commit b4b43676df5dee35660524374b8ed45e108a5a8c Author: Janaki Lahorani Date: Sun Sep 16 18:17:16 2018 -0700 HIVE-20570: Fix plan for query with hive.optimize.union.remove set to true diff --git ql/src/java/org/apache/hadoop/hive/ql/optimizer/GenMRFileSink1.java ql/src/java/org/apache/hadoop/hive/ql/optimizer/GenMRFileSink1.java index 2f00c220649ecd846a2c08c31cab00cfa5aef290..25c6b24f46771ba12327851285af15ffb9b65bcd 100644 --- ql/src/java/org/apache/hadoop/hive/ql/optimizer/GenMRFileSink1.java +++ ql/src/java/org/apache/hadoop/hive/ql/optimizer/GenMRFileSink1.java @@ -116,7 +116,9 @@ public Object process(Node nd, Stack stack, NodeProcessorCtx opProcCtx, } FileSinkDesc fileSinkDesc = fsOp.getConf(); - if (fileSinkDesc.isLinkedFileSink()) { + // There are linked file sink operators and child tasks are present + if (fileSinkDesc.isLinkedFileSink() && (currTask.getChildTasks() != null) && + (currTask.getChildTasks().size() == 1)) { Map> linkedFileDescTasks = ctx.getLinkedFileDescTasks(); if (linkedFileDescTasks == null) { @@ -124,12 +126,8 @@ public Object process(Node nd, Stack stack, NodeProcessorCtx opProcCtx, ctx.setLinkedFileDescTasks(linkedFileDescTasks); } - // The child tasks may be null in case of a select - if ((currTask.getChildTasks() != null) && - (currTask.getChildTasks().size() == 1)) { - for (FileSinkDesc fileDesc : fileSinkDesc.getLinkedFileSinkDesc()) { - linkedFileDescTasks.put(fileDesc, currTask.getChildTasks().get(0)); - } + for (FileSinkDesc fileDesc : fileSinkDesc.getLinkedFileSinkDesc()) { + linkedFileDescTasks.put(fileDesc, currTask.getChildTasks().get(0)); } } diff --git ql/src/test/queries/clientpositive/union_remove_plan.q ql/src/test/queries/clientpositive/union_remove_plan.q new file mode 100644 index 0000000000000000000000000000000000000000..c8f1f5cf3afe79643c9f6dd0502300dadc31cc40 --- /dev/null +++ ql/src/test/queries/clientpositive/union_remove_plan.q @@ -0,0 +1,14 @@ + create table if not exists test_table(column1 string, column2 int); + insert into test_table values('a',1),('b',2); + + set hive.optimize.union.remove=true; + set mapred.input.dir.recursive=true; + + explain + select column1 from test_table group by column1 + union all + select column1 from test_table group by column1; + +select column1 from test_table group by column1 +union all +select column1 from test_table group by column1; diff --git ql/src/test/results/clientpositive/skewjoin_union_remove_1.q.out ql/src/test/results/clientpositive/skewjoin_union_remove_1.q.out index a7351df5d721d8d6630ce687e30d31117dd62544..5d0a6bb281e2d857912f082a5eedd48defd69424 100644 --- ql/src/test/results/clientpositive/skewjoin_union_remove_1.q.out +++ ql/src/test/results/clientpositive/skewjoin_union_remove_1.q.out @@ -43,7 +43,7 @@ POSTHOOK: type: QUERY STAGE DEPENDENCIES: Stage-1 is a root stage Stage-3 is a root stage - Stage-0 depends on stages: Stage-1 + Stage-0 depends on stages: Stage-1, Stage-3 STAGE PLANS: Stage: Stage-1 @@ -185,7 +185,7 @@ POSTHOOK: type: QUERY STAGE DEPENDENCIES: Stage-1 is a root stage Stage-3 is a root stage - Stage-0 depends on stages: Stage-1 + Stage-0 depends on stages: Stage-1, Stage-3 STAGE PLANS: Stage: Stage-1 diff --git ql/src/test/results/clientpositive/skewjoin_union_remove_2.q.out ql/src/test/results/clientpositive/skewjoin_union_remove_2.q.out index 31df384548d21f93466997c0df3b1025a6798ff6..0d467fe836998902ac2389e00926cb5405f48959 100644 --- ql/src/test/results/clientpositive/skewjoin_union_remove_2.q.out +++ ql/src/test/results/clientpositive/skewjoin_union_remove_2.q.out @@ -59,7 +59,7 @@ POSTHOOK: type: QUERY STAGE DEPENDENCIES: Stage-1 is a root stage Stage-4 is a root stage - Stage-0 depends on stages: Stage-1 + Stage-0 depends on stages: Stage-1, Stage-4 STAGE PLANS: Stage: Stage-1 diff --git ql/src/test/results/clientpositive/union_remove_plan.q.out ql/src/test/results/clientpositive/union_remove_plan.q.out new file mode 100644 index 0000000000000000000000000000000000000000..5fac5acc2db625cabdf7499b8f20d625da28f5d9 --- /dev/null +++ ql/src/test/results/clientpositive/union_remove_plan.q.out @@ -0,0 +1,126 @@ +PREHOOK: query: create table if not exists test_table(column1 string, column2 int) +PREHOOK: type: CREATETABLE +PREHOOK: Output: database:default +PREHOOK: Output: default@test_table +POSTHOOK: query: create table if not exists test_table(column1 string, column2 int) +POSTHOOK: type: CREATETABLE +POSTHOOK: Output: database:default +POSTHOOK: Output: default@test_table +PREHOOK: query: insert into test_table values('a',1),('b',2) +PREHOOK: type: QUERY +PREHOOK: Input: _dummy_database@_dummy_table +PREHOOK: Output: default@test_table +POSTHOOK: query: insert into test_table values('a',1),('b',2) +POSTHOOK: type: QUERY +POSTHOOK: Input: _dummy_database@_dummy_table +POSTHOOK: Output: default@test_table +POSTHOOK: Lineage: test_table.column1 SCRIPT [] +POSTHOOK: Lineage: test_table.column2 SCRIPT [] +PREHOOK: query: explain + select column1 from test_table group by column1 + union all + select column1 from test_table group by column1 +PREHOOK: type: QUERY +POSTHOOK: query: explain + select column1 from test_table group by column1 + union all + select column1 from test_table group by column1 +POSTHOOK: type: QUERY +STAGE DEPENDENCIES: + Stage-1 is a root stage + Stage-2 is a root stage + Stage-0 depends on stages: Stage-1, Stage-2 + +STAGE PLANS: + Stage: Stage-1 + Map Reduce + Map Operator Tree: + TableScan + alias: test_table + Statistics: Num rows: 2 Data size: 6 Basic stats: COMPLETE Column stats: NONE + Select Operator + expressions: column1 (type: string) + outputColumnNames: column1 + Statistics: Num rows: 2 Data size: 6 Basic stats: COMPLETE Column stats: NONE + Group By Operator + keys: column1 (type: string) + mode: hash + outputColumnNames: _col0 + Statistics: Num rows: 2 Data size: 6 Basic stats: COMPLETE Column stats: NONE + Reduce Output Operator + key expressions: _col0 (type: string) + sort order: + + Map-reduce partition columns: _col0 (type: string) + Statistics: Num rows: 2 Data size: 6 Basic stats: COMPLETE Column stats: NONE + Execution mode: vectorized + Reduce Operator Tree: + Group By Operator + keys: KEY._col0 (type: string) + mode: mergepartial + outputColumnNames: _col0 + Statistics: Num rows: 1 Data size: 3 Basic stats: COMPLETE Column stats: NONE + File Output Operator + compressed: false + Statistics: Num rows: 1 Data size: 3 Basic stats: COMPLETE Column stats: NONE + table: + input format: org.apache.hadoop.mapred.SequenceFileInputFormat + output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat + serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe + + Stage: Stage-2 + Map Reduce + Map Operator Tree: + TableScan + alias: test_table + Statistics: Num rows: 2 Data size: 6 Basic stats: COMPLETE Column stats: NONE + Select Operator + expressions: column1 (type: string) + outputColumnNames: column1 + Statistics: Num rows: 2 Data size: 6 Basic stats: COMPLETE Column stats: NONE + Group By Operator + keys: column1 (type: string) + mode: hash + outputColumnNames: _col0 + Statistics: Num rows: 2 Data size: 6 Basic stats: COMPLETE Column stats: NONE + Reduce Output Operator + key expressions: _col0 (type: string) + sort order: + + Map-reduce partition columns: _col0 (type: string) + Statistics: Num rows: 2 Data size: 6 Basic stats: COMPLETE Column stats: NONE + Execution mode: vectorized + Reduce Operator Tree: + Group By Operator + keys: KEY._col0 (type: string) + mode: mergepartial + outputColumnNames: _col0 + Statistics: Num rows: 1 Data size: 3 Basic stats: COMPLETE Column stats: NONE + File Output Operator + compressed: false + Statistics: Num rows: 1 Data size: 3 Basic stats: COMPLETE Column stats: NONE + table: + input format: org.apache.hadoop.mapred.SequenceFileInputFormat + output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat + serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe + + Stage: Stage-0 + Fetch Operator + limit: -1 + Processor Tree: + ListSink + +PREHOOK: query: select column1 from test_table group by column1 +union all +select column1 from test_table group by column1 +PREHOOK: type: QUERY +PREHOOK: Input: default@test_table +#### A masked pattern was here #### +POSTHOOK: query: select column1 from test_table group by column1 +union all +select column1 from test_table group by column1 +POSTHOOK: type: QUERY +POSTHOOK: Input: default@test_table +#### A masked pattern was here #### +a +b +a +b