[
https://issues.apache.org/jira/browse/PIG-2148?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13056867#comment-13056867
]
Eric Yang commented on PIG-2148:
--------------------------------
+1 on Dmitriy's proposal. Alan, if we can use the directory structure proposed
by Owen in HADOOP-6255, it would be helpful to trim down the size of pig.jar
file, and setup dependency properly from single $PREFIX/share/[hadoop|hbase]
location to refer to hadoop and hbase jar files.
> pig script should have command line option to use pig-withouthadoop.jar
> instead of pig-core.jar
> -----------------------------------------------------------------------------------------------
>
> Key: PIG-2148
> URL: https://issues.apache.org/jira/browse/PIG-2148
> Project: Pig
> Issue Type: New Feature
> Components: impl
> Affects Versions: 0.9.0
> Reporter: Alan Gates
> Assignee: Alan Gates
> Priority: Minor
> Fix For: 0.10
>
> Attachments: PIG-2148.patch
>
>
> If you use Pig downloaded from Apache with Hadoop 0.20.203 or CDH3, you have
> to use the pig-withouthadoop.jar instead of pig-core.jar. However bin/pig
> uses pig-core.jar, meaning you can't use the shell script to start Pig. We
> need to add a -without option to bin/pig so that it will pick up the
> pig-withouthadoop.jar.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
