrewhereIntoReadSteps.cpp @@ -50,6 +50,17 @@ void fillRequiredColumns(const ActionsDAG::Node * node, std::unordered_map DAGNodeRef +/// ResultNode -> DAGNodeRef using OriginalToNewNodeMap = std::unordered_map; using NodeNameToLastUsedStepMap = std::unordered_map; @@ -70,7 +81,6 @@ const ActionsDAG::Node & addClonedDAGToDAG( OriginalToNewNodeMap & node_remap, NodeNameToLastUsedStepMap & node_to_step_map) { - //const String & node_name = original_dag_node->result_name; /// Look for the node in the map of already known nodes if (node_remap.contains(original_dag_node)) { @@ -82,9 +92,11 @@ const ActionsDAG::Node & addClonedDAGToDAG( /// If the node is known from the previous steps, add it as an input, except for constants if (original_dag_node->type != ActionsDAG::ActionType::COLUMN) { - node_ref.dag->addOrReplaceInOutputs(*node_ref.node); + // addToOutputsIfNotAlreadyAdded(*node_ref.dag, node_ref.node); + node_ref.dag->getOutputs().push_back(node_ref.node); + const auto & new_node = new_dag->addInput(node_ref.node->result_name, node_ref.node->result_type); - node_remap[original_dag_node] = {new_dag.get(), &new_node}; /// TODO: here we update the node reference. Is it always correct? + node_remap[original_dag_node] = {new_dag.get(), &new_node}; /// Remember the index of the last step which reuses this node. /// We cannot remove this node from the outputs before that step. @@ -207,7 +219,11 @@ const ActionsDAG::Node & addAndTrue( /// 6. Find all outputs of the original DAG /// 7. Find all outputs that were computed in the already built DAGs, mark these nodes as outputs in the steps where they were computed /// 8. Add computation of the remaining outputs to the last step with the procedure similar to 4 -bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionActionsSettings & actions_settings, PrewhereExprInfo & prewhere) +bool tryBuildPrewhereSteps( + PrewhereInfoPtr prewhere_info, + const ExpressionActionsSettings & actions_settings, + PrewhereExprInfo & prewhere, + bool force_shirt_circuit_execution) { if (!prewhere_info) return true; @@ -275,26 +291,16 @@ bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionAction /// Add AND function to combine the conditions FunctionOverloadResolverPtr func_builder_and = std::make_unique(std::make_shared()); const auto & and_function_node = addFunction(step_dag, func_builder_and, new_condition_nodes); //, node_remap); - //step_dag->addOrReplaceInOutputs(and_function_node); result_node = &and_function_node; } else { result_node = new_condition_nodes.front(); /// Check if explicit cast is needed for the condition to serve as a filter. - //const auto result_type_name = result_node->result_type->getName(); - if (isUInt8(removeNullable(removeLowCardinality(result_node->result_type)))) - { - /// No need to cast - //step_dag->addOrReplaceInOutputs(result_node); - //result_name = result_node.result_name; - } - else + if (!isUInt8(removeNullable(removeLowCardinality(result_node->result_type)))) { /// Build "condition AND True" expression to "cast" the condition to UInt8 or Nullable(UInt8) depending on its type. result_node = &addAndTrue(step_dag, *result_node); //, node_remap); - //step_dag->addOrReplaceInOutputs(cast_node); - //result_name = &cast_node.result_name; } } @@ -334,7 +340,6 @@ bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionAction const auto & cast_node = addCast(last_step_dag, and_node, output->result_type); //, node_remap); /// Add alias for the result with the name of the PREWHERE column const auto & prewhere_result_node = last_step_dag->addAlias(cast_node, output->result_name); - //last_step_dag->addOrReplaceInOutputs(prewhere_result_node); last_step_dag->getOutputs().push_back(&prewhere_result_node); steps.back().result_node = &prewhere_result_node; } @@ -358,7 +363,7 @@ bool tryBuildPrewhereSteps(PrewhereInfoPtr prewhere_info, const ExpressionAction /// Don't remove if it's in the list of original outputs .remove_filter_column = step.original_node && !all_outputs.contains(step.original_node) && node_to_step[step.original_node] <= step_index, - .need_filter = true, + .need_filter = force_shirt_circuit_execution, .perform_alter_conversions = true, }; From 92114f3c749bb78811ece644123b3d81e011e56f Mon Sep 17 00:00:00 2001 From: Nikolai Kochetov Date: Mon, 11 Nov 2024 18:01:24 +0000 Subject: [PATCH 638/680] Fixing typos. --- src/Processors/QueryPlan/FilterStep.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/Processors/QueryPlan/FilterStep.cpp b/src/Processors/QueryPlan/FilterStep.cpp index 5bf55f67208..af9e3f0c515 100644 --- a/src/Processors/QueryPlan/FilterStep.cpp +++ b/src/Processors/QueryPlan/FilterStep.cpp @@ -143,7 +143,7 @@ void FilterStep::transformPipeline(QueryPipelineBuilder & pipeline, const BuildQ { std::vector and_atoms; - /// Spliting AND filter condition to steps under the setting, which is enabled with merge_filters optimization. + /// Splitting AND filter condition to steps under the setting, which is enabled with merge_filters optimization. /// This is needed to support short-circuit properly. if (settings.enable_multiple_filters_transforms_for_and_chain && !actions_dag.hasStatefulFunctions()) and_atoms = splitAndChainIntoMultipleFilters(actions_dag, filter_column_name); From 621cb60446cb17f0366f49b86c3432eed5db3716 Mon Sep 17 00:00:00 2001 From: Peter Nguyen Date: Mon, 11 Nov 2024 11:12:01 -0800 Subject: [PATCH 639/680] Fix 'was was' typo in sql-reference/statements/alter/column.md --- docs/en/sql-reference/statements/alter/column.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/en/sql-reference/statements/alter/column.md b/docs/en/sql-reference/statements/alter/column.md index 29df041ccc6..fb16dacb7c8 100644 --- a/docs/en/sql-reference/statements/alter/column.md +++ b/docs/en/sql-reference/statements/alter/column.md @@ -279,7 +279,7 @@ For columns with a new or updated `MATERIALIZED` value expression, all existing For columns with a new or updated `DEFAULT` value expression, the behavior depends on the ClickHouse version: - In ClickHouse < v24.2, all existing rows are rewritten. -- ClickHouse >= v24.2 distinguishes if a row value in a column with `DEFAULT` value expression was explicitly specified when it was inserted, or not, i.e. calculated from the `DEFAULT` value expression. If the value was explicitly specified, ClickHouse keeps it as is. If the value was was calculated, ClickHouse changes it to the new or updated `MATERIALIZED` value expression. +- ClickHouse >= v24.2 distinguishes if a row value in a column with `DEFAULT` value expression was explicitly specified when it was inserted, or not, i.e. calculated from the `DEFAULT` value expression. If the value was explicitly specified, ClickHouse keeps it as is. If the value was calculated, ClickHouse changes it to the new or updated `MATERIALIZED` value expression. Syntax: From b05d3ed6df35b2e66c81bc8d7b9077a82758dcf1 Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Mon, 11 Nov 2024 22:43:03 +0100 Subject: [PATCH 640/680] impl --- src/Analyzer/Resolve/QueryAnalyzer.cpp | 3 + .../ExecuteScalarSubqueriesVisitor.cpp | 7 ++- src/Interpreters/PreparedSets.cpp | 19 +++--- src/Interpreters/ProcessorsProfileLog.cpp | 62 ++++++++++++++++++- src/Interpreters/ProcessorsProfileLog.h | 1 + src/Interpreters/executeQuery.cpp | 48 +------------- 6 files changed, 80 insertions(+), 60 deletions(-) diff --git a/src/Analyzer/Resolve/QueryAnalyzer.cpp b/src/Analyzer/Resolve/QueryAnalyzer.cpp index 390418494e7..03ebd893c47 100644 --- a/src/Analyzer/Resolve/QueryAnalyzer.cpp +++ b/src/Analyzer/Resolve/QueryAnalyzer.cpp @@ -1,3 +1,4 @@ +#include #include #include @@ -676,6 +677,8 @@ void QueryAnalyzer::evaluateScalarSubqueryIfNeeded(QueryTreeNodePtr & node, Iden "tuple"}); } } + + logProcessorProfile(context, io.pipeline.getProcessors()); } scalars_cache.emplace(node_with_hash, scalar_block); diff --git a/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp b/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp index d4da038c089..9ae2ffc208d 100644 --- a/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp +++ b/src/Interpreters/ExecuteScalarSubqueriesVisitor.cpp @@ -5,9 +5,11 @@ #include #include #include +#include #include #include #include +#include #include #include #include @@ -19,9 +21,8 @@ #include #include #include -#include #include -#include +#include namespace ProfileEvents { @@ -246,6 +247,8 @@ void ExecuteScalarSubqueriesMatcher::visit(const ASTSubquery & subquery, ASTPtr if (tmp_block.rows() != 0) throw Exception(ErrorCodes::INCORRECT_RESULT_OF_SCALAR_SUBQUERY, "Scalar subquery returned more than one row"); + + logProcessorProfile(data.getContext(), io.pipeline.getProcessors()); } block = materializeBlock(block); diff --git a/src/Interpreters/PreparedSets.cpp b/src/Interpreters/PreparedSets.cpp index 538108165fb..c69e2f84d42 100644 --- a/src/Interpreters/PreparedSets.cpp +++ b/src/Interpreters/PreparedSets.cpp @@ -1,21 +1,22 @@ #include #include -#include -#include -#include -#include -#include +#include #include -#include +#include +#include +#include +#include +#include #include #include +#include #include +#include #include #include -#include -#include #include #include +#include namespace DB { @@ -239,6 +240,8 @@ SetPtr FutureSetFromSubquery::buildOrderedSetInplace(const ContextPtr & context) if (!set_and_key->set->isCreated()) return nullptr; + logProcessorProfile(context, pipeline.getProcessors()); + return set_and_key->set; } diff --git a/src/Interpreters/ProcessorsProfileLog.cpp b/src/Interpreters/ProcessorsProfileLog.cpp index 8a646b5d0e7..d7811e5e9e2 100644 --- a/src/Interpreters/ProcessorsProfileLog.cpp +++ b/src/Interpreters/ProcessorsProfileLog.cpp @@ -1,5 +1,6 @@ #include +#include #include #include #include @@ -8,16 +9,19 @@ #include #include #include +#include #include #include #include -#include - -#include namespace DB { +namespace Setting +{ +extern const SettingsBool log_processors_profiles; +} + ColumnsDescription ProcessorProfileLogElement::getColumnsDescription() { return ColumnsDescription @@ -81,5 +85,57 @@ void ProcessorProfileLogElement::appendToBlock(MutableColumns & columns) const columns[i++]->insert(output_bytes); } +void logProcessorProfile(ContextPtr context, const Processors & processors) +{ + const Settings & settings = context->getSettingsRef(); + if (settings[Setting::log_processors_profiles]) + { + if (auto processors_profile_log = context->getProcessorsProfileLog()) + { + ProcessorProfileLogElement processor_elem; + const auto time_now = std::chrono::system_clock::now(); + processor_elem.event_time = timeInSeconds(time_now); + processor_elem.event_time_microseconds = timeInMicroseconds(time_now); + processor_elem.initial_query_id = context->getInitialQueryId(); + processor_elem.query_id = context->getCurrentQueryId(); + + auto get_proc_id = [](const IProcessor & proc) -> UInt64 { return reinterpret_cast(&proc); }; + + for (const auto & processor : processors) + { + std::vector parents; + for (const auto & port : processor->getOutputs()) + { + if (!port.isConnected()) + continue; + const IProcessor & next = port.getInputPort().getProcessor(); + parents.push_back(get_proc_id(next)); + } + + processor_elem.id = get_proc_id(*processor); + processor_elem.parent_ids = std::move(parents); + + processor_elem.plan_step = reinterpret_cast(processor->getQueryPlanStep()); + processor_elem.plan_step_name = processor->getPlanStepName(); + processor_elem.plan_step_description = processor->getPlanStepDescription(); + processor_elem.plan_group = processor->getQueryPlanStepGroup(); + + processor_elem.processor_name = processor->getName(); + + processor_elem.elapsed_us = static_cast(processor->getElapsedNs() / 1000U); + processor_elem.input_wait_elapsed_us = static_cast(processor->getInputWaitElapsedNs() / 1000U); + processor_elem.output_wait_elapsed_us = static_cast(processor->getOutputWaitElapsedNs() / 1000U); + + auto stats = processor->getProcessorDataStats(); + processor_elem.input_rows = stats.input_rows; + processor_elem.input_bytes = stats.input_bytes; + processor_elem.output_rows = stats.output_rows; + processor_elem.output_bytes = stats.output_bytes; + + processors_profile_log->add(processor_elem); + } + } + } +} } diff --git a/src/Interpreters/ProcessorsProfileLog.h b/src/Interpreters/ProcessorsProfileLog.h index fbf52f45f56..9cc2ab6c7f0 100644 --- a/src/Interpreters/ProcessorsProfileLog.h +++ b/src/Interpreters/ProcessorsProfileLog.h @@ -50,4 +50,5 @@ public: using SystemLog::SystemLog; }; +void logProcessorProfile(ContextPtr context, const Processors & processors); } diff --git a/src/Interpreters/executeQuery.cpp b/src/Interpreters/executeQuery.cpp index 9250c069283..fa28fa04ab1 100644 --- a/src/Interpreters/executeQuery.cpp +++ b/src/Interpreters/executeQuery.cpp @@ -117,7 +117,6 @@ namespace Setting extern const SettingsOverflowMode join_overflow_mode; extern const SettingsString log_comment; extern const SettingsBool log_formatted_queries; - extern const SettingsBool log_processors_profiles; extern const SettingsBool log_profile_events; extern const SettingsUInt64 log_queries_cut_to_length; extern const SettingsBool log_queries; @@ -551,53 +550,8 @@ void logQueryFinish( if (auto query_log = context->getQueryLog()) query_log->add(elem); } - if (settings[Setting::log_processors_profiles]) - { - if (auto processors_profile_log = context->getProcessorsProfileLog()) - { - ProcessorProfileLogElement processor_elem; - processor_elem.event_time = elem.event_time; - processor_elem.event_time_microseconds = elem.event_time_microseconds; - processor_elem.initial_query_id = elem.client_info.initial_query_id; - processor_elem.query_id = elem.client_info.current_query_id; - auto get_proc_id = [](const IProcessor & proc) -> UInt64 { return reinterpret_cast(&proc); }; - - for (const auto & processor : query_pipeline.getProcessors()) - { - std::vector parents; - for (const auto & port : processor->getOutputs()) - { - if (!port.isConnected()) - continue; - const IProcessor & next = port.getInputPort().getProcessor(); - parents.push_back(get_proc_id(next)); - } - - processor_elem.id = get_proc_id(*processor); - processor_elem.parent_ids = std::move(parents); - - processor_elem.plan_step = reinterpret_cast(processor->getQueryPlanStep()); - processor_elem.plan_step_name = processor->getPlanStepName(); - processor_elem.plan_step_description = processor->getPlanStepDescription(); - processor_elem.plan_group = processor->getQueryPlanStepGroup(); - - processor_elem.processor_name = processor->getName(); - - processor_elem.elapsed_us = static_cast(processor->getElapsedNs() / 1000U); - processor_elem.input_wait_elapsed_us = static_cast(processor->getInputWaitElapsedNs() / 1000U); - processor_elem.output_wait_elapsed_us = static_cast(processor->getOutputWaitElapsedNs() / 1000U); - - auto stats = processor->getProcessorDataStats(); - processor_elem.input_rows = stats.input_rows; - processor_elem.input_bytes = stats.input_bytes; - processor_elem.output_rows = stats.output_rows; - processor_elem.output_bytes = stats.output_bytes; - - processors_profile_log->add(processor_elem); - } - } - } + logProcessorProfile(context, query_pipeline.getProcessors()); logQueryMetricLogFinish(context, internal, elem.client_info.current_query_id, time_now, std::make_shared(info)); } From bd71442ea26a5263b56e6774c6938fcb24dea432 Mon Sep 17 00:00:00 2001 From: Nikita Taranov Date: Mon, 11 Nov 2024 22:45:39 +0100 Subject: [PATCH 641/680] add test --- .../03270_processors_profile_log_3.reference | 2 + .../03270_processors_profile_log_3.sh | 96 +++++++++++++++++++ 2 files changed, 98 insertions(+) create mode 100644 tests/queries/0_stateless/03270_processors_profile_log_3.reference create mode 100755 tests/queries/0_stateless/03270_processors_profile_log_3.sh diff --git a/tests/queries/0_stateless/03270_processors_profile_log_3.reference b/tests/queries/0_stateless/03270_processors_profile_log_3.reference new file mode 100644 index 00000000000..6ed281c757a --- /dev/null +++ b/tests/queries/0_stateless/03270_processors_profile_log_3.reference @@ -0,0 +1,2 @@ +1 +1 diff --git a/tests/queries/0_stateless/03270_processors_profile_log_3.sh b/tests/queries/0_stateless/03270_processors_profile_log_3.sh new file mode 100755 index 00000000000..eb86a9f6352 --- /dev/null +++ b/tests/queries/0_stateless/03270_processors_profile_log_3.sh @@ -0,0 +1,96 @@ +#!/usr/bin/env bash + +set -e + +CUR_DIR=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd) +# shellcheck source=../shell_config.sh +. "$CUR_DIR"/../shell_config.sh + + +$CLICKHOUSE_CLIENT -q " + CREATE TABLE t + ( + a UInt32, + b UInt32 + ) + ENGINE = MergeTree + ORDER BY (a, b); + + INSERT INTO t SELECT number, number FROM numbers(1000); +" + +query_id="03270_processors_profile_log_3_$RANDOM" + +$CLICKHOUSE_CLIENT --query_id="$query_id" -q " + SET log_processors_profiles = 1; + + WITH + t0 AS + ( + SELECT * + FROM numbers(1000) + ), + t1 AS + ( + SELECT number * 3 AS b + FROM t0 + ) + SELECT b * 3 + FROM t + WHERE a IN (t1) + FORMAT Null; +" + +$CLICKHOUSE_CLIENT --query_id="$query_id" -q " + SYSTEM FLUSH LOGS; + + SELECT sum(elapsed_us) > 0 + FROM system.processors_profile_log + WHERE event_date >= yesterday() AND query_id = '$query_id' AND name = 'CreatingSetsTransform'; +" + +##################################################################### + +$CLICKHOUSE_CLIENT -q " + CREATE TABLE t1 + ( + st FixedString(54) + ) + ENGINE = MergeTree + ORDER BY tuple(); + + INSERT INTO t1 VALUES + ('abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRTUVWXYZ'), + ('\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0'), + ('IIIIIIIIII\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0'); +" + +query_id="03270_processors_profile_log_3_$RANDOM" + +$CLICKHOUSE_CLIENT --query_id="$query_id" -q " + SET log_processors_profiles = 1; + SET max_threads=2; -- no merging when max_threads=1 + + WITH + ( + SELECT groupConcat(',')(st) + FROM t1 + ORDER BY ALL + ) AS a, + ( + SELECT groupConcat(',')(CAST(st, 'String')) + FROM t1 + ORDER BY ALL + ) AS b + SELECT a = b + FORMAT Null; +" + +$CLICKHOUSE_CLIENT --query_id="$query_id" -q " + SYSTEM FLUSH LOGS; + + SELECT sum(elapsed_us) > 0 + FROM system.processors_profile_log + WHERE event_date >= yesterday() AND query_id = '$query_id' AND name = 'MergingSortedTransform'; +" + From ec27bd2e51cce45c1c199b2f88ddea31b1e20839 Mon Sep 17 00:00:00 2001 From: Alexey Milovidov Date: Tue, 12 Nov 2024 01:23:01 +0100 Subject: [PATCH 642/680] Remove ridiculous code bloat --- .../AggregateFunctionDeltaSumTimestamp.cpp | 69 ++++++++++++++---- src/AggregateFunctions/Helpers.h | 70 +------------------ 2 files changed, 58 insertions(+), 81 deletions(-) diff --git a/src/AggregateFunctions/AggregateFunctionDeltaSumTimestamp.cpp b/src/AggregateFunctions/AggregateFunctionDeltaSumTimestamp.cpp index 5819c533fd9..79f0f2b328e 100644 --- a/src/AggregateFunctions/AggregateFunctionDeltaSumTimestamp.cpp +++ b/src/AggregateFunctions/AggregateFunctionDeltaSumTimestamp.cpp @@ -22,6 +22,13 @@ namespace ErrorCodes namespace { +/** Due to a lack of proper code review, this code was contributed with a multiplication of template instantiations + * over all pairs of data types, and we deeply regret that. + * + * We cannot remove all combinations, because the binary representation of serialized data has to remain the same, + * but we can partially heal the wound by treating unsigned and signed data types in the same way. + */ + template struct AggregationFunctionDeltaSumTimestampData { @@ -37,23 +44,22 @@ template class AggregationFunctionDeltaSumTimestamp final : public IAggregateFunctionDataHelper< AggregationFunctionDeltaSumTimestampData, - AggregationFunctionDeltaSumTimestamp - > + AggregationFunctionDeltaSumTimestamp> { public: AggregationFunctionDeltaSumTimestamp(const DataTypes & arguments, const Array & params) : IAggregateFunctionDataHelper< AggregationFunctionDeltaSumTimestampData, - AggregationFunctionDeltaSumTimestamp - >{arguments, params, createResultType()} - {} + AggregationFunctionDeltaSumTimestamp>{arguments, params, createResultType()} + { + } AggregationFunctionDeltaSumTimestamp() : IAggregateFunctionDataHelper< AggregationFunctionDeltaSumTimestampData, - AggregationFunctionDeltaSumTimestamp - >{} - {} + AggregationFunctionDeltaSumTimestamp>{} + { + } bool allocatesMemoryInArena() const override { return false; } @@ -63,8 +69,8 @@ public: void NO_SANITIZE_UNDEFINED ALWAYS_INLINE add(AggregateDataPtr __restrict place, const IColumn ** columns, size_t row_num, Arena *) const override { - auto value = assert_cast &>(*columns[0]).getData()[row_num]; - auto ts = assert_cast &>(*columns[1]).getData()[row_num]; + auto value = unalignedLoad(columns[0]->getRawData().data() + row_num * sizeof(ValueType)); + auto ts = unalignedLoad(columns[1]->getRawData().data() + row_num * sizeof(TimestampType)); auto & data = this->data(place); @@ -172,10 +178,49 @@ public: void insertResultInto(AggregateDataPtr __restrict place, IColumn & to, Arena *) const override { - assert_cast &>(to).getData().push_back(this->data(place).sum); + static_cast(to).template insertRawData( + reinterpret_cast(&this->data(place).sum)); } }; + + +template class AggregateFunctionTemplate, typename... TArgs> +static IAggregateFunction * createWithTwoTypesSecond(const IDataType & second_type, TArgs && ... args) +{ + WhichDataType which(second_type); + + if (which.idx == TypeIndex::UInt32) return new AggregateFunctionTemplate(args...); + if (which.idx == TypeIndex::UInt64) return new AggregateFunctionTemplate(args...); + if (which.idx == TypeIndex::Int32) return new AggregateFunctionTemplate(args...); + if (which.idx == TypeIndex::Int64) return new AggregateFunctionTemplate(args...); + if (which.idx == TypeIndex::Float32) return new AggregateFunctionTemplate(args...); + if (which.idx == TypeIndex::Float64) return new AggregateFunctionTemplate(args...); + if (which.idx == TypeIndex::Date) return new AggregateFunctionTemplate(args...); + if (which.idx == TypeIndex::DateTime) return new AggregateFunctionTemplate(args...); + + return nullptr; +} + +template