You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hive.apache.org by ha...@apache.org on 2013/11/16 20:22:43 UTC

svn commit: r1542556 - in /hive/trunk/ql/src: java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java test/queries/clientpositive/vector_left_outer_join.q test/results/clientpositive/vector_left_outer_join.q.out

Author: hashutosh
Date: Sat Nov 16 19:22:42 2013
New Revision: 1542556

URL: http://svn.apache.org/r1542556
Log:
HIVE-5813 : Multi-way Left outer join fails in vectorized mode (Ashutosh Chauhan via Thejas Nair, Eric Hanson & Remus Rusanu)

Added:
    hive/trunk/ql/src/test/queries/clientpositive/vector_left_outer_join.q
    hive/trunk/ql/src/test/results/clientpositive/vector_left_outer_join.q.out
Modified:
    hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java

Modified: hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java
URL: http://svn.apache.org/viewvc/hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java?rev=1542556&r1=1542555&r2=1542556&view=diff
==============================================================================
--- hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java (original)
+++ hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/CommonJoinOperator.java Sat Nov 16 19:22:42 2013
@@ -434,7 +434,7 @@ public abstract class CommonJoinOperator
       }
     }
     if (forward) {
-      internalForward(forwardCache, null);
+      internalForward(forwardCache, outputObjInspector);
       countAfterReport = 0;
     }
   }

Added: hive/trunk/ql/src/test/queries/clientpositive/vector_left_outer_join.q
URL: http://svn.apache.org/viewvc/hive/trunk/ql/src/test/queries/clientpositive/vector_left_outer_join.q?rev=1542556&view=auto
==============================================================================
--- hive/trunk/ql/src/test/queries/clientpositive/vector_left_outer_join.q (added)
+++ hive/trunk/ql/src/test/queries/clientpositive/vector_left_outer_join.q Sat Nov 16 19:22:42 2013
@@ -0,0 +1,21 @@
+set hive.vectorized.execution.enabled=true;
+set hive.auto.convert.join=true;
+explain 
+select count(*) from (select c.ctinyint 
+from alltypesorc c
+left outer join alltypesorc cd
+  on cd.cint = c.cint 
+left outer join alltypesorc hd
+  on hd.ctinyint = c.ctinyint
+) t1
+;
+select count(*) from (select c.ctinyint
+from alltypesorc c
+left outer join alltypesorc cd
+  on cd.cint = c.cint 
+left outer join alltypesorc hd
+  on hd.ctinyint = c.ctinyint
+) t1;
+
+set hive.auto.convert.join=false;
+set hive.vectorized.execution.enabled=false;

Added: hive/trunk/ql/src/test/results/clientpositive/vector_left_outer_join.q.out
URL: http://svn.apache.org/viewvc/hive/trunk/ql/src/test/results/clientpositive/vector_left_outer_join.q.out?rev=1542556&view=auto
==============================================================================
--- hive/trunk/ql/src/test/results/clientpositive/vector_left_outer_join.q.out (added)
+++ hive/trunk/ql/src/test/results/clientpositive/vector_left_outer_join.q.out Sat Nov 16 19:22:42 2013
@@ -0,0 +1,156 @@
+PREHOOK: query: explain 
+select count(*) from (select c.ctinyint 
+from alltypesorc c
+left outer join alltypesorc cd
+  on cd.cint = c.cint 
+left outer join alltypesorc hd
+  on hd.ctinyint = c.ctinyint
+) t1
+PREHOOK: type: QUERY
+POSTHOOK: query: explain 
+select count(*) from (select c.ctinyint 
+from alltypesorc c
+left outer join alltypesorc cd
+  on cd.cint = c.cint 
+left outer join alltypesorc hd
+  on hd.ctinyint = c.ctinyint
+) t1
+POSTHOOK: type: QUERY
+ABSTRACT SYNTAX TREE:
+  (TOK_QUERY (TOK_FROM (TOK_SUBQUERY (TOK_QUERY (TOK_FROM (TOK_LEFTOUTERJOIN (TOK_LEFTOUTERJOIN (TOK_TABREF (TOK_TABNAME alltypesorc) c) (TOK_TABREF (TOK_TABNAME alltypesorc) cd) (= (. (TOK_TABLE_OR_COL cd) cint) (. (TOK_TABLE_OR_COL c) cint))) (TOK_TABREF (TOK_TABNAME alltypesorc) hd) (= (. (TOK_TABLE_OR_COL hd) ctinyint) (. (TOK_TABLE_OR_COL c) ctinyint)))) (TOK_INSERT (TOK_DESTINATION (TOK_DIR TOK_TMP_FILE)) (TOK_SELECT (TOK_SELEXPR (. (TOK_TABLE_OR_COL c) ctinyint))))) t1)) (TOK_INSERT (TOK_DESTINATION (TOK_DIR TOK_TMP_FILE)) (TOK_SELECT (TOK_SELEXPR (TOK_FUNCTIONSTAR count)))))
+
+STAGE DEPENDENCIES:
+  Stage-7 is a root stage
+  Stage-2 depends on stages: Stage-7
+  Stage-0 is a root stage
+
+STAGE PLANS:
+  Stage: Stage-7
+    Map Reduce Local Work
+      Alias -> Map Local Tables:
+        t1:cd 
+          Fetch Operator
+            limit: -1
+        t1:hd 
+          Fetch Operator
+            limit: -1
+      Alias -> Map Local Operator Tree:
+        t1:cd 
+          TableScan
+            alias: cd
+            HashTable Sink Operator
+              condition expressions:
+                0 {ctinyint}
+                1 
+              handleSkewJoin: false
+              keys:
+                0 [Column[cint]]
+                1 [Column[cint]]
+              Position of Big Table: 0
+        t1:hd 
+          TableScan
+            alias: hd
+            HashTable Sink Operator
+              condition expressions:
+                0 
+                1 
+              handleSkewJoin: false
+              keys:
+                0 [Column[_col0]]
+                1 [Column[ctinyint]]
+              Position of Big Table: 0
+
+  Stage: Stage-2
+    Map Reduce
+      Alias -> Map Operator Tree:
+        t1:c 
+          TableScan
+            alias: c
+            Map Join Operator
+              condition map:
+                   Left Outer Join0 to 1
+              condition expressions:
+                0 {ctinyint}
+                1 
+              handleSkewJoin: false
+              keys:
+                0 [Column[cint]]
+                1 [Column[cint]]
+              outputColumnNames: _col0
+              Position of Big Table: 0
+              Vectorized execution: true
+              Map Join Operator
+                condition map:
+                     Left Outer Join0 to 1
+                condition expressions:
+                  0 
+                  1 
+                handleSkewJoin: false
+                keys:
+                  0 [Column[_col0]]
+                  1 [Column[ctinyint]]
+                Position of Big Table: 0
+                Vectorized execution: true
+                Select Operator
+                  Vectorized execution: true
+                  Group By Operator
+                    aggregations:
+                          expr: count()
+                    bucketGroup: false
+                    mode: hash
+                    outputColumnNames: _col0
+                    Vectorized execution: true
+                    Reduce Output Operator
+                      sort order: 
+                      tag: -1
+                      value expressions:
+                            expr: _col0
+                            type: bigint
+      Local Work:
+        Map Reduce Local Work
+      Reduce Operator Tree:
+        Group By Operator
+          aggregations:
+                expr: count(VALUE._col0)
+          bucketGroup: false
+          mode: mergepartial
+          outputColumnNames: _col0
+          Select Operator
+            expressions:
+                  expr: _col0
+                  type: bigint
+            outputColumnNames: _col0
+            File Output Operator
+              compressed: false
+              GlobalTableId: 0
+              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
+
+  Stage: Stage-0
+    Fetch Operator
+      limit: -1
+
+
+PREHOOK: query: select count(*) from (select c.ctinyint
+from alltypesorc c
+left outer join alltypesorc cd
+  on cd.cint = c.cint 
+left outer join alltypesorc hd
+  on hd.ctinyint = c.ctinyint
+) t1
+PREHOOK: type: QUERY
+PREHOOK: Input: default@alltypesorc
+#### A masked pattern was here ####
+POSTHOOK: query: select count(*) from (select c.ctinyint
+from alltypesorc c
+left outer join alltypesorc cd
+  on cd.cint = c.cint 
+left outer join alltypesorc hd
+  on hd.ctinyint = c.ctinyint
+) t1
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@alltypesorc
+#### A masked pattern was here ####
+192735557