Luong Hoang Phuc created PIG-4296:
-------------------------------------
Summary: Cannot filter partition by 'matches'
Key: PIG-4296
URL: https://issues.apache.org/jira/browse/PIG-4296
Project: Pig
Issue Type: Bug
Affects Versions: 0.12.0
Environment: Cloudera CDH 5.2, hive 0.13, pig 0.12
Reporter: Luong Hoang Phuc
I use Cloudera CDH 5.2 with hive 0.13 and pig 0.12, and I have a hive table
with partition named 'g'. When i try to run:
A = LOAD 'test' USING org.apache.hcatalog.pig.HCatLoader() AS (a:chararray,
b:chararray, game:chararray, g:chararray);
B = FILTER A BY g matches 'abc';
DUMP B;
I receive error:
Caused by: java.io.IOException: MetaException(message:Error parsing partition
filter; lexer error: null; exception NoViableAltException(11@[]))
at
org.apache.hcatalog.mapreduce.HCatInputFormat.setFilter(HCatInputFormat.java:111)
at org.apache.hcatalog.pig.HCatLoader.setLocation(HCatLoader.java:120)
at
org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.JobControlCompiler.getJob(JobControlCompiler.java:498)
... 19 more
Caused by: MetaException(message:Error parsing partition filter; lexer error:
null; exception NoViableAltException(11@[]))
at
org.apache.hadoop.hive.metastore.api.ThriftHiveMetastore$get_partitions_by_filter_result$get_partitions_by_filter_resultStandardScheme.read(ThriftHiveMetastore.java:64602)
at
org.apache.hadoop.hive.metastore.api.ThriftHiveMetastore$get_partitions_by_filter_result$get_partitions_by_filter_resultStandardScheme.read(ThriftHiveMetastore.java:64569)
at
org.apache.hadoop.hive.metastore.api.ThriftHiveMetastore$get_partitions_by_filter_result.read(ThriftHiveMetastore.java:64503)
at org.apache.thrift.TServiceClient.receiveBase(TServiceClient.java:78)
at
org.apache.hadoop.hive.metastore.api.ThriftHiveMetastore$Client.recv_get_partitions_by_filter(ThriftHiveMetastore.java:1942)
at
org.apache.hadoop.hive.metastore.api.ThriftHiveMetastore$Client.get_partitions_by_filter(ThriftHiveMetastore.java:1926)
at
org.apache.hadoop.hive.metastore.HiveMetaStoreClient.listPartitionsByFilter(HiveMetaStoreClient.java:914)
at
org.apache.hcatalog.mapreduce.InitializeInput.getInputJobInfo(InitializeInput.java:113)
at
org.apache.hcatalog.mapreduce.InitializeInput.setInput(InitializeInput.java:86)
at
org.apache.hcatalog.mapreduce.HCatInputFormat.setFilter(HCatInputFormat.java:109)
... 21 more
I tested 'matches' with any field (not partition) it's ok.
With my old Cloudera CDH 4.6.0 hive 0.11 and pig 0.11 it's still OK with
'matches' partition.
Please check if I have any mistake. Thank you.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
