diff options
author | Alexander Smirnov <alex@ydb.tech> | 2025-04-18 04:38:27 +0000 |
---|---|---|
committer | Alexander Smirnov <alex@ydb.tech> | 2025-04-18 04:38:27 +0000 |
commit | 7ffed5509ed1a1b5633adacf4368ef034e48660e (patch) | |
tree | 2e742ad254cf199d620fdd75cb984f4115165b3f /yql | |
parent | 332dc5ddfae67b6af527d54cd25a1971d8d78147 (diff) | |
parent | 5d15e3b0913f6b537a7940a584fe1347e8551c93 (diff) | |
download | ydb-7ffed5509ed1a1b5633adacf4368ef034e48660e.tar.gz |
Merge pull request #17385 from ydb-platform/merge-libs-250418-0050
Diffstat (limited to 'yql')
37 files changed, 699 insertions, 57 deletions
diff --git a/yql/essentials/core/common_opt/yql_co_flow2.cpp b/yql/essentials/core/common_opt/yql_co_flow2.cpp index 7d521b2ac47..8ebb5243ffa 100644 --- a/yql/essentials/core/common_opt/yql_co_flow2.cpp +++ b/yql/essentials/core/common_opt/yql_co_flow2.cpp @@ -2027,7 +2027,100 @@ void RegisterCoFlowCallables2(TCallableOptimizerMap& map) { return ret; } - return node; + // Add PruneKeys to EquiJoin + static const char optName[] = "EmitPruneKeys"; + if (!IsOptimizerEnabled<optName>(*optCtx.Types) || IsOptimizerDisabled<optName>(*optCtx.Types)) { + return node; + } + auto equiJoin = TCoEquiJoin(node); + if (HasSetting(equiJoin.Arg(equiJoin.ArgCount() - 1).Ref(), "prune_keys_added")) { + return node; + } + + THashMap<TStringBuf, THashSet<TStringBuf>> columnsForPruneKeysExtractor; + GetPruneKeysColumnsForJoinLeaves(equiJoin.Arg(equiJoin.ArgCount() - 2).Cast<TCoEquiJoinTuple>(), columnsForPruneKeysExtractor); + + TExprNode::TListType children; + bool hasChanges = false; + for (size_t i = 0; i + 2 < equiJoin.ArgCount(); ++i) { + auto child = equiJoin.Arg(i).Cast<TCoEquiJoinInput>(); + auto list = child.List(); + auto scope = child.Scope(); + + if (!scope.Ref().IsAtom()) { + children.push_back(equiJoin.Arg(i).Ptr()); + continue; + } + + auto itemNames = columnsForPruneKeysExtractor.find(scope.Ref().Content()); + if (itemNames == columnsForPruneKeysExtractor.end() || itemNames->second.empty()) { + children.push_back(equiJoin.Arg(i).Ptr()); + continue; + } + + if (auto distinct = list.Ref().GetConstraint<TDistinctConstraintNode>()) { + if (distinct->ContainsCompleteSet(std::vector<std::string_view>(itemNames->second.cbegin(), itemNames->second.cend()))) { + children.push_back(equiJoin.Arg(i).Ptr()); + continue; + } + } + + bool isOrdered = false; + if (auto sorted = list.Ref().GetConstraint<TSortedConstraintNode>()) { + for (const auto& item : sorted->GetContent()) { + size_t foundItemNamesCount = 0; + for (const auto& path : item.first) { + if (itemNames->second.contains(path.front())) { + foundItemNamesCount++; + } + } + if (foundItemNamesCount == itemNames->second.size()) { + isOrdered = true; + break; + } + } + } + + auto pruneKeysCallable = isOrdered ? "PruneAdjacentKeys" : "PruneKeys"; + YQL_CLOG(DEBUG, Core) << "Add " << pruneKeysCallable << " to EquiJoin input #" << i << ", label " << scope.Ref().Content(); + children.push_back(ctx.Builder(child.Pos()) + .List() + .Callable(0, pruneKeysCallable) + .Add(0, list.Ptr()) + .Lambda(1) + .Param("item") + .List(0) + .Do([&](TExprNodeBuilder& parent) -> TExprNodeBuilder & { + ui32 i = 0; + for (const auto& column : itemNames->second) { + parent.Callable(i++, "Member") + .Arg(0, "item") + .Atom(1, column) + .Seal(); + } + return parent; + }) + .Seal() + .Seal() + .Seal() + .Add(1, scope.Ptr()) + .Seal() + .Build()); + hasChanges = true; + } + + if (!hasChanges) { + return node; + } + + children.push_back(equiJoin.Arg(equiJoin.ArgCount() - 2).Ptr()); + children.push_back(AddSetting( + equiJoin.Arg(equiJoin.ArgCount() - 1).Ref(), + equiJoin.Arg(equiJoin.ArgCount() - 1).Pos(), + "prune_keys_added", + nullptr, + ctx)); + return ctx.ChangeChildren(*node, std::move(children)); }; map["ExtractMembers"] = [](const TExprNode::TPtr& node, TExprContext& ctx, TOptimizeContext& optCtx) { diff --git a/yql/essentials/core/peephole_opt/yql_opt_peephole_physical.cpp b/yql/essentials/core/peephole_opt/yql_opt_peephole_physical.cpp index 5755bf1dbf5..28ee1ead8f5 100644 --- a/yql/essentials/core/peephole_opt/yql_opt_peephole_physical.cpp +++ b/yql/essentials/core/peephole_opt/yql_opt_peephole_physical.cpp @@ -2760,19 +2760,11 @@ TExprNode::TPtr ExpandListHas(const TExprNode::TPtr& input, TExprContext& ctx) { return RewriteSearchByKeyForTypesMismatch<true, true>(input, ctx); } -TExprNode::TPtr ExpandPruneAdjacentKeys(const TExprNode::TPtr& input, TExprContext& ctx) { - const auto type = input->Head().GetTypeAnn(); - const auto& keyExtractorLambda = input->ChildRef(1); - - YQL_ENSURE(type->GetKind() == ETypeAnnotationKind::List || type->GetKind() == ETypeAnnotationKind::Stream); - - const auto elemType = type->GetKind() == ETypeAnnotationKind::List - ? type->Cast<TListExprType>()->GetItemType() - : type->Cast<TStreamExprType>()->GetItemType(); - const auto optionalElemType = *ctx.MakeType<TOptionalExprType>(elemType); +TExprNode::TPtr ExpandPruneAdjacentKeys(const TExprNode::TPtr& input, TExprContext& ctx, TTypeAnnotationContext& /*typesCtx*/) { + const auto& keyExtractorLambda = input->ChildPtr(1); YQL_CLOG(DEBUG, CorePeepHole) << "Expand " << input->Content(); - return ctx.Builder(input->Pos()) + return KeepConstraints(ctx.Builder(input->Pos()) .Callable("OrderedFlatMap") .Callable(0, "Fold1Map") .Add(0, input->HeadPtr()) @@ -2799,7 +2791,11 @@ TExprNode::TPtr ExpandPruneAdjacentKeys(const TExprNode::TPtr& input, TExprConte .Seal() .Seal() .Callable(1, "Nothing") - .Add(0, ExpandType(input->Pos(), optionalElemType, ctx)) + .Callable(0, "OptionalType") + .Callable(0, "TypeOf") + .Arg(0, "item") + .Seal() + .Seal() .Seal() .Callable(2, "Just") .Arg(0, "item") @@ -2814,13 +2810,16 @@ TExprNode::TPtr ExpandPruneAdjacentKeys(const TExprNode::TPtr& input, TExprConte .Arg(0, "item") .Seal() .Seal() - .Build(); + .Build(), *input, ctx); } -TExprNode::TPtr ExpandPruneKeys(const TExprNode::TPtr& input, TExprContext& ctx) { +TExprNode::TPtr ExpandPruneKeys(const TExprNode::TPtr& input, TExprContext& ctx, TTypeAnnotationContext& typesCtx) { const auto type = input->Head().GetTypeAnn(); - const auto& keyExtractorLambda = input->ChildRef(1); - YQL_ENSURE(type->GetKind() == ETypeAnnotationKind::List || type->GetKind() == ETypeAnnotationKind::Stream); + auto keyExtractorLambda = input->ChildPtr(1); + + YQL_ENSURE(type->GetKind() == ETypeAnnotationKind::Flow + || type->GetKind() == ETypeAnnotationKind::List + || type->GetKind() == ETypeAnnotationKind::Stream); auto initHandler = ctx.Builder(input->Pos()) .Lambda() @@ -2865,6 +2864,39 @@ TExprNode::TPtr ExpandPruneKeys(const TExprNode::TPtr& input, TExprContext& ctx) .Seal() .Build(); } else { + // Slight copy of GetDictionaryKeyTypes to check if keyExtractorLambda result type is complicated + // mkql CombineCore supports only simple types; for others we should add pickling + bool keyExtractorLambdaShouldBePickled = false; + auto itemType = keyExtractorLambda->GetTypeAnn(); + if (itemType->GetKind() == ETypeAnnotationKind::Optional) { + itemType = itemType->Cast<TOptionalExprType>()->GetItemType(); + } + + if (itemType->GetKind() == ETypeAnnotationKind::Tuple) { + auto tuple = itemType->Cast<TTupleExprType>(); + for (const auto& item : tuple->GetItems()) { + if (!IsDataOrOptionalOfData(item)) { + keyExtractorLambdaShouldBePickled = true; + break; + } + } + } else if (itemType->GetKind() != ETypeAnnotationKind::Data) { + keyExtractorLambdaShouldBePickled = true; + } + + if (keyExtractorLambdaShouldBePickled) { + keyExtractorLambda = ctx.Builder(input->Pos()) + .Lambda() + .Param("item") + .Callable(0, "StablePickle") + .Apply(0, keyExtractorLambda) + .With(0, "item") + .Seal() + .Seal() + .Seal() + .Build(); + } + return ctx.Builder(input->Pos()) .Callable("CombineCore") .Add(0, input->HeadPtr()) @@ -2872,6 +2904,7 @@ TExprNode::TPtr ExpandPruneKeys(const TExprNode::TPtr& input, TExprContext& ctx) .Add(2, initHandler) .Add(3, updateHandler) .Add(4, finishHandler) + .Atom(5, ToString(typesCtx.PruneKeysMemLimit)) .Seal() .Build(); } @@ -8906,8 +8939,6 @@ struct TPeepHoleRules { {"CheckedDiv", &ExpandCheckedDiv}, {"CheckedMod", &ExpandCheckedMod}, {"CheckedMinus", &ExpandCheckedMinus}, - {"PruneAdjacentKeys", &ExpandPruneAdjacentKeys}, - {"PruneKeys", &ExpandPruneKeys}, {"JsonValue", &ExpandJsonValue}, {"JsonExists", &ExpandJsonExists}, {"EmptyIterator", &DropDependsOnFromEmptyIterator}, @@ -8926,6 +8957,8 @@ struct TPeepHoleRules { {"CostsOf", &ExpandCostsOf}, {"JsonQuery", &ExpandJsonQuery}, {"MatchRecognize", &ExpandMatchRecognize}, + {"PruneAdjacentKeys", &ExpandPruneAdjacentKeys}, + {"PruneKeys", &ExpandPruneKeys}, {"CalcOverWindow", &ExpandCalcOverWindow}, {"CalcOverSessionWindow", &ExpandCalcOverWindow}, {"CalcOverWindowGroup", &ExpandCalcOverWindow}, diff --git a/yql/essentials/core/yql_expr_constraint.cpp b/yql/essentials/core/yql_expr_constraint.cpp index e67557d9645..7fcf560a2cf 100644 --- a/yql/essentials/core/yql_expr_constraint.cpp +++ b/yql/essentials/core/yql_expr_constraint.cpp @@ -822,9 +822,23 @@ private: } if constexpr (Adjacent) { - return CopyAllFrom<0>(input, output, ctx); + if (const auto status = CopyAllFrom<0>(input, output, ctx); status != TStatus::Ok) { + return status; + } + + TPartOfConstraintBase::TSetType keys = GetPathsToKeys<true>(input->Child(1)->Tail(), input->Child(1)->Head().Head()); + TPartOfConstraintBase::TSetOfSetsType uniqueKeys; + for (const auto& elem : keys) { + uniqueKeys.insert(TPartOfConstraintBase::TSetType{elem}); + } + if (!keys.empty()) { + input->AddConstraint(ctx.MakeConstraint<TUniqueConstraintNode>(TUniqueConstraintNode::TContentType{uniqueKeys})); + input->AddConstraint(ctx.MakeConstraint<TDistinctConstraintNode>(TDistinctConstraintNode::TContentType{uniqueKeys})); + } + return TStatus::Ok; } - return FromFirst<TEmptyConstraintNode, TUniqueConstraintNode, TPartOfUniqueConstraintNode, TDistinctConstraintNode, TPartOfDistinctConstraintNode>(input, output, ctx); + + return FromFirst<TEmptyConstraintNode>(input, output, ctx); } template<class TConstraint> diff --git a/yql/essentials/core/yql_join.cpp b/yql/essentials/core/yql_join.cpp index 7cca45604d1..5ce3fba268b 100644 --- a/yql/essentials/core/yql_join.cpp +++ b/yql/essentials/core/yql_join.cpp @@ -820,6 +820,10 @@ IGraphTransformer::TStatus ValidateEquiJoinOptions(TPositionHandle positionHandl // do nothing } else if (optionName == "multiple_joins") { // do nothing + } else if (optionName == "prune_keys_added") { + if (!EnsureTupleSize(*child, 1, ctx)) { + return IGraphTransformer::TStatus::Error; + } } else { YQL_ENSURE(false, "Cached join option '" << optionName << "' not handled"); } @@ -2007,7 +2011,7 @@ void GatherJoinInputs(const TExprNode::TPtr& expr, const TExprNode& row, } bool IsCachedJoinOption(TStringBuf name) { - static THashSet<TStringBuf> CachedJoinOptions = {"preferred_sort", "cbo_passed", "multiple_joins"}; + static THashSet<TStringBuf> CachedJoinOptions = {"preferred_sort", "cbo_passed", "multiple_joins", "prune_keys_added"}; return CachedJoinOptions.contains(name); } @@ -2016,4 +2020,31 @@ bool IsCachedJoinLinkOption(TStringBuf name) { return CachedJoinLinkOptions.contains(name); } +void GetPruneKeysColumnsForJoinLeaves(const TCoEquiJoinTuple& joinTree, THashMap<TStringBuf, THashSet<TStringBuf>>& columnsForPruneKeysExtractor) { + auto settings = GetEquiJoinLinkSettings(joinTree.Options().Ref()); + TStringBuf joinKind = joinTree.Type().Value(); + + auto left = joinTree.LeftScope(); + if (!left.Maybe<TCoAtom>()) { + GetPruneKeysColumnsForJoinLeaves(left.Cast<TCoEquiJoinTuple>(), columnsForPruneKeysExtractor); + } else { + if (joinKind == "RightSemi" || joinKind == "RightOnly" || settings.LeftHints.contains("any")) { + if (!settings.LeftHints.contains("unique")) { + CollectEquiJoinKeyColumnsFromLeaf(joinTree.LeftKeys().Ref(), columnsForPruneKeysExtractor); + } + } + } + + auto right = joinTree.RightScope(); + if (!right.Maybe<TCoAtom>()) { + GetPruneKeysColumnsForJoinLeaves(right.Cast<TCoEquiJoinTuple>(), columnsForPruneKeysExtractor); + } else { + if (joinKind == "LeftSemi" || joinKind == "LeftOnly" || settings.RightHints.contains("any")) { + if (!settings.RightHints.contains("unique")) { + CollectEquiJoinKeyColumnsFromLeaf(joinTree.RightKeys().Ref(), columnsForPruneKeysExtractor); + } + } + } +} + } // namespace NYql diff --git a/yql/essentials/core/yql_join.h b/yql/essentials/core/yql_join.h index a313aceefa3..26b417cf542 100644 --- a/yql/essentials/core/yql_join.h +++ b/yql/essentials/core/yql_join.h @@ -180,4 +180,6 @@ void GatherJoinInputs(const TExprNode::TPtr& expr, const TExprNode& row, bool IsCachedJoinOption(TStringBuf name); bool IsCachedJoinLinkOption(TStringBuf name); +void GetPruneKeysColumnsForJoinLeaves(const NNodes::TCoEquiJoinTuple& joinTree, THashMap<TStringBuf, THashSet<TStringBuf>>& columnsForPruneKeysExtractor); + } diff --git a/yql/essentials/core/yql_type_annotation.h b/yql/essentials/core/yql_type_annotation.h index bc09c963f4b..a43650a01f4 100644 --- a/yql/essentials/core/yql_type_annotation.h +++ b/yql/essentials/core/yql_type_annotation.h @@ -452,6 +452,7 @@ struct TTypeAnnotationContext: public TThrRefBase { THashSet<TString> PeepholeFlags; bool StreamLookupJoin = false; ui32 MaxAggPushdownPredicates = 6; // algorithm complexity is O(2^N) + ui32 PruneKeysMemLimit = 128 * 1024 * 1024; TMaybe<TColumnOrder> LookupColumnOrder(const TExprNode& node) const; IGraphTransformer::TStatus SetColumnOrder(const TExprNode& node, const TColumnOrder& columnOrder, TExprContext& ctx); diff --git a/yql/essentials/data/language/pragmas_opensource.json b/yql/essentials/data/language/pragmas_opensource.json index 26cc1c777a0..a4a3ad73364 100644 --- a/yql/essentials/data/language/pragmas_opensource.json +++ b/yql/essentials/data/language/pragmas_opensource.json @@ -1 +1 @@ -[{"name":"yt.Annotations"},{"name":"yt.ApplyStoredConstraints"},{"name":"yt.Auth"},{"name":"yt.AutoMerge"},{"name":"yt.BatchListFolderConcurrency"},{"name":"yt.BinaryExpirationInterval"},{"name":"yt.BinaryTmpFolder"},{"name":"yt.BlockMapJoin"},{"name":"yt.BlockReaderSupportedDataTypes"},{"name":"yt.BlockReaderSupportedTypes"},{"name":"yt.BufferRowCount"},{"name":"yt.ClientMapTimeout"},{"name":"yt.ColumnGroupMode"},{"name":"yt.CombineCoreLimit"},{"name":"yt.CommonJoinCoreLimit"},{"name":"yt.CompactForDistinct"},{"name":"yt.CoreDumpPath"},{"name":"yt.DQRPCReaderInflight"},{"name":"yt.DQRPCReaderTimeout"},{"name":"yt.DataSizePerJob"},{"name":"yt.DataSizePerMapJob"},{"name":"yt.DataSizePerPartition"},{"name":"yt.DataSizePerSortJob"},{"name":"yt.DefaultCalcMemoryLimit"},{"name":"yt.DefaultCluster"},{"name":"yt.DefaultLocalityTimeout"},{"name":"yt.DefaultMapSelectivityFactor"},{"name":"yt.DefaultMaxJobFails"},{"name":"yt.DefaultMemoryDigestLowerBound"},{"name":"yt.DefaultMemoryLimit"},{"name":"yt.DefaultMemoryReserveFactor"},{"name":"yt.DefaultOperationWeight"},{"name":"yt.DefaultRuntimeCluster"},{"name":"yt.Description"},{"name":"yt.DisableFuseOperations"},{"name":"yt.DisableJobSplitting"},{"name":"yt.DisableOptimizers"},{"name":"yt.DockerImage"},{"name":"yt.DqPruneKeyFilterLambda"},{"name":"yt.DropUnusedKeysFromKeyFilter"},{"name":"yt.EnableDynamicStoreReadInDQ"},{"name":"yt.EnableFuseMapToMapReduce"},{"name":"yt.EnforceJobUtc"},{"name":"yt.ErasureCodecCpu"},{"name":"yt.ErasureCodecCpuForDq"},{"name":"yt.EvaluationTableSizeLimit"},{"name":"yt.ExpirationDeadline"},{"name":"yt.ExpirationInterval"},{"name":"yt.ExtendTableLimit"},{"name":"yt.ExtendedStatsMaxChunkCount"},{"name":"yt.ExternalTx"},{"name":"yt.ExtraTmpfsSize"},{"name":"yt.FileCacheTtl"},{"name":"yt.FmrOperationSpec"},{"name":"yt.FolderInlineDataLimit"},{"name":"yt.FolderInlineItemsLimit"},{"name":"yt.ForceInferSchema"},{"name":"yt.ForceJobSizeAdjuster"},{"name":"yt.ForceTmpSecurity"},{"name":"yt.GeobaseDownloadUrl"},{"name":"yt.HybridDqDataSizeLimitForOrdered"},{"name":"yt.HybridDqDataSizeLimitForUnordered"},{"name":"yt.HybridDqExecution"},{"name":"yt.HybridDqExecutionFallback"},{"name":"yt.IgnoreTypeV3"},{"name":"yt.IgnoreWeakSchema"},{"name":"yt.IgnoreYamrDsv"},{"name":"yt.InferSchema"},{"name":"yt.InferSchemaMode"},{"name":"yt.InferSchemaTableCountThreshold"},{"name":"yt.InflightTempTablesLimit"},{"name":"yt.IntermediateAccount"},{"name":"yt.IntermediateDataMedium"},{"name":"yt.IntermediateReplicationFactor"},{"name":"yt.JavascriptCpu"},{"name":"yt.JobBlockInput"},{"name":"yt.JobBlockInputSupportedDataTypes"},{"name":"yt.JobBlockInputSupportedTypes"},{"name":"yt.JobBlockOutput"},{"name":"yt.JobBlockOutputSupportedDataTypes"},{"name":"yt.JobBlockOutputSupportedTypes"},{"name":"yt.JobBlockTableContent"},{"name":"yt.JobEnv"},{"name":"yt.JoinAllowColumnRenames"},{"name":"yt.JoinCollectColumnarStatistics"},{"name":"yt.JoinColumnarStatisticsFetcherMode"},{"name":"yt.JoinCommonUseMapMultiOut"},{"name":"yt.JoinEnableStarJoin"},{"name":"yt.JoinMergeForce"},{"name":"yt.JoinMergeReduceJobMaxSize"},{"name":"yt.JoinMergeSetTopLevelFullSort"},{"name":"yt.JoinMergeTablesLimit"},{"name":"yt.JoinMergeUnsortedFactor"},{"name":"yt.JoinMergeUseSmallAsPrimary"},{"name":"yt.JoinUseColumnarStatistics"},{"name":"yt.JoinWaitAllInputs"},{"name":"yt.KeepTempTables"},{"name":"yt.KeyFilterForStartsWith"},{"name":"yt.LLVMMemSize"},{"name":"yt.LLVMNodeCountLimit"},{"name":"yt.LLVMPerNodeMemSize"},{"name":"yt.LayerPaths"},{"name":"yt.LocalCalcLimit"},{"name":"yt.LookupJoinLimit"},{"name":"yt.LookupJoinMaxRows"},{"name":"yt.MapJoinLimit"},{"name":"yt.MapJoinShardCount"},{"name":"yt.MapJoinShardMinRows"},{"name":"yt.MapJoinUseFlow"},{"name":"yt.MapLocalityTimeout"},{"name":"yt.MaxChunksForDqRead"},{"name":"yt.MaxColumnGroups"},{"name":"yt.MaxCpuUsageToFuseMultiOuts"},{"name":"yt.MaxExtraJobMemoryToFuseOperations"},{"name":"yt.MaxInputTables"},{"name":"yt.MaxInputTablesForSortedMerge"},{"name":"yt.MaxJobCount"},{"name":"yt.MaxKeyRangeCount"},{"name":"yt.MaxKeyWeight"},{"name":"yt.MaxOperationFiles"},{"name":"yt.MaxOutputTables"},{"name":"yt.MaxReplicationFactorToFuseMultiOuts"},{"name":"yt.MaxReplicationFactorToFuseOperations"},{"name":"yt.MaxRowWeight"},{"name":"yt.MaxSpeculativeJobCountPerTask"},{"name":"yt.MergeAdjacentPointRanges"},{"name":"yt.MinColumnGroupSize"},{"name":"yt.MinLocalityInputDataWeight"},{"name":"yt.MinPublishedAvgChunkSize"},{"name":"yt.MinTempAvgChunkSize"},{"name":"yt.NativeYtTypeCompatibility"},{"name":"yt.NetworkProject"},{"name":"yt.NightlyCompress"},{"name":"yt.OperationReaders"},{"name":"yt.OperationSpec"},{"name":"yt.OptimizeFor"},{"name":"yt.Owners"},{"name":"yt.ParallelOperationsLimit"},{"name":"yt.PartitionByConstantKeysViaMap"},{"name":"yt.Pool"},{"name":"yt.PoolTrees"},{"name":"yt.PrimaryMedium"},{"name":"yt.PruneKeyFilterLambda"},{"name":"yt.PruneQLFilterLambda"},{"name":"yt.PublishedAutoMerge"},{"name":"yt.PublishedCompressionCodec"},{"name":"yt.PublishedErasureCodec"},{"name":"yt.PublishedMedia"},{"name":"yt.PublishedPrimaryMedium"},{"name":"yt.PublishedReplicationFactor"},{"name":"yt.PythonCpu"},{"name":"yt.QueryCacheChunkLimit"},{"name":"yt.QueryCacheIgnoreTableRevision"},{"name":"yt.QueryCacheMode"},{"name":"yt.QueryCacheSalt"},{"name":"yt.QueryCacheTtl"},{"name":"yt.QueryCacheUseExpirationTimeout"},{"name":"yt.QueryCacheUseForCalc"},{"name":"yt.ReduceLocalityTimeout"},{"name":"yt.ReleaseTempData"},{"name":"yt.ReportEquiJoinStats"},{"name":"yt.RuntimeCluster"},{"name":"yt.RuntimeClusterSelection"},{"name":"yt.SamplingIoBlockSize"},{"name":"yt.SchedulingTag"},{"name":"yt.SchedulingTagFilter"},{"name":"yt.ScriptCpu"},{"name":"yt.SortLocalityTimeout"},{"name":"yt.StartedBy"},{"name":"yt.StaticPool"},{"name":"yt.SuspendIfAccountLimitExceeded"},{"name":"yt.SwitchLimit"},{"name":"yt.TableContentColumnarStatistics"},{"name":"yt.TableContentCompressLevel"},{"name":"yt.TableContentDeliveryMode"},{"name":"yt.TableContentLocalExecution"},{"name":"yt.TableContentMaxChunksForNativeDelivery"},{"name":"yt.TableContentMaxInputTables"},{"name":"yt.TableContentMinAvgChunkSize"},{"name":"yt.TableContentTmpFolder"},{"name":"yt.TableContentUseSkiff"},{"name":"yt.TablesTmpFolder"},{"name":"yt.TempTablesTtl"},{"name":"yt.TemporaryAutoMerge"},{"name":"yt.TemporaryCompressionCodec"},{"name":"yt.TemporaryErasureCodec"},{"name":"yt.TemporaryMedia"},{"name":"yt.TemporaryPrimaryMedium"},{"name":"yt.TemporaryReplicationFactor"},{"name":"yt.TentativePoolTrees"},{"name":"yt.TentativeTreeEligibilityMaxJobDurationRatio"},{"name":"yt.TentativeTreeEligibilityMinJobDuration"},{"name":"yt.TentativeTreeEligibilitySampleJobCount"},{"name":"yt.TmpFolder"},{"name":"yt.TopSortMaxLimit"},{"name":"yt.TopSortRowMultiplierPerJob"},{"name":"yt.TopSortSizePerJob"},{"name":"yt.UseAggPhases"},{"name":"yt.UseColumnGroupsFromInputTables"},{"name":"yt.UseColumnarStatistics"},{"name":"yt.UseDefaultTentativePoolTrees"},{"name":"yt.UseFlow"},{"name":"yt.UseIntermediateSchema"},{"name":"yt.UseIntermediateStreams"},{"name":"yt.UseNativeDescSort"},{"name":"yt.UseNativeYtTypes"},{"name":"yt.UseNewPredicateExtraction"},{"name":"yt.UsePartitionsByKeysForFinalAgg"},{"name":"yt.UseQLFilter"},{"name":"yt.UseRPCReaderInDQ"},{"name":"yt.UseSkiff"},{"name":"yt.UseSystemColumns"},{"name":"yt.UseTmpfs"},{"name":"yt.UseTypeV2"},{"name":"yt.UseYqlRowSpecCompactForm"},{"name":"yt.UserSlots"},{"name":"yt.ViewIsolation"},{"name":"yt.WideFlowLimit"},{"name":"dq.AggregateStatsByStage"},{"name":"dq.AnalyticsHopping"},{"name":"dq.AnalyzeQuery"},{"name":"dq.ChannelBufferSize"},{"name":"dq.ChunkSizeLimit"},{"name":"dq.CollectCoreDumps"},{"name":"dq.ComputeActorType"},{"name":"dq.DataSizePerJob"},{"name":"dq.DisableCheckpoints"},{"name":"dq.DisableLLVMForBlockStages"},{"name":"dq.EnableChannelStats"},{"name":"dq.EnableComputeActor"},{"name":"dq.EnableDqReplicate"},{"name":"dq.EnableFullResultWrite"},{"name":"dq.EnableInsert"},{"name":"dq.EnableSpillingInChannels"},{"name":"dq.EnableSpillingNodes"},{"name":"dq.EnableStrip"},{"name":"dq.ExportStats"},{"name":"dq.FallbackPolicy"},{"name":"dq.HashJoinMode"},{"name":"dq.HashShuffleMaxTasks"},{"name":"dq.HashShuffleTasksRatio"},{"name":"dq.MaxDataSizePerJob"},{"name":"dq.MaxDataSizePerQuery"},{"name":"dq.MaxNetworkRetries"},{"name":"dq.MaxRetries"},{"name":"dq.MaxTasksPerOperation"},{"name":"dq.MaxTasksPerStage"},{"name":"dq.MemoryLimit"},{"name":"dq.OptLLVM"},{"name":"dq.OutputChunkMaxSize"},{"name":"dq.ParallelOperationsLimit"},{"name":"dq.PingTimeoutMs"},{"name":"dq.PullRequestTimeoutMs"},{"name":"dq.QueryTimeout"},{"name":"dq.RetryBackoffMs"},{"name":"dq.Scheduler"},{"name":"dq.SpillingEngine"},{"name":"dq.SplitStageOnDqReplicate"},{"name":"dq.TaskRunnerStats"},{"name":"dq.UseAggPhases"},{"name":"dq.UseBlockReader"},{"name":"dq.UseFastPickleTransport"},{"name":"dq.UseFinalizeByKey"},{"name":"dq.UseGraceJoinCoreForMap"},{"name":"dq.UseOOBTransport"},{"name":"dq.UseSimpleYtReader"},{"name":"dq.UseWideBlockChannels"},{"name":"dq.UseWideChannels"},{"name":"dq.WatermarksEnableIdlePartitions"},{"name":"dq.WatermarksGranularityMs"},{"name":"dq.WatermarksLateArrivalDelayMs"},{"name":"dq.WatermarksMode"},{"name":"dq.WorkerFilter"},{"name":"dq.WorkersPerOperation"},{"name":"AllowDotInAlias"},{"name":"AllowUnnamedColumns"},{"name":"AnsiCurrentRow"},{"name":"AnsiImplicitCrossJoin"},{"name":"AnsiInForEmptyOrNullableItemsCollections"},{"name":"AnsiLike"},{"name":"AnsiOptionalAs"},{"name":"AnsiRankForNullableKeys"},{"name":"AutoCommit"},{"name":"BlockEngine"},{"name":"BlockEngineEnable"},{"name":"BlockEngineForce"},{"name":"BogousStarInGroupByOverJoin"},{"name":"CheckedOps"},{"name":"ClassicDivision"},{"name":"CoalesceJoinKeysOnQualifiedAll"},{"name":"CompactGroupBy"},{"name":"CompactNamedExprs"},{"name":"CostBasedOptimizer"},{"name":"DataWatermarks"},{"name":"DirectRead"},{"name":"DisableAnsiCurrentRow"},{"name":"DisableAnsiImplicitCrossJoin"},{"name":"DisableAnsiInForEmptyOrNullableItemsCollections"},{"name":"DisableAnsiLike"},{"name":"DisableAnsiOptionalAs"},{"name":"DisableAnsiRankForNullableKeys"},{"name":"DisableBlockEngineEnable"},{"name":"DisableBlockEngineForce"},{"name":"DisableBogousStarInGroupByOverJoin"},{"name":"DisableCoalesceJoinKeysOnQualifiedAll"},{"name":"DisableCompactGroupBy"},{"name":"DisableCompactNamedExprs"},{"name":"DisableDistinctOverKeys"},{"name":"DisableDistinctOverWindow"},{"name":"DisableDqEngineEnable"},{"name":"DisableDqEngineForce"},{"name":"DisableEmitAggApply"},{"name":"DisableEmitStartsWith"},{"name":"DisableEmitTableSource"},{"name":"DisableEmitUnionMerge"},{"name":"DisableFilterPushdownOverJoinOptionalSide"},{"name":"DisableFlexibleTypes"},{"name":"DisableJsonQueryReturnsJsonDocument"},{"name":"DisableOrderedColumns"},{"name":"DisablePullUpFlatMapOverJoin"},{"name":"DisableRegexUseRe2"},{"name":"DisableRotateJoinTree"},{"name":"DisableSeqMode"},{"name":"DisableSimpleColumns"},{"name":"DisableStrictJoinKeyTypes"},{"name":"DisableUnicodeLiterals"},{"name":"DisableUnorderedResult"},{"name":"DisableUnorderedSubqueries"},{"name":"DisableUseBlocks"},{"name":"DisableValidateUnusedExprs"},{"name":"DisableWarnOnAnsiAliasShadowing"},{"name":"DisableWarnUntypedStringLiterals"},{"name":"DiscoveryMode"},{"name":"DistinctOverKeys"},{"name":"DistinctOverWindow"},{"name":"DqEngine"},{"name":"DqEngineEnable"},{"name":"DqEngineForce"},{"name":"EmitAggApply"},{"name":"EmitStartsWith"},{"name":"EmitTableSource"},{"name":"EmitUnionMerge"},{"name":"EnableSystemColumns"},{"name":"Engine"},{"name":"ErrorMsg"},{"name":"FeatureR010"},{"name":"File"},{"name":"FileOption"},{"name":"FilterPushdownOverJoinOptionalSide"},{"name":"FlexibleTypes"},{"name":"Folder"},{"name":"Greetings"},{"name":"GroupByCubeLimit"},{"name":"GroupByLimit"},{"name":"JsonQueryReturnsJsonDocument"},{"name":"Library"},{"name":"OrderedColumns"},{"name":"OverrideLibrary"},{"name":"Package"},{"name":"PackageVersion"},{"name":"PathPrefix"},{"name":"PositionalUnionAll"},{"name":"PqReadBy"},{"name":"PullUpFlatMapOverJoin"},{"name":"RefSelect"},{"name":"RegexUseRe2"},{"name":"ResultRowsLimit"},{"name":"ResultSizeLimit"},{"name":"RotateJoinTree"},{"name":"RuntimeLogLevel"},{"name":"SampleSelect"},{"name":"SeqMode"},{"name":"SimpleColumns"},{"name":"StrictJoinKeyTypes"},{"name":"Udf"},{"name":"UnicodeLiterals"},{"name":"UnorderedResult"},{"name":"UnorderedSubqueries"},{"name":"UseBlocks"},{"name":"UseTablePrefixForEach"},{"name":"ValidateUnusedExprs"},{"name":"WarnOnAnsiAliasShadowing"},{"name":"WarnUnnamedColumns"},{"name":"WarnUntypedStringLiterals"},{"name":"Warning"},{"name":"WarningMsg"},{"name":"yson.AutoConvert"},{"name":"yson.CastToString"},{"name":"yson.DisableCastToString"},{"name":"yson.DisableStrict"},{"name":"yson.Strict"}] +[{"name":"yt.Annotations"},{"name":"yt.ApplyStoredConstraints"},{"name":"yt.Auth"},{"name":"yt.AutoMerge"},{"name":"yt.BatchListFolderConcurrency"},{"name":"yt.BinaryExpirationInterval"},{"name":"yt.BinaryTmpFolder"},{"name":"yt.BlockMapJoin"},{"name":"yt.BlockReaderSupportedDataTypes"},{"name":"yt.BlockReaderSupportedTypes"},{"name":"yt.BufferRowCount"},{"name":"yt.ClientMapTimeout"},{"name":"yt.ColumnGroupMode"},{"name":"yt.CombineCoreLimit"},{"name":"yt.CommonJoinCoreLimit"},{"name":"yt.CompactForDistinct"},{"name":"yt.CoreDumpPath"},{"name":"yt.DQRPCReaderInflight"},{"name":"yt.DQRPCReaderTimeout"},{"name":"yt.DataSizePerJob"},{"name":"yt.DataSizePerMapJob"},{"name":"yt.DataSizePerPartition"},{"name":"yt.DataSizePerSortJob"},{"name":"yt.DefaultCalcMemoryLimit"},{"name":"yt.DefaultCluster"},{"name":"yt.DefaultLocalityTimeout"},{"name":"yt.DefaultMapSelectivityFactor"},{"name":"yt.DefaultMaxJobFails"},{"name":"yt.DefaultMemoryDigestLowerBound"},{"name":"yt.DefaultMemoryLimit"},{"name":"yt.DefaultMemoryReserveFactor"},{"name":"yt.DefaultOperationWeight"},{"name":"yt.DefaultRuntimeCluster"},{"name":"yt.Description"},{"name":"yt.DisableFuseOperations"},{"name":"yt.DisableJobSplitting"},{"name":"yt.DisableOptimizers"},{"name":"yt.DockerImage"},{"name":"yt.DqPruneKeyFilterLambda"},{"name":"yt.DropUnusedKeysFromKeyFilter"},{"name":"yt.EnableDynamicStoreReadInDQ"},{"name":"yt.EnableFuseMapToMapReduce"},{"name":"yt.EnforceJobUtc"},{"name":"yt.ErasureCodecCpu"},{"name":"yt.ErasureCodecCpuForDq"},{"name":"yt.EvaluationTableSizeLimit"},{"name":"yt.ExpirationDeadline"},{"name":"yt.ExpirationInterval"},{"name":"yt.ExtendTableLimit"},{"name":"yt.ExtendedStatsMaxChunkCount"},{"name":"yt.ExternalTx"},{"name":"yt.ExtraTmpfsSize"},{"name":"yt.FileCacheTtl"},{"name":"yt.FmrOperationSpec"},{"name":"yt.FolderInlineDataLimit"},{"name":"yt.FolderInlineItemsLimit"},{"name":"yt.ForceInferSchema"},{"name":"yt.ForceJobSizeAdjuster"},{"name":"yt.ForceTmpSecurity"},{"name":"yt.GeobaseDownloadUrl"},{"name":"yt.HybridDqDataSizeLimitForOrdered"},{"name":"yt.HybridDqDataSizeLimitForUnordered"},{"name":"yt.HybridDqExecution"},{"name":"yt.HybridDqExecutionFallback"},{"name":"yt.IgnoreTypeV3"},{"name":"yt.IgnoreWeakSchema"},{"name":"yt.IgnoreYamrDsv"},{"name":"yt.InferSchema"},{"name":"yt.InferSchemaMode"},{"name":"yt.InferSchemaTableCountThreshold"},{"name":"yt.InflightTempTablesLimit"},{"name":"yt.IntermediateAccount"},{"name":"yt.IntermediateDataMedium"},{"name":"yt.IntermediateReplicationFactor"},{"name":"yt.JavascriptCpu"},{"name":"yt.JobBlockInput"},{"name":"yt.JobBlockInputSupportedDataTypes"},{"name":"yt.JobBlockInputSupportedTypes"},{"name":"yt.JobBlockOutput"},{"name":"yt.JobBlockOutputSupportedDataTypes"},{"name":"yt.JobBlockOutputSupportedTypes"},{"name":"yt.JobBlockTableContent"},{"name":"yt.JobEnv"},{"name":"yt.JoinAllowColumnRenames"},{"name":"yt.JoinCollectColumnarStatistics"},{"name":"yt.JoinColumnarStatisticsFetcherMode"},{"name":"yt.JoinCommonUseMapMultiOut"},{"name":"yt.JoinEnableStarJoin"},{"name":"yt.JoinMergeForce"},{"name":"yt.JoinMergeReduceJobMaxSize"},{"name":"yt.JoinMergeSetTopLevelFullSort"},{"name":"yt.JoinMergeTablesLimit"},{"name":"yt.JoinMergeUnsortedFactor"},{"name":"yt.JoinMergeUseSmallAsPrimary"},{"name":"yt.JoinUseColumnarStatistics"},{"name":"yt.JoinWaitAllInputs"},{"name":"yt.KeepTempTables"},{"name":"yt.KeyFilterForStartsWith"},{"name":"yt.LLVMMemSize"},{"name":"yt.LLVMNodeCountLimit"},{"name":"yt.LLVMPerNodeMemSize"},{"name":"yt.LayerPaths"},{"name":"yt.LocalCalcLimit"},{"name":"yt.LookupJoinLimit"},{"name":"yt.LookupJoinMaxRows"},{"name":"yt.MapJoinLimit"},{"name":"yt.MapJoinShardCount"},{"name":"yt.MapJoinShardMinRows"},{"name":"yt.MapJoinUseFlow"},{"name":"yt.MapLocalityTimeout"},{"name":"yt.MaxChunksForDqRead"},{"name":"yt.MaxColumnGroups"},{"name":"yt.MaxCpuUsageToFuseMultiOuts"},{"name":"yt.MaxExtraJobMemoryToFuseOperations"},{"name":"yt.MaxInputTables"},{"name":"yt.MaxInputTablesForSortedMerge"},{"name":"yt.MaxJobCount"},{"name":"yt.MaxKeyRangeCount"},{"name":"yt.MaxKeyWeight"},{"name":"yt.MaxOperationFiles"},{"name":"yt.MaxOutputTables"},{"name":"yt.MaxReplicationFactorToFuseMultiOuts"},{"name":"yt.MaxReplicationFactorToFuseOperations"},{"name":"yt.MaxRowWeight"},{"name":"yt.MaxSpeculativeJobCountPerTask"},{"name":"yt.MergeAdjacentPointRanges"},{"name":"yt.MinColumnGroupSize"},{"name":"yt.MinLocalityInputDataWeight"},{"name":"yt.MinPublishedAvgChunkSize"},{"name":"yt.MinTempAvgChunkSize"},{"name":"yt.NativeYtTypeCompatibility"},{"name":"yt.NetworkProject"},{"name":"yt.NightlyCompress"},{"name":"yt.OperationReaders"},{"name":"yt.OperationSpec"},{"name":"yt.OptimizeFor"},{"name":"yt.Owners"},{"name":"yt.ParallelOperationsLimit"},{"name":"yt.PartitionByConstantKeysViaMap"},{"name":"yt.Pool"},{"name":"yt.PoolTrees"},{"name":"yt.PrimaryMedium"},{"name":"yt.PruneKeyFilterLambda"},{"name":"yt.PruneQLFilterLambda"},{"name":"yt.PublishedAutoMerge"},{"name":"yt.PublishedCompressionCodec"},{"name":"yt.PublishedErasureCodec"},{"name":"yt.PublishedMedia"},{"name":"yt.PublishedPrimaryMedium"},{"name":"yt.PublishedReplicationFactor"},{"name":"yt.PythonCpu"},{"name":"yt.QueryCacheChunkLimit"},{"name":"yt.QueryCacheIgnoreTableRevision"},{"name":"yt.QueryCacheMode"},{"name":"yt.QueryCacheSalt"},{"name":"yt.QueryCacheTtl"},{"name":"yt.QueryCacheUseExpirationTimeout"},{"name":"yt.QueryCacheUseForCalc"},{"name":"yt.ReduceLocalityTimeout"},{"name":"yt.ReleaseTempData"},{"name":"yt.ReportEquiJoinStats"},{"name":"yt.RuntimeCluster"},{"name":"yt.RuntimeClusterSelection"},{"name":"yt.SamplingIoBlockSize"},{"name":"yt.SchedulingTag"},{"name":"yt.SchedulingTagFilter"},{"name":"yt.ScriptCpu"},{"name":"yt.SortLocalityTimeout"},{"name":"yt.StartedBy"},{"name":"yt.StaticPool"},{"name":"yt.SuspendIfAccountLimitExceeded"},{"name":"yt.SwitchLimit"},{"name":"yt.TableContentColumnarStatistics"},{"name":"yt.TableContentCompressLevel"},{"name":"yt.TableContentDeliveryMode"},{"name":"yt.TableContentLocalExecution"},{"name":"yt.TableContentMaxChunksForNativeDelivery"},{"name":"yt.TableContentMaxInputTables"},{"name":"yt.TableContentMinAvgChunkSize"},{"name":"yt.TableContentTmpFolder"},{"name":"yt.TableContentUseSkiff"},{"name":"yt.TablesTmpFolder"},{"name":"yt.TempTablesTtl"},{"name":"yt.TemporaryAutoMerge"},{"name":"yt.TemporaryCompressionCodec"},{"name":"yt.TemporaryErasureCodec"},{"name":"yt.TemporaryMedia"},{"name":"yt.TemporaryPrimaryMedium"},{"name":"yt.TemporaryReplicationFactor"},{"name":"yt.TentativePoolTrees"},{"name":"yt.TentativeTreeEligibilityMaxJobDurationRatio"},{"name":"yt.TentativeTreeEligibilityMinJobDuration"},{"name":"yt.TentativeTreeEligibilitySampleJobCount"},{"name":"yt.TmpFolder"},{"name":"yt.TopSortMaxLimit"},{"name":"yt.TopSortRowMultiplierPerJob"},{"name":"yt.TopSortSizePerJob"},{"name":"yt.UseAggPhases"},{"name":"yt.UseColumnGroupsFromInputTables"},{"name":"yt.UseColumnarStatistics"},{"name":"yt.UseDefaultTentativePoolTrees"},{"name":"yt.UseFlow"},{"name":"yt.UseIntermediateSchema"},{"name":"yt.UseIntermediateStreams"},{"name":"yt.UseNativeDescSort"},{"name":"yt.UseNativeDynamicTableRead"},{"name":"yt.UseNativeYtTypes"},{"name":"yt.UseNewPredicateExtraction"},{"name":"yt.UsePartitionsByKeysForFinalAgg"},{"name":"yt.UseQLFilter"},{"name":"yt.UseRPCReaderInDQ"},{"name":"yt.UseSkiff"},{"name":"yt.UseSystemColumns"},{"name":"yt.UseTmpfs"},{"name":"yt.UseTypeV2"},{"name":"yt.UseYqlRowSpecCompactForm"},{"name":"yt.UserSlots"},{"name":"yt.ViewIsolation"},{"name":"yt.WideFlowLimit"},{"name":"dq.AggregateStatsByStage"},{"name":"dq.AnalyticsHopping"},{"name":"dq.AnalyzeQuery"},{"name":"dq.ChannelBufferSize"},{"name":"dq.ChunkSizeLimit"},{"name":"dq.CollectCoreDumps"},{"name":"dq.ComputeActorType"},{"name":"dq.DataSizePerJob"},{"name":"dq.DisableCheckpoints"},{"name":"dq.DisableLLVMForBlockStages"},{"name":"dq.EnableChannelStats"},{"name":"dq.EnableComputeActor"},{"name":"dq.EnableDqReplicate"},{"name":"dq.EnableFullResultWrite"},{"name":"dq.EnableInsert"},{"name":"dq.EnableSpillingInChannels"},{"name":"dq.EnableSpillingNodes"},{"name":"dq.EnableStrip"},{"name":"dq.ExportStats"},{"name":"dq.FallbackPolicy"},{"name":"dq.HashJoinMode"},{"name":"dq.HashShuffleMaxTasks"},{"name":"dq.HashShuffleTasksRatio"},{"name":"dq.MaxDataSizePerJob"},{"name":"dq.MaxDataSizePerQuery"},{"name":"dq.MaxNetworkRetries"},{"name":"dq.MaxRetries"},{"name":"dq.MaxTasksPerOperation"},{"name":"dq.MaxTasksPerStage"},{"name":"dq.MemoryLimit"},{"name":"dq.OptLLVM"},{"name":"dq.OutputChunkMaxSize"},{"name":"dq.ParallelOperationsLimit"},{"name":"dq.PingTimeoutMs"},{"name":"dq.PullRequestTimeoutMs"},{"name":"dq.QueryTimeout"},{"name":"dq.RetryBackoffMs"},{"name":"dq.Scheduler"},{"name":"dq.SpillingEngine"},{"name":"dq.SplitStageOnDqReplicate"},{"name":"dq.TaskRunnerStats"},{"name":"dq.UseAggPhases"},{"name":"dq.UseBlockReader"},{"name":"dq.UseFastPickleTransport"},{"name":"dq.UseFinalizeByKey"},{"name":"dq.UseGraceJoinCoreForMap"},{"name":"dq.UseOOBTransport"},{"name":"dq.UseSimpleYtReader"},{"name":"dq.UseWideBlockChannels"},{"name":"dq.UseWideChannels"},{"name":"dq.WatermarksEnableIdlePartitions"},{"name":"dq.WatermarksGranularityMs"},{"name":"dq.WatermarksLateArrivalDelayMs"},{"name":"dq.WatermarksMode"},{"name":"dq.WorkerFilter"},{"name":"dq.WorkersPerOperation"},{"name":"AllowDotInAlias"},{"name":"AllowUnnamedColumns"},{"name":"AnsiCurrentRow"},{"name":"AnsiImplicitCrossJoin"},{"name":"AnsiInForEmptyOrNullableItemsCollections"},{"name":"AnsiLike"},{"name":"AnsiOptionalAs"},{"name":"AnsiRankForNullableKeys"},{"name":"AutoCommit"},{"name":"BlockEngine"},{"name":"BlockEngineEnable"},{"name":"BlockEngineForce"},{"name":"BogousStarInGroupByOverJoin"},{"name":"CheckedOps"},{"name":"ClassicDivision"},{"name":"CoalesceJoinKeysOnQualifiedAll"},{"name":"CompactGroupBy"},{"name":"CompactNamedExprs"},{"name":"CostBasedOptimizer"},{"name":"DataWatermarks"},{"name":"DirectRead"},{"name":"DisableAnsiCurrentRow"},{"name":"DisableAnsiImplicitCrossJoin"},{"name":"DisableAnsiInForEmptyOrNullableItemsCollections"},{"name":"DisableAnsiLike"},{"name":"DisableAnsiOptionalAs"},{"name":"DisableAnsiRankForNullableKeys"},{"name":"DisableBlockEngineEnable"},{"name":"DisableBlockEngineForce"},{"name":"DisableBogousStarInGroupByOverJoin"},{"name":"DisableCoalesceJoinKeysOnQualifiedAll"},{"name":"DisableCompactGroupBy"},{"name":"DisableCompactNamedExprs"},{"name":"DisableDistinctOverKeys"},{"name":"DisableDistinctOverWindow"},{"name":"DisableDqEngineEnable"},{"name":"DisableDqEngineForce"},{"name":"DisableEmitAggApply"},{"name":"DisableEmitStartsWith"},{"name":"DisableEmitTableSource"},{"name":"DisableEmitUnionMerge"},{"name":"DisableFilterPushdownOverJoinOptionalSide"},{"name":"DisableFlexibleTypes"},{"name":"DisableGroupByExprAfterWhere"},{"name":"DisableJsonQueryReturnsJsonDocument"},{"name":"DisableOrderedColumns"},{"name":"DisablePullUpFlatMapOverJoin"},{"name":"DisableRegexUseRe2"},{"name":"DisableRotateJoinTree"},{"name":"DisableSeqMode"},{"name":"DisableSimpleColumns"},{"name":"DisableStrictJoinKeyTypes"},{"name":"DisableUnicodeLiterals"},{"name":"DisableUnorderedResult"},{"name":"DisableUnorderedSubqueries"},{"name":"DisableUseBlocks"},{"name":"DisableValidateUnusedExprs"},{"name":"DisableWarnOnAnsiAliasShadowing"},{"name":"DisableWarnUntypedStringLiterals"},{"name":"DiscoveryMode"},{"name":"DistinctOverKeys"},{"name":"DistinctOverWindow"},{"name":"DqEngine"},{"name":"DqEngineEnable"},{"name":"DqEngineForce"},{"name":"EmitAggApply"},{"name":"EmitStartsWith"},{"name":"EmitTableSource"},{"name":"EmitUnionMerge"},{"name":"EnableSystemColumns"},{"name":"Engine"},{"name":"ErrorMsg"},{"name":"FeatureR010"},{"name":"File"},{"name":"FileOption"},{"name":"FilterPushdownOverJoinOptionalSide"},{"name":"FlexibleTypes"},{"name":"Folder"},{"name":"Greetings"},{"name":"GroupByCubeLimit"},{"name":"GroupByExprAfterWhere"},{"name":"GroupByLimit"},{"name":"JsonQueryReturnsJsonDocument"},{"name":"Library"},{"name":"OrderedColumns"},{"name":"OverrideLibrary"},{"name":"Package"},{"name":"PackageVersion"},{"name":"PathPrefix"},{"name":"PositionalUnionAll"},{"name":"PqReadBy"},{"name":"PullUpFlatMapOverJoin"},{"name":"RefSelect"},{"name":"RegexUseRe2"},{"name":"ResultRowsLimit"},{"name":"ResultSizeLimit"},{"name":"RotateJoinTree"},{"name":"RuntimeLogLevel"},{"name":"SampleSelect"},{"name":"SeqMode"},{"name":"SimpleColumns"},{"name":"StrictJoinKeyTypes"},{"name":"Udf"},{"name":"UnicodeLiterals"},{"name":"UnorderedResult"},{"name":"UnorderedSubqueries"},{"name":"UseBlocks"},{"name":"UseTablePrefixForEach"},{"name":"ValidateUnusedExprs"},{"name":"WarnOnAnsiAliasShadowing"},{"name":"WarnUnnamedColumns"},{"name":"WarnUntypedStringLiterals"},{"name":"Warning"},{"name":"WarningMsg"},{"name":"yson.AutoConvert"},{"name":"yson.CastToString"},{"name":"yson.DisableCastToString"},{"name":"yson.DisableStrict"},{"name":"yson.Strict"}] diff --git a/yql/essentials/minikql/comp_nodes/mkql_condense.cpp b/yql/essentials/minikql/comp_nodes/mkql_condense.cpp index ced3a95d5ab..04181778482 100644 --- a/yql/essentials/minikql/comp_nodes/mkql_condense.cpp +++ b/yql/essentials/minikql/comp_nodes/mkql_condense.cpp @@ -30,7 +30,7 @@ public: NUdf::TUnboxedValuePod DoCalculate(NUdf::TUnboxedValue& state, TComputationContext& ctx) const { if (state.IsFinish()) { - return static_cast<const NUdf::TUnboxedValuePod&>(state); + return state; } if (state.IsInvalid()) { diff --git a/yql/essentials/minikql/comp_nodes/mkql_condense1.cpp b/yql/essentials/minikql/comp_nodes/mkql_condense1.cpp index 850a7bff042..b2cda5a5872 100644 --- a/yql/essentials/minikql/comp_nodes/mkql_condense1.cpp +++ b/yql/essentials/minikql/comp_nodes/mkql_condense1.cpp @@ -30,7 +30,7 @@ public: NUdf::TUnboxedValuePod DoCalculate(NUdf::TUnboxedValue& state, TComputationContext& ctx) const { if (state.IsFinish()) { - return static_cast<const NUdf::TUnboxedValuePod&>(state); + return state; } else if (state.HasValue()) { if constexpr (UseCtx) { CleanupCurrentContext(); diff --git a/yql/essentials/minikql/comp_nodes/mkql_flatmap.cpp b/yql/essentials/minikql/comp_nodes/mkql_flatmap.cpp index 923974e9e7e..69c42c7afbd 100644 --- a/yql/essentials/minikql/comp_nodes/mkql_flatmap.cpp +++ b/yql/essentials/minikql/comp_nodes/mkql_flatmap.cpp @@ -219,7 +219,7 @@ public: } if (state.IsFinish()) { - return NUdf::TUnboxedValuePod::MakeFinish(); + return state; } while (true) { diff --git a/yql/essentials/minikql/comp_nodes/mkql_multimap.cpp b/yql/essentials/minikql/comp_nodes/mkql_multimap.cpp index 1a0639b9e9b..5e33c133c31 100644 --- a/yql/essentials/minikql/comp_nodes/mkql_multimap.cpp +++ b/yql/essentials/minikql/comp_nodes/mkql_multimap.cpp @@ -21,8 +21,9 @@ public: {} NUdf::TUnboxedValuePod DoCalculate(NUdf::TUnboxedValue& state, TComputationContext& ctx) const { - if (state.IsFinish()) - return NUdf::TUnboxedValuePod::MakeFinish(); + if (state.IsFinish()) { + return state; + } const auto pos = state.IsInvalid() ? 0ULL : state.Get<ui64>(); if (!pos) { @@ -412,8 +413,9 @@ public: {} NUdf::TUnboxedValuePod DoCalculate(NUdf::TUnboxedValue& state, TComputationContext& ctx) const { - if (state.IsFinish()) - return NUdf::TUnboxedValuePod::MakeFinish(); + if (state.IsFinish()) { + return state; + } const auto pos = state.IsInvalid() ? 0ULL : state.Get<ui64>(); if (!pos) { diff --git a/yql/essentials/minikql/comp_nodes/mkql_squeeze_to_list.cpp b/yql/essentials/minikql/comp_nodes/mkql_squeeze_to_list.cpp index 548eb0937df..55bd62b112b 100644 --- a/yql/essentials/minikql/comp_nodes/mkql_squeeze_to_list.cpp +++ b/yql/essentials/minikql/comp_nodes/mkql_squeeze_to_list.cpp @@ -47,7 +47,7 @@ public: NUdf::TUnboxedValuePod DoCalculate(NUdf::TUnboxedValue& state, TComputationContext& ctx) const { if (state.IsFinish()) { - return NUdf::TUnboxedValuePod::MakeFinish(); + return state; } else if (state.IsInvalid()) { MakeState(ctx, Limit->GetValue(ctx).GetOrDefault(std::numeric_limits<ui64>::max()), state); } diff --git a/yql/essentials/minikql/comp_nodes/mkql_todict.cpp b/yql/essentials/minikql/comp_nodes/mkql_todict.cpp index 738f88232a9..0ca6c68166a 100644 --- a/yql/essentials/minikql/comp_nodes/mkql_todict.cpp +++ b/yql/essentials/minikql/comp_nodes/mkql_todict.cpp @@ -985,7 +985,7 @@ public: NUdf::TUnboxedValuePod DoCalculate(NUdf::TUnboxedValue& state, TComputationContext& ctx) const { if (state.IsFinish()) { - return state.Release(); + return state; } else if (state.IsInvalid()) { MakeState(ctx, state); } @@ -1162,7 +1162,7 @@ public: NUdf::TUnboxedValuePod DoCalculate(NUdf::TUnboxedValue& state, TComputationContext& ctx) const { if (state.IsFinish()) { - return state.Release(); + return state; } else if (state.IsInvalid()) { MakeState(ctx, state); } diff --git a/yql/essentials/minikql/comp_nodes/ut/mkql_todict_ut.cpp b/yql/essentials/minikql/comp_nodes/ut/mkql_todict_ut.cpp index abb4f5d85e9..f09b9529d26 100644 --- a/yql/essentials/minikql/comp_nodes/ut/mkql_todict_ut.cpp +++ b/yql/essentials/minikql/comp_nodes/ut/mkql_todict_ut.cpp @@ -147,6 +147,10 @@ Y_UNIT_TEST_SUITE(TMiniKQLToDictTest) { status = res.Fetch(v); UNIT_ASSERT_VALUES_EQUAL(NUdf::EFetchStatus::Finish, status); + // XXX: Check whether the internal state is not released + // and the sentinel is still set (see more info in YQL-19866). + status = res.Fetch(v); + UNIT_ASSERT_VALUES_EQUAL(NUdf::EFetchStatus::Finish, status); }; for (auto stream : {true, false}) { @@ -201,6 +205,10 @@ Y_UNIT_TEST_SUITE(TMiniKQLToDictTest) { status = res.Fetch(v); UNIT_ASSERT_VALUES_EQUAL(NUdf::EFetchStatus::Finish, status); + // XXX: Check whether the internal state is not released + // and the sentinel is still set (see more info in YQL-19866). + status = res.Fetch(v); + UNIT_ASSERT_VALUES_EQUAL(NUdf::EFetchStatus::Finish, status); }; for (auto hashed : {true, false}) { diff --git a/yql/essentials/sql/v1/context.cpp b/yql/essentials/sql/v1/context.cpp index 1a0a1f4b18d..7f3d5433b96 100644 --- a/yql/essentials/sql/v1/context.cpp +++ b/yql/essentials/sql/v1/context.cpp @@ -69,6 +69,7 @@ THashMap<TStringBuf, TPragmaField> CTX_PRAGMA_FIELDS = { {"EmitUnionMerge", &TContext::EmitUnionMerge}, {"SeqMode", &TContext::SeqMode}, {"DistinctOverKeys", &TContext::DistinctOverKeys}, + {"GroupByExprAfterWhere", &TContext::GroupByExprAfterWhere}, }; typedef TMaybe<bool> TContext::*TPragmaMaybeField; @@ -104,6 +105,10 @@ TContext::TContext(const TLexers& lexers, const TParsers& parsers, , WarningPolicy(settings.IsReplay) , BlockEngineEnable(Settings.BlockDefaultAuto->Allow()) { + if (settings.LangVer >= MakeLangVersion(2025, 2)) { + GroupByExprAfterWhere = true; + } + for (auto lib : settings.Libraries) { Libraries.emplace(lib, TLibraryStuff()); } diff --git a/yql/essentials/sql/v1/context.h b/yql/essentials/sql/v1/context.h index 3bdfa1ceab4..e3f70f7515e 100644 --- a/yql/essentials/sql/v1/context.h +++ b/yql/essentials/sql/v1/context.h @@ -373,6 +373,7 @@ namespace NSQLTranslationV1 { bool DistinctOverWindow = false; bool SeqMode = false; bool DistinctOverKeys = false; + bool GroupByExprAfterWhere = false; bool EmitUnionMerge = false; TVector<size_t> ForAllStatementsParts; diff --git a/yql/essentials/sql/v1/select.cpp b/yql/essentials/sql/v1/select.cpp index c4290f3268f..cb72dc95f5f 100644 --- a/yql/essentials/sql/v1/select.cpp +++ b/yql/essentials/sql/v1/select.cpp @@ -1744,6 +1744,11 @@ public: if (Flatten) { block = L(block, Y("let", "core", Y(ordered ? "OrderedFlatMap" : "FlatMap", "core", BuildLambda(Pos, Y("row"), Flatten, "res")))); } + if (ctx.GroupByExprAfterWhere) { + if (auto filter = Source->BuildFilter(ctx, "core"); filter) { + block = L(block, Y("let", "core", filter)); + } + } if (PreaggregatedMap) { block = L(block, Y("let", "core", PreaggregatedMap)); if (Source->IsCompositeSource() && !Columns.QualifiedAll) { @@ -1752,9 +1757,10 @@ public: } else if (Source->IsCompositeSource() && !Columns.QualifiedAll) { block = L(block, Y("let", "origcore", "core")); } - auto filter = Source->BuildFilter(ctx, "core"); - if (filter) { - block = L(block, Y("let", "core", filter)); + if (!ctx.GroupByExprAfterWhere) { + if (auto filter = Source->BuildFilter(ctx, "core"); filter) { + block = L(block, Y("let", "core", filter)); + } } if (Aggregate) { block = L(block, Y("let", "core", Aggregate)); diff --git a/yql/essentials/sql/v1/sql_query.cpp b/yql/essentials/sql/v1/sql_query.cpp index b60a338d02f..d64a7bf8d61 100644 --- a/yql/essentials/sql/v1/sql_query.cpp +++ b/yql/essentials/sql/v1/sql_query.cpp @@ -3393,6 +3393,12 @@ TNodePtr TSqlQuery::PragmaStatement(const TRule_pragma_stmt& stmt, bool& success } else if (normalizedPragma == "disabledistinctoverkeys") { Ctx.DistinctOverKeys = false; Ctx.IncrementMonCounter("sql_pragma", "DisableDistinctOverKeys"); + } else if (normalizedPragma == "groupbyexprafterwhere") { + Ctx.GroupByExprAfterWhere = true; + Ctx.IncrementMonCounter("sql_pragma", "GroupByExprAfterWhere"); + } else if (normalizedPragma == "disablegroupbyexprafterwhere") { + Ctx.GroupByExprAfterWhere = false; + Ctx.IncrementMonCounter("sql_pragma", "DisableGroupByExprAfterWhere"); } else if (normalizedPragma == "engine") { Ctx.IncrementMonCounter("sql_pragma", "Engine"); diff --git a/yql/essentials/tests/common/test_framework/test_file_common.py b/yql/essentials/tests/common/test_framework/test_file_common.py index 240182e0056..9c91f8a1671 100644 --- a/yql/essentials/tests/common/test_framework/test_file_common.py +++ b/yql/essentials/tests/common/test_framework/test_file_common.py @@ -88,7 +88,7 @@ def get_sql_query(provider, suite, case, config, data_path=None, template='.sql' def run_file_no_cache(provider, suite, case, cfg, config, yql_http_file_server, yqlrun_binary=None, extra_args=[], force_blocks=False, allow_llvm=True, data_path=None, - run_sql=True, cfg_postprocess=None): + run_sql=True, cfg_postprocess=None, langver=None): check_provider(provider, config) sql_query = get_sql_query(provider, suite, case, config, data_path, template='.sql' if run_sql else '.yqls') @@ -119,7 +119,8 @@ def run_file_no_cache(provider, suite, case, cfg, config, yql_http_file_server, gateway_config=get_gateways_config(http_files, yql_http_file_server, force_blocks=force_blocks, is_hybrid=is_hybrid(provider), allow_llvm=allow_llvm, postprocess_func=cfg_postprocess), extra_args=extra_args, - udfs_dir=yql_binary_path('yql/essentials/tests/common/test_framework/udfs_deps') + udfs_dir=yql_binary_path('yql/essentials/tests/common/test_framework/udfs_deps'), + langver=langver ) res, tables_res = execute( @@ -156,12 +157,14 @@ def run_file_no_cache(provider, suite, case, cfg, config, yql_http_file_server, def run_file(provider, suite, case, cfg, config, yql_http_file_server, yqlrun_binary=None, - extra_args=[], force_blocks=False, allow_llvm=True, data_path=None, run_sql=True, cfg_postprocess=None): + extra_args=[], force_blocks=False, allow_llvm=True, data_path=None, run_sql=True, + cfg_postprocess=None, langver=None): if (suite, case, cfg) not in run_file.cache: run_file.cache[(suite, case, cfg)] = \ run_file_no_cache(provider, suite, case, cfg, config, yql_http_file_server, yqlrun_binary, extra_args, force_blocks=force_blocks, allow_llvm=allow_llvm, - data_path=data_path, run_sql=run_sql, cfg_postprocess=cfg_postprocess) + data_path=data_path, run_sql=run_sql, cfg_postprocess=cfg_postprocess, + langver=langver) return run_file.cache[(suite, case, cfg)] diff --git a/yql/essentials/tests/common/test_framework/test_utils.py b/yql/essentials/tests/common/test_framework/test_utils.py index 0865245fd9b..e760e19a342 100644 --- a/yql/essentials/tests/common/test_framework/test_utils.py +++ b/yql/essentials/tests/common/test_framework/test_utils.py @@ -144,6 +144,7 @@ def validate_cfg(result): "yt_file", "os", "param", + "langver", ), "Unknown command in .cfg: %s" % (r[0]) diff --git a/yql/essentials/tests/common/test_framework/yql_utils.py b/yql/essentials/tests/common/test_framework/yql_utils.py index 3e4a4afa3fe..2d59a1fa13f 100644 --- a/yql/essentials/tests/common/test_framework/yql_utils.py +++ b/yql/essentials/tests/common/test_framework/yql_utils.py @@ -496,6 +496,13 @@ def is_xfail(cfg): return False +def get_langver(cfg): + for item in cfg: + if item[0] == 'langver': + return item[1] + return None + + def is_skip_forceblocks(cfg): for item in cfg: if item[0] == 'skip_forceblocks': diff --git a/yql/essentials/tests/common/test_framework/yqlrun.py b/yql/essentials/tests/common/test_framework/yqlrun.py index e37bf7c38d3..49086ba7e88 100644 --- a/yql/essentials/tests/common/test_framework/yqlrun.py +++ b/yql/essentials/tests/common/test_framework/yqlrun.py @@ -25,7 +25,8 @@ FIX_DIR_PREFIXES = { class YQLRun(object): - def __init__(self, udfs_dir=None, prov='yt', use_sql2yql=False, keep_temp=True, binary=None, gateway_config=None, fs_config=None, extra_args=[], cfg_dir=None, support_udfs=True): + def __init__(self, udfs_dir=None, prov='yt', use_sql2yql=False, keep_temp=True, binary=None, gateway_config=None, + fs_config=None, extra_args=[], cfg_dir=None, support_udfs=True, langver=None): if binary is None: self.yqlrun_binary = yql_utils.yql_binary_path(os.getenv('YQL_YQLRUN_PATH') or 'yql/tools/yqlrun/yqlrun') else: @@ -80,6 +81,8 @@ class YQLRun(object): flags = yql_utils.get_param('SQL_FLAGS').split(',') self.gateway_config.SqlCore.TranslationFlags.extend(flags) + self.langver = langver + def yql_exec(self, program=None, program_file=None, files=None, urls=None, run_sql=False, verbose=False, check_error=True, tables=None, pretty_plan=True, wait=True, parameters={}, extra_env={}, require_udf_resolver=False, scan_udfs=True): @@ -173,6 +176,9 @@ class YQLRun(object): if ansi_lexer: cmd += '--ansi-lexer ' + if self.langver is not None: + cmd += '--langver=%s ' % (self.langver,) + if self.keep_temp and prov != 'pure': cmd += '--keep-temp ' diff --git a/yql/essentials/tests/s-expressions/minirun/pure.py b/yql/essentials/tests/s-expressions/minirun/pure.py index 38318d9fb8a..576230fd00b 100644 --- a/yql/essentials/tests/s-expressions/minirun/pure.py +++ b/yql/essentials/tests/s-expressions/minirun/pure.py @@ -9,12 +9,13 @@ from yql_utils import execute, get_tables, get_files, get_http_files, \ KSV_ATTR, yql_binary_path, is_xfail, is_canonize_peephole, is_peephole_use_blocks, is_canonize_lineage, \ is_skip_forceblocks, get_param, normalize_source_code_path, replace_vals, get_gateway_cfg_suffix, \ do_custom_query_check, stable_result_file, stable_table_file, is_with_final_result_issues, \ - normalize_result + normalize_result, get_langver from yqlrun import YQLRun from test_utils import get_config, get_parameters_json from test_file_common import run_file, run_file_no_cache, get_gateways_config, get_sql_query +DEFAULT_LANG_VER = '2025.01' ASTDIFF_PATH = yql_binary_path('yql/essentials/tools/astdiff/astdiff') MINIRUN_PATH = yql_binary_path('yql/essentials/tools/minirun/minirun') DATA_PATH = yatest.common.source_path('yql/essentials/tests/s-expressions/suites') @@ -25,6 +26,9 @@ def run_test(suite, case, cfg, tmpdir, what, yql_http_file_server): pytest.skip('non-trivial gateways.conf') config = get_config(suite, case, cfg, data_path=DATA_PATH) + langver = get_langver(config) + if langver is None: + langver = DEFAULT_LANG_VER xfail = is_xfail(config) if xfail and what != 'Results': @@ -38,7 +42,8 @@ def run_test(suite, case, cfg, tmpdir, what, yql_http_file_server): if is_with_final_result_issues(config): extra_final_args += ['--with-final-issues'] (res, tables_res) = run_file('pure', suite, case, cfg, config, yql_http_file_server, MINIRUN_PATH, - extra_args=extra_final_args, allow_llvm=False, data_path=DATA_PATH, run_sql=False) + extra_args=extra_final_args, allow_llvm=False, data_path=DATA_PATH, + run_sql=False, langver=langver) to_canonize = [] assert not tables_res @@ -72,7 +77,8 @@ def run_test(suite, case, cfg, tmpdir, what, yql_http_file_server): keep_temp=False, gateway_config=get_gateways_config(http_files, yql_http_file_server, allow_llvm=is_llvm), udfs_dir=yql_binary_path('yql/essentials/tests/common/test_framework/udfs_deps'), - binary=MINIRUN_PATH + binary=MINIRUN_PATH, + langver=langver ) opt_res, opt_tables_res = execute( diff --git a/yql/essentials/tests/sql/minirun/part5/canondata/result.json b/yql/essentials/tests/sql/minirun/part5/canondata/result.json index 579677b9314..ec17a6eec0b 100644 --- a/yql/essentials/tests/sql/minirun/part5/canondata/result.json +++ b/yql/essentials/tests/sql/minirun/part5/canondata/result.json @@ -237,6 +237,20 @@ "uri": "https://{canondata_backend}/1871002/ab54d2c5acdb4e70fca2cf294e5ea9c225baab0c/resource.tar.gz#test.test_aggr_factory-transform_output-default.txt-Results_/results.txt" } ], + "test.test[aggregate-group_by_expr_after_where_ver--Debug]": [ + { + "checksum": "d0da2a1dc674dc86930994f910eb1c2b", + "size": 253, + "uri": "https://{canondata_backend}/1130705/42827d049e4963da219fd249860b757d672765ec/resource.tar.gz#test.test_aggregate-group_by_expr_after_where_ver--Debug_/opt.yql" + } + ], + "test.test[aggregate-group_by_expr_after_where_ver--Results]": [ + { + "checksum": "1578f70c73d3ed10cf371be4ce4e70ec", + "size": 690, + "uri": "https://{canondata_backend}/1130705/42827d049e4963da219fd249860b757d672765ec/resource.tar.gz#test.test_aggregate-group_by_expr_after_where_ver--Results_/results.txt" + } + ], "test.test[aggregate-hopping-default.txt-Debug]": [ { "checksum": "bb23bcbf33c639f7faf92653dd668f3c", @@ -1044,6 +1058,20 @@ "uri": "https://{canondata_backend}/1936273/19f08c34eba9366d29ee0ffb8eb99e637c34fd97/resource.tar.gz#test.test_join-convert_check_key_mem2-default.txt-Results_/results.txt" } ], + "test.test[join-prune_keys-default.txt-Debug]": [ + { + "checksum": "a706e6bd4285c96ad5120d701109fc82", + "size": 3669, + "uri": "https://{canondata_backend}/1871102/392832e505c55eb371c9d3241b89c96b5a837c8f/resource.tar.gz#test.test_join-prune_keys-default.txt-Debug_/opt.yql" + } + ], + "test.test[join-prune_keys-default.txt-Results]": [ + { + "checksum": "af076a3334031b8cdaa6969f064a4616", + "size": 18160, + "uri": "https://{canondata_backend}/1130705/620da5a4f19baef17c32a4b3c699ec3c3091ada5/resource.tar.gz#test.test_join-prune_keys-default.txt-Results_/results.txt" + } + ], "test.test[json-json_exists/common_syntax-default.txt-Debug]": [ { "checksum": "1559e7b19e1d1827f8f1ea62929effb1", diff --git a/yql/essentials/tests/sql/minirun/part6/canondata/result.json b/yql/essentials/tests/sql/minirun/part6/canondata/result.json index 26ca736cfcf..60e9fee9459 100644 --- a/yql/essentials/tests/sql/minirun/part6/canondata/result.json +++ b/yql/essentials/tests/sql/minirun/part6/canondata/result.json @@ -195,6 +195,20 @@ "uri": "https://{canondata_backend}/1936273/614fe8dff439fd011c07c47361f2a1d0d854297f/resource.tar.gz#test.test_aggregate-distinct_over_keys-default.txt-Results_/results.txt" } ], + "test.test[aggregate-group_by_expr_after_where-default.txt-Debug]": [ + { + "checksum": "d0da2a1dc674dc86930994f910eb1c2b", + "size": 253, + "uri": "https://{canondata_backend}/1925842/9a344928381729abc8381a7e8ada7e10e2ba51fe/resource.tar.gz#test.test_aggregate-group_by_expr_after_where-default.txt-Debug_/opt.yql" + } + ], + "test.test[aggregate-group_by_expr_after_where-default.txt-Results]": [ + { + "checksum": "1578f70c73d3ed10cf371be4ce4e70ec", + "size": 690, + "uri": "https://{canondata_backend}/1925842/9a344928381729abc8381a7e8ada7e10e2ba51fe/resource.tar.gz#test.test_aggregate-group_by_expr_after_where-default.txt-Results_/results.txt" + } + ], "test.test[ansi_idents-escaping-default.txt-Debug]": [ { "checksum": "13dd30dd58fd993aa21441bec427f12b", diff --git a/yql/essentials/tests/sql/minirun/part7/canondata/result.json b/yql/essentials/tests/sql/minirun/part7/canondata/result.json index e8c9926ae75..8083b15dac8 100644 --- a/yql/essentials/tests/sql/minirun/part7/canondata/result.json +++ b/yql/essentials/tests/sql/minirun/part7/canondata/result.json @@ -1232,16 +1232,16 @@ ], "test.test[select-prune_keys-default.txt-Debug]": [ { - "checksum": "95e58e469ce10fce0d0d5e55c0cf3baf", - "size": 3292, - "uri": "https://{canondata_backend}/1931696/04008bc01ad4f562f8e03ad2bc296f7a64a78489/resource.tar.gz#test.test_select-prune_keys-default.txt-Debug_/opt.yql" + "checksum": "1dc9319da10b4c1343b6ceb5d0abb5b7", + "size": 3788, + "uri": "https://{canondata_backend}/212715/392992a262a39acb9e6e49f104e5800a3e731eb6/resource.tar.gz#test.test_select-prune_keys-default.txt-Debug_/opt.yql" } ], "test.test[select-prune_keys-default.txt-Results]": [ { - "checksum": "f53b6976b6a5e1ee5e1d331c25963930", - "size": 27670, - "uri": "https://{canondata_backend}/1931696/04008bc01ad4f562f8e03ad2bc296f7a64a78489/resource.tar.gz#test.test_select-prune_keys-default.txt-Results_/results.txt" + "checksum": "46c226ab1c00ed4a9a5b57c6639f15b2", + "size": 30306, + "uri": "https://{canondata_backend}/212715/392992a262a39acb9e6e49f104e5800a3e731eb6/resource.tar.gz#test.test_select-prune_keys-default.txt-Results_/results.txt" } ], "test.test[union-union_positional_mix-default.txt-Debug]": [ diff --git a/yql/essentials/tests/sql/minirun/pure.py b/yql/essentials/tests/sql/minirun/pure.py index c3a78adf354..3ec17bb2fa0 100644 --- a/yql/essentials/tests/sql/minirun/pure.py +++ b/yql/essentials/tests/sql/minirun/pure.py @@ -9,12 +9,13 @@ from yql_utils import execute, get_tables, get_files, get_http_files, \ KSV_ATTR, yql_binary_path, is_xfail, is_canonize_peephole, is_peephole_use_blocks, is_canonize_lineage, \ is_skip_forceblocks, get_param, normalize_source_code_path, replace_vals, get_gateway_cfg_suffix, \ do_custom_query_check, stable_result_file, stable_table_file, is_with_final_result_issues, \ - normalize_result + normalize_result, get_langver from yqlrun import YQLRun from test_utils import get_config, get_parameters_json from test_file_common import run_file, run_file_no_cache, get_gateways_config, get_sql_query +DEFAULT_LANG_VER = '2025.01' DATA_PATH = yatest.common.source_path('yql/essentials/tests/sql/suites') ASTDIFF_PATH = yql_binary_path('yql/essentials/tools/astdiff/astdiff') MINIRUN_PATH = yql_binary_path('yql/essentials/tools/minirun/minirun') @@ -39,6 +40,10 @@ def run_test(suite, case, cfg, tmpdir, what, yql_http_file_server): config = get_config(suite, case, cfg, data_path = DATA_PATH) + langver = get_langver(config) + if langver is None: + langver = DEFAULT_LANG_VER + xfail = is_xfail(config) if xfail and what != 'Results': pytest.skip('xfail is not supported in this mode') @@ -51,7 +56,8 @@ def run_test(suite, case, cfg, tmpdir, what, yql_http_file_server): if is_with_final_result_issues(config): extra_final_args += ['--with-final-issues'] (res, tables_res) = run_file('pure', suite, case, cfg, config, yql_http_file_server, MINIRUN_PATH, - extra_args=extra_final_args, allow_llvm=False, data_path=DATA_PATH) + extra_args=extra_final_args, allow_llvm=False, data_path=DATA_PATH, + langver=langver) to_canonize = [] assert xfail or os.path.exists(res.results_file) @@ -67,7 +73,8 @@ def run_test(suite, case, cfg, tmpdir, what, yql_http_file_server): force_blocks = is_peephole_use_blocks(config) (res, tables_res) = run_file_no_cache('pure', suite, case, cfg, config, yql_http_file_server, force_blocks=force_blocks, extra_args=['--peephole'], - data_path=DATA_PATH, yqlrun_binary=MINIRUN_PATH) + data_path=DATA_PATH, yqlrun_binary=MINIRUN_PATH, + langver=langver) return [yatest.common.canonical_file(res.opt_file, diff_tool=ASTDIFF_PATH)] if what == 'Results': @@ -100,7 +107,8 @@ def run_test(suite, case, cfg, tmpdir, what, yql_http_file_server): keep_temp=False, gateway_config=get_gateways_config(http_files, yql_http_file_server, allow_llvm=is_llvm, force_blocks=is_blocks), udfs_dir=yql_binary_path('yql/essentials/tests/common/test_framework/udfs_deps'), - binary=MINIRUN_PATH + binary=MINIRUN_PATH, + langver=langver ) opt_res, opt_tables_res = execute( diff --git a/yql/essentials/tests/sql/sql2yql/canondata/result.json b/yql/essentials/tests/sql/sql2yql/canondata/result.json index 4cb7e6cc7af..1022b92f85c 100644 --- a/yql/essentials/tests/sql/sql2yql/canondata/result.json +++ b/yql/essentials/tests/sql/sql2yql/canondata/result.json @@ -937,6 +937,20 @@ "uri": "https://{canondata_backend}/1936273/e22f8123b51c2802f50d5a8d4626267f2f28e9ab/resource.tar.gz#test_sql2yql.test_aggregate-distinct_over_keys_/sql.yql" } ], + "test_sql2yql.test[aggregate-group_by_expr_after_where]": [ + { + "checksum": "4799645fef77850f5f5f07de2d1b8bc2", + "size": 1762, + "uri": "https://{canondata_backend}/1899731/49525280cc90ece19469c3347e616ee12710ec2c/resource.tar.gz#test_sql2yql.test_aggregate-group_by_expr_after_where_/sql.yql" + } + ], + "test_sql2yql.test[aggregate-group_by_expr_after_where_ver]": [ + { + "checksum": "d7d81aab522ef18bb07e619c426594c5", + "size": 1762, + "uri": "https://{canondata_backend}/1597364/99b2cf59a9975dbc2994ead01aa9dcbd784b5279/resource.tar.gz#test_sql2yql.test_aggregate-group_by_expr_after_where_ver_/sql.yql" + } + ], "test_sql2yql.test[aggregate-group_by_rollup_rename]": [ { "checksum": "bc5b27508587d82ba3e9d0a752d25dcc", @@ -3933,6 +3947,13 @@ "uri": "https://{canondata_backend}/1942173/99e88108149e222741552e7e6cddef041d6a2846/resource.tar.gz#test_sql2yql.test_join-left_join_with_self_aggr_/sql.yql" } ], + "test_sql2yql.test[join-prune_keys]": [ + { + "checksum": "a04490e5ef3a567ca13cb1ed88272bd8", + "size": 22535, + "uri": "https://{canondata_backend}/1599023/2a161150407124ac83c2566a6542b19a53abfccb/resource.tar.gz#test_sql2yql.test_join-prune_keys_/sql.yql" + } + ], "test_sql2yql.test[join-yql-19192]": [ { "checksum": "fffdf1cbb40643da9daf9bdf3edec121", @@ -7022,9 +7043,9 @@ ], "test_sql2yql.test[select-prune_keys]": [ { - "checksum": "55346f77548ef19f9a09d2f1d3f6f466", - "size": 17765, - "uri": "https://{canondata_backend}/1871182/906a4c4e540bb8746f8d7595500d4d1c9f664846/resource.tar.gz#test_sql2yql.test_select-prune_keys_/sql.yql" + "checksum": "f7da5706622461ab177712e6c348c61b", + "size": 19536, + "uri": "https://{canondata_backend}/1814674/c8d78993e8e9976f1e3fae2197140afe33195365/resource.tar.gz#test_sql2yql.test_select-prune_keys_/sql.yql" } ], "test_sql2yql.test[select-result_label]": [ @@ -8103,6 +8124,16 @@ "uri": "file://test_sql_format.test_aggregate-distinct_over_keys_/formatted.sql" } ], + "test_sql_format.test[aggregate-group_by_expr_after_where]": [ + { + "uri": "file://test_sql_format.test_aggregate-group_by_expr_after_where_/formatted.sql" + } + ], + "test_sql_format.test[aggregate-group_by_expr_after_where_ver]": [ + { + "uri": "file://test_sql_format.test_aggregate-group_by_expr_after_where_ver_/formatted.sql" + } + ], "test_sql_format.test[aggregate-group_by_rollup_rename]": [ { "uri": "file://test_sql_format.test_aggregate-group_by_rollup_rename_/formatted.sql" @@ -10243,6 +10274,11 @@ "uri": "file://test_sql_format.test_join-left_join_with_self_aggr_/formatted.sql" } ], + "test_sql_format.test[join-prune_keys]": [ + { + "uri": "file://test_sql_format.test_join-prune_keys_/formatted.sql" + } + ], "test_sql_format.test[join-yql-19192]": [ { "uri": "file://test_sql_format.test_join-yql-19192_/formatted.sql" diff --git a/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_aggregate-group_by_expr_after_where_/formatted.sql b/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_aggregate-group_by_expr_after_where_/formatted.sql new file mode 100644 index 00000000000..cb490bcb76c --- /dev/null +++ b/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_aggregate-group_by_expr_after_where_/formatted.sql @@ -0,0 +1,13 @@ +PRAGMA GroupByExprAfterWhere; + +SELECT + x +FROM ( + SELECT + 1 AS x +) +WHERE + x == 1 +GROUP BY + -x AS x +; diff --git a/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_aggregate-group_by_expr_after_where_ver_/formatted.sql b/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_aggregate-group_by_expr_after_where_ver_/formatted.sql new file mode 100644 index 00000000000..20d25d53d24 --- /dev/null +++ b/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_aggregate-group_by_expr_after_where_ver_/formatted.sql @@ -0,0 +1,11 @@ +SELECT + x +FROM ( + SELECT + 1 AS x +) +WHERE + x == 1 +GROUP BY + -x AS x +; diff --git a/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_join-prune_keys_/formatted.sql b/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_join-prune_keys_/formatted.sql new file mode 100644 index 00000000000..0f53bcbe441 --- /dev/null +++ b/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_join-prune_keys_/formatted.sql @@ -0,0 +1,238 @@ +PRAGMA config.flags('OptimizerFlags', 'EmitPruneKeys'); + +$a = ( + SELECT + * + FROM + as_table([ + <|x: 1, t: 1|>, + <|x: 1, t: 1|>, + <|x: 1, t: 2|>, + <|x: 3, t: 1|>, + <|x: 3, t: 4|>, + <|x: 3, t: 2|>, + ]) +); + +$b = ( + SELECT + * + FROM + as_table([ + <|x: 1, y: 1|>, + <|x: 1, y: 2|>, + <|x: 1, y: 3|>, + <|x: 1, y: 3|>, + <|x: 2, y: 3|>, + <|x: 2, y: 4|>, + ]) +); + +$c = ( + SELECT + * + FROM + as_table([ + <|x: 1|>, + <|x: 1|>, + <|x: 1|>, + <|x: 1|>, + <|x: 2|>, + <|x: 2|>, + ]) +); + +-- PruneKeys +SELECT + a.* +FROM + $a AS a +WHERE + a.x IN ( + SELECT + x + FROM + $b + ) +; -- PruneKeys + +SELECT + a.* +FROM + $a AS a +WHERE + a.x IN ( + SELECT + /*+ distinct(x) */ x + FROM + $b + ) +; -- nothing + +SELECT + a.* +FROM + $a AS a +WHERE + a.x IN ( + SELECT + x + FROM + $c + ) +; -- PruneKeys + +SELECT + a.* +FROM + $a AS a +LEFT SEMI JOIN + $b AS b +ON + a.x == b.x +; -- PruneKeys(b) + +SELECT + a.* +FROM + $b AS b +RIGHT SEMI JOIN + $a AS a +ON + b.x == a.x +; -- PruneKeys(b) + +SELECT + a.x, + a.t, + b.x +FROM ANY + $a AS a +JOIN + $b AS b +ON + a.x == b.x +; -- PruneKeys(a) + +SELECT + a.x, + a.t, + b.x +FROM + $a AS a +JOIN ANY + $b AS b +ON + a.x == b.x +; -- PruneKeys(b) + +$a_sorted = ( + SELECT + * + FROM + $a + ASSUME ORDER BY + x +); + +$b_sorted = ( + SELECT + * + FROM + $b + ASSUME ORDER BY + x +); + +$c_sorted = ( + SELECT + * + FROM + $c + ASSUME ORDER BY + x +); + +-- PruneAdjacentKeys +SELECT + a.* +FROM + $a AS a +WHERE + a.x IN ( + SELECT + x + FROM + $b_sorted + ) +; -- PruneAdjacentKeys + +SELECT + a.* +FROM + $a AS a +WHERE + a.x IN ( + SELECT + /*+ distinct(x) */ x + FROM + $b_sorted + ) +; -- nothing + +SELECT + a.* +FROM + $a AS a +WHERE + a.x IN ( + SELECT + x + FROM + $c_sorted + ) +; -- PruneAdjacentKeys + +SELECT + a.* +FROM + $a AS a +LEFT SEMI JOIN + $b_sorted AS b +ON + a.x == b.x +; -- PruneAdjacentKeys(b_sorted) + +SELECT + a.* +FROM + $b_sorted AS b +RIGHT SEMI JOIN + $a AS a +ON + b.x == a.x +; -- PruneAdjacentKeys(b_sorted) + +SELECT + a.x, + a.t, + b.x +FROM ANY + $a_sorted AS a +JOIN + $b AS b +ON + a.x == b.x +; -- PruneAdjacentKeys(a_sorted) + +SELECT + a.x, + a.t, + b.x +FROM + $a AS a +JOIN ANY + $b_sorted AS b +ON + a.x == b.x +; -- PruneAdjacentKeys(b_sorted) diff --git a/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_select-prune_keys_/formatted.sql b/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_select-prune_keys_/formatted.sql index 65a0e7c3a79..80c2fa06403 100644 --- a/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_select-prune_keys_/formatted.sql +++ b/yql/essentials/tests/sql/sql2yql/canondata/test_sql_format.test_select-prune_keys_/formatted.sql @@ -29,6 +29,14 @@ SELECT ListLength(Yql::PruneKeys(AsList(1, 1, 1, 3, 3, 3, 3), $mod2)) ; +SELECT + Yql::PruneAdjacentKeys(AsList(NULL, NULL, NULL, 1, 1, 2, 3, 3, 4, 5), $id) +; + +SELECT + Yql::PruneKeys(AsList(1, NULL, 1, NULL, 1, NULL, 1), $id) +; + -- optimize tests $get_a = ($x) -> { RETURN <|a: $x.a|>; diff --git a/yql/essentials/tests/sql/suites/aggregate/group_by_expr_after_where.sql b/yql/essentials/tests/sql/suites/aggregate/group_by_expr_after_where.sql new file mode 100644 index 00000000000..8f23c1f49cd --- /dev/null +++ b/yql/essentials/tests/sql/suites/aggregate/group_by_expr_after_where.sql @@ -0,0 +1,4 @@ +pragma GroupByExprAfterWhere; +select x from (select 1 as x) +where x = 1 +group by -x as x diff --git a/yql/essentials/tests/sql/suites/aggregate/group_by_expr_after_where_ver.cfg b/yql/essentials/tests/sql/suites/aggregate/group_by_expr_after_where_ver.cfg new file mode 100644 index 00000000000..367bc6a9ec0 --- /dev/null +++ b/yql/essentials/tests/sql/suites/aggregate/group_by_expr_after_where_ver.cfg @@ -0,0 +1 @@ +langver 2025.02 diff --git a/yql/essentials/tests/sql/suites/aggregate/group_by_expr_after_where_ver.sql b/yql/essentials/tests/sql/suites/aggregate/group_by_expr_after_where_ver.sql new file mode 100644 index 00000000000..e0542689369 --- /dev/null +++ b/yql/essentials/tests/sql/suites/aggregate/group_by_expr_after_where_ver.sql @@ -0,0 +1,3 @@ +select x from (select 1 as x) +where x = 1 +group by -x as x diff --git a/yql/essentials/tests/sql/suites/join/prune_keys.sql b/yql/essentials/tests/sql/suites/join/prune_keys.sql new file mode 100644 index 00000000000..9f10de9ace7 --- /dev/null +++ b/yql/essentials/tests/sql/suites/join/prune_keys.sql @@ -0,0 +1,50 @@ +pragma config.flags('OptimizerFlags', 'EmitPruneKeys'); + +$a = select * from as_table([ + <|x:1, t:1|>, + <|x:1, t:1|>, + <|x:1, t:2|>, + <|x:3, t:1|>, + <|x:3, t:4|>, + <|x:3, t:2|>, + ]); + +$b = select * from as_table([ + <|x:1, y:1|>, + <|x:1, y:2|>, + <|x:1, y:3|>, + <|x:1, y:3|>, + <|x:2, y:3|>, + <|x:2, y:4|>, + ]); + +$c = select * from as_table([ + <|x:1|>, + <|x:1|>, + <|x:1|>, + <|x:1|>, + <|x:2|>, + <|x:2|>, + ]); + +-- PruneKeys +select a.* from $a as a where a.x in (select x from $b); -- PruneKeys +select a.* from $a as a where a.x in (select /*+ distinct(x) */ x from $b); -- nothing +select a.* from $a as a where a.x in (select x from $c); -- PruneKeys +select a.* from $a as a left semi join $b as b on a.x = b.x; -- PruneKeys(b) +select a.* from $b as b right semi join $a as a on b.x = a.x; -- PruneKeys(b) +select a.x, a.t, b.x from any $a as a join $b as b on a.x == b.x; -- PruneKeys(a) +select a.x, a.t, b.x from $a as a join any $b as b on a.x == b.x; -- PruneKeys(b) + +$a_sorted = select * from $a assume order by x; +$b_sorted = select * from $b assume order by x; +$c_sorted = select * from $c assume order by x; + +-- PruneAdjacentKeys +select a.* from $a as a where a.x in (select x from $b_sorted); -- PruneAdjacentKeys +select a.* from $a as a where a.x in (select /*+ distinct(x) */ x from $b_sorted); -- nothing +select a.* from $a as a where a.x in (select x from $c_sorted); -- PruneAdjacentKeys +select a.* from $a as a left semi join $b_sorted as b on a.x = b.x; -- PruneAdjacentKeys(b_sorted) +select a.* from $b_sorted as b right semi join $a as a on b.x = a.x; -- PruneAdjacentKeys(b_sorted) +select a.x, a.t, b.x from any $a_sorted as a join $b as b on a.x == b.x; -- PruneAdjacentKeys(a_sorted) +select a.x, a.t, b.x from $a as a join any $b_sorted as b on a.x == b.x; -- PruneAdjacentKeys(b_sorted) diff --git a/yql/essentials/tests/sql/suites/select/prune_keys.sql b/yql/essentials/tests/sql/suites/select/prune_keys.sql index 9cf5cb3ced7..f541bcfc61d 100644 --- a/yql/essentials/tests/sql/suites/select/prune_keys.sql +++ b/yql/essentials/tests/sql/suites/select/prune_keys.sql @@ -11,6 +11,9 @@ SELECT Yql::PruneKeys([], $id); $mod2 = ($x) -> { RETURN $x % 2; }; SELECT ListLength(Yql::PruneKeys(AsList(1,1,1,3,3,3,3), $mod2)); +SELECT Yql::PruneAdjacentKeys(AsList(null,null,null,1,1,2,3,3,4,5), $id); +SELECT Yql::PruneKeys(AsList(1,null,1,null,1,null,1), $id); + -- optimize tests $get_a = ($x) -> { RETURN <|a:$x.a|>; }; |