Author: gunther
Date: Thu Sep 11 19:02:53 2014
New Revision: 1624364
URL: http://svn.apache.org/r1624364
Log:
HIVE-8051: Some union queries fail with dynamic partition pruning on tez
(Gunther Hagleitner, reviewed by Vikram Dixit K)
Modified:
hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/parse/AppMasterEventProcessor.java
hive/trunk/ql/src/test/queries/clientpositive/dynamic_partition_pruning_2.q
hive/trunk/ql/src/test/results/clientpositive/tez/dynamic_partition_pruning_2.q.out
Modified:
hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/parse/AppMasterEventProcessor.java
URL:
http://svn.apache.org/viewvc/hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/parse/AppMasterEventProcessor.java?rev=1624364&r1=1624363&r2=1624364&view=diff
==============================================================================
---
hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/parse/AppMasterEventProcessor.java
(original)
+++
hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/parse/AppMasterEventProcessor.java
Thu Sep 11 19:02:53 2014
@@ -56,6 +56,7 @@ public class AppMasterEventProcessor imp
} else {
events = new ArrayList<AppMasterEventOperator>();
}
+ events.add(event);
context.tsToEventMap.put(desc.getTableScan(), events);
return true;
}
Modified:
hive/trunk/ql/src/test/queries/clientpositive/dynamic_partition_pruning_2.q
URL:
http://svn.apache.org/viewvc/hive/trunk/ql/src/test/queries/clientpositive/dynamic_partition_pruning_2.q?rev=1624364&r1=1624363&r2=1624364&view=diff
==============================================================================
--- hive/trunk/ql/src/test/queries/clientpositive/dynamic_partition_pruning_2.q
(original)
+++ hive/trunk/ql/src/test/queries/clientpositive/dynamic_partition_pruning_2.q
Thu Sep 11 19:02:53 2014
@@ -39,3 +39,12 @@ and
d1.label in ('foo', 'bar')
GROUP BY d1.label
ORDER BY d1.label;
+
+EXPLAIN
+SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id AND label = 'foo'
+UNION ALL
+SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id AND label = 'bar';
+
+SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id AND label = 'foo'
+UNION ALL
+SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id AND label = 'bar';
\ No newline at end of file
Modified:
hive/trunk/ql/src/test/results/clientpositive/tez/dynamic_partition_pruning_2.q.out
URL:
http://svn.apache.org/viewvc/hive/trunk/ql/src/test/results/clientpositive/tez/dynamic_partition_pruning_2.q.out?rev=1624364&r1=1624363&r2=1624364&view=diff
==============================================================================
---
hive/trunk/ql/src/test/results/clientpositive/tez/dynamic_partition_pruning_2.q.out
(original)
+++
hive/trunk/ql/src/test/results/clientpositive/tez/dynamic_partition_pruning_2.q.out
Thu Sep 11 19:02:53 2014
@@ -267,3 +267,176 @@ POSTHOOK: Input: default@dim_shops
#### A masked pattern was here ####
bar 3 15
foo 3 6
+PREHOOK: query: EXPLAIN
+SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id AND label = 'foo'
+UNION ALL
+SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id AND label = 'bar'
+PREHOOK: type: QUERY
+POSTHOOK: query: EXPLAIN
+SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id AND label = 'foo'
+UNION ALL
+SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id AND label = 'bar'
+POSTHOOK: type: QUERY
+STAGE DEPENDENCIES:
+ Stage-1 is a root stage
+ Stage-0 depends on stages: Stage-1
+
+STAGE PLANS:
+ Stage: Stage-1
+ Tez
+ Edges:
+ Map 3 <- Map 1 (BROADCAST_EDGE), Union 4 (CONTAINS)
+ Map 5 <- Map 2 (BROADCAST_EDGE), Union 4 (CONTAINS)
+#### A masked pattern was here ####
+ Vertices:
+ Map 1
+ Map Operator Tree:
+ TableScan
+ alias: dim_shops
+ filterExpr: (id is not null and (label = 'bar')) (type:
boolean)
+ Statistics: Num rows: 2 Data size: 18 Basic stats: COMPLETE
Column stats: NONE
+ Filter Operator
+ predicate: (id is not null and (label = 'bar')) (type:
boolean)
+ Statistics: Num rows: 0 Data size: 0 Basic stats: NONE
Column stats: NONE
+ Reduce Output Operator
+ key expressions: id (type: int)
+ sort order: +
+ Map-reduce partition columns: id (type: int)
+ Statistics: Num rows: 0 Data size: 0 Basic stats: NONE
Column stats: NONE
+ Select Operator
+ expressions: id (type: int)
+ outputColumnNames: _col0
+ Statistics: Num rows: 0 Data size: 0 Basic stats: NONE
Column stats: NONE
+ Group By Operator
+ keys: _col0 (type: int)
+ mode: hash
+ outputColumnNames: _col0
+ Statistics: Num rows: 0 Data size: 0 Basic stats: NONE
Column stats: NONE
+ Dynamic Partitioning Event Operator
+ Target Input: agg_01
+ Partition key expr: dim_shops_id
+ Statistics: Num rows: 0 Data size: 0 Basic stats:
NONE Column stats: NONE
+ Target column: dim_shops_id
+ Target Vertex: Map 3
+ Map 2
+ Map Operator Tree:
+ TableScan
+ alias: dim_shops
+ filterExpr: (id is not null and (label = 'foo')) (type:
boolean)
+ Statistics: Num rows: 2 Data size: 18 Basic stats: COMPLETE
Column stats: NONE
+ Filter Operator
+ predicate: (id is not null and (label = 'foo')) (type:
boolean)
+ Statistics: Num rows: 0 Data size: 0 Basic stats: NONE
Column stats: NONE
+ Reduce Output Operator
+ key expressions: id (type: int)
+ sort order: +
+ Map-reduce partition columns: id (type: int)
+ Statistics: Num rows: 0 Data size: 0 Basic stats: NONE
Column stats: NONE
+ Select Operator
+ expressions: id (type: int)
+ outputColumnNames: _col0
+ Statistics: Num rows: 0 Data size: 0 Basic stats: NONE
Column stats: NONE
+ Group By Operator
+ keys: _col0 (type: int)
+ mode: hash
+ outputColumnNames: _col0
+ Statistics: Num rows: 0 Data size: 0 Basic stats: NONE
Column stats: NONE
+ Dynamic Partitioning Event Operator
+ Target Input: agg_01
+ Partition key expr: dim_shops_id
+ Statistics: Num rows: 0 Data size: 0 Basic stats:
NONE Column stats: NONE
+ Target column: dim_shops_id
+ Target Vertex: Map 5
+ Map 3
+ Map Operator Tree:
+ TableScan
+ alias: agg_01
+ filterExpr: dim_shops_id is not null (type: boolean)
+ Map Join Operator
+ condition map:
+ Inner Join 0 to 1
+ condition expressions:
+ 0 {amount} {dim_shops_id}
+ 1 {id}
+ keys:
+ 0 dim_shops_id (type: int)
+ 1 id (type: int)
+ outputColumnNames: _col0, _col1, _col5
+ Filter Operator
+ predicate: (_col1 = _col5) (type: boolean)
+ Select Operator
+ expressions: _col0 (type: decimal(10,0))
+ outputColumnNames: _col0
+ Select Operator
+ expressions: _col0 (type: decimal(10,0))
+ outputColumnNames: _col0
+ File Output Operator
+ compressed: false
+ table:
+ input format:
org.apache.hadoop.mapred.TextInputFormat
+ output format:
org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat
+ serde:
org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+ Map 5
+ Map Operator Tree:
+ TableScan
+ alias: agg_01
+ filterExpr: dim_shops_id is not null (type: boolean)
+ Map Join Operator
+ condition map:
+ Inner Join 0 to 1
+ condition expressions:
+ 0 {amount} {dim_shops_id}
+ 1 {id}
+ keys:
+ 0 dim_shops_id (type: int)
+ 1 id (type: int)
+ outputColumnNames: _col0, _col1, _col5
+ Filter Operator
+ predicate: (_col1 = _col5) (type: boolean)
+ Select Operator
+ expressions: _col0 (type: decimal(10,0))
+ outputColumnNames: _col0
+ Select Operator
+ expressions: _col0 (type: decimal(10,0))
+ outputColumnNames: _col0
+ File Output Operator
+ compressed: false
+ table:
+ input format:
org.apache.hadoop.mapred.TextInputFormat
+ output format:
org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat
+ serde:
org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe
+ Union 4
+ Vertex: Union 4
+
+ Stage: Stage-0
+ Fetch Operator
+ limit: -1
+ Processor Tree:
+ ListSink
+
+PREHOOK: query: SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id
AND label = 'foo'
+UNION ALL
+SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id AND label = 'bar'
+PREHOOK: type: QUERY
+PREHOOK: Input: default@agg_01
+PREHOOK: Input: default@agg_01@dim_shops_id=1
+PREHOOK: Input: default@agg_01@dim_shops_id=2
+PREHOOK: Input: default@agg_01@dim_shops_id=3
+PREHOOK: Input: default@dim_shops
+#### A masked pattern was here ####
+POSTHOOK: query: SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id
AND label = 'foo'
+UNION ALL
+SELECT amount FROM agg_01, dim_shops WHERE dim_shops_id = id AND label = 'bar'
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@agg_01
+POSTHOOK: Input: default@agg_01@dim_shops_id=1
+POSTHOOK: Input: default@agg_01@dim_shops_id=2
+POSTHOOK: Input: default@agg_01@dim_shops_id=3
+POSTHOOK: Input: default@dim_shops
+#### A masked pattern was here ####
+4
+5
+6
+1
+2
+3
