Author: rohini
Date: Mon Oct 26 18:35:32 2015
New Revision: 1710672

URL: http://svn.apache.org/viewvc?rev=1710672&view=rev
Log:
PIG-4712: [Pig on Tez] NPE in Bloom UDF after Union (rohini)

Modified:
    pig/trunk/CHANGES.txt
    
pig/trunk/src/org/apache/pig/backend/hadoop/executionengine/physicalLayer/expressionOperators/POUserFunc.java
    pig/trunk/test/e2e/pig/tests/nightly.conf

Modified: pig/trunk/CHANGES.txt
URL: 
http://svn.apache.org/viewvc/pig/trunk/CHANGES.txt?rev=1710672&r1=1710671&r2=1710672&view=diff
==============================================================================
--- pig/trunk/CHANGES.txt (original)
+++ pig/trunk/CHANGES.txt Mon Oct 26 18:35:32 2015
@@ -57,6 +57,8 @@ PIG-4639: Add better parser for Apache H
 
 BUG FIXES
 
+PIG-4712: [Pig on Tez] NPE in Bloom UDF after Union (rohini)
+
 PIG-4707: [Pig on Tez] Streaming job hangs with pig.exec.mapPartAgg=true 
(rohini)
 
 PIG-4703: TezOperator.stores shall not ship to backend (daijy)

Modified: 
pig/trunk/src/org/apache/pig/backend/hadoop/executionengine/physicalLayer/expressionOperators/POUserFunc.java
URL: 
http://svn.apache.org/viewvc/pig/trunk/src/org/apache/pig/backend/hadoop/executionengine/physicalLayer/expressionOperators/POUserFunc.java?rev=1710672&r1=1710671&r2=1710672&view=diff
==============================================================================
--- 
pig/trunk/src/org/apache/pig/backend/hadoop/executionengine/physicalLayer/expressionOperators/POUserFunc.java
 (original)
+++ 
pig/trunk/src/org/apache/pig/backend/hadoop/executionengine/physicalLayer/expressionOperators/POUserFunc.java
 Mon Oct 26 18:35:32 2015
@@ -566,6 +566,8 @@ public class POUserFunc extends Expressi
             requestedParallelism, null, funcSpec.clone());
         clone.setResultType(resultType);
         clone.signature = signature;
+        clone.cacheFiles = cacheFiles;
+        clone.shipFiles = shipFiles;
         return clone;
     }
 

Modified: pig/trunk/test/e2e/pig/tests/nightly.conf
URL: 
http://svn.apache.org/viewvc/pig/trunk/test/e2e/pig/tests/nightly.conf?rev=1710672&r1=1710671&r2=1710672&view=diff
==============================================================================
--- pig/trunk/test/e2e/pig/tests/nightly.conf (original)
+++ pig/trunk/test/e2e/pig/tests/nightly.conf Mon Oct 26 18:35:32 2015
@@ -5148,6 +5148,30 @@ store C into ':OUTPATH:';\,
                                 C = load ':INPATH:/singlefile/votertab10k'as 
(name:chararray, age:int, reg:chararray, contrib:float);
                                 D = join C by name, B by name;
                                 store D into ':OUTPATH:';",
+                    },{
+                        'num' => 4,
+                        'pig' => "set pig.optimizer.rules.disabled 
PushUpFilter;
+                                define bb BuildBloom('Hash.JENKINS_HASH', 
'fixed', '128', '3');
+                                A = LOAD ':INPATH:/singlefile/studenttab10k' 
AS (name:chararray, age:int, gpa:double);
+                                B = filter A by name == 'alice allen';
+                                C = group B all;
+                                D = foreach C generate bb(B.name);
+                                store D into ':HDFSTMP:/mybloom_4';
+                                exec;
+                                define bloom Bloom(':HDFSTMP:/mybloom_4');
+                                E = LOAD ':INPATH:/singlefile/studenttab10k' 
AS (name:chararray, age:int, gpa:double);
+                                F = LOAD ':INPATH:/singlefile/studenttab10k' 
AS (name:chararray, age:int, gpa:double);
+                                G = union E, F;
+                                -- PushUpFilter is disabled to avoid filter 
being pushed before union
+                                H = filter G by bloom(name);
+                                store H into ':OUTPATH:';",
+                        'notmq' => 1,
+                        'verify_pig_script' => "
+                                A = LOAD ':INPATH:/singlefile/studenttab10k' 
AS (name, age:int ,gpa:double);
+                                B = LOAD ':INPATH:/singlefile/studenttab10k' 
AS (name, age:int ,gpa:double);
+                                C = UNION A,B;
+                                D = filter C by name == 'alice allen';
+                                store D into ':OUTPATH:';",
                     }
                 ],
             },{


Reply via email to