[
https://issues.apache.org/jira/browse/PIG-3088?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13637362#comment-13637362
]
Prashant Kommireddi commented on PIG-3088:
------------------------------------------
Hey Jon, the patch does not apply cleanly.
grep "hunks FAILED" pig3088
13 out of 13 hunks FAILED -- saving rejects to file
src/org/apache/pig/newplan/logical/rules/PushDownForEachFlatten.java.rej
6 out of 8 hunks FAILED -- saving rejects to file
src/org/apache/pig/parser/LogicalPlanBuilder.java.rej
Russell, making this default behavior could result in duplicated alias.
Consider the following when "z1" is a common field name between the 2
relations.
{code}
grunt> a = load 'a' as (x1, y1, z1);
grunt> b = load 'a' as (x2, y2, z1);
grunt> c = join a by x1, b by x2;
grunt> describe c;
c: {a::x1: bytearray,a::y1: bytearray,a::z1: bytearray,b::x2: bytearray,b::y2:
bytearray,b::z1: bytearray}
{code}
> Add a builtin udf which removes prefixes
> ----------------------------------------
>
> Key: PIG-3088
> URL: https://issues.apache.org/jira/browse/PIG-3088
> Project: Pig
> Issue Type: New Feature
> Reporter: Jonathan Coveney
> Assignee: Jonathan Coveney
> Fix For: 0.12
>
> Attachments: PIG-3088-0.patch
>
>
> This is something that I always hear people complaining about. Note that this
> depends on the FlattenOutput annotation.
> This UDF supports the following.
> {code}
> a = load 'a' as (x1, y1, z1);
> b = load 'a' as (x2, y2, z2);
> c = join a by x1, b by x2;
> describe c;
> --c: {a::x1: bytearray,a::y1: bytearray,a::z1: bytearray,b::x2:
> bytearray,b::y2: bytearray,b::z2: bytearray}
> d = foreach c generate RemovePrefix(*);
> describe d;
> --d: {x1: bytearray,y1: bytearray,z1: bytearray,x2: bytearray,y2:
> bytearray,z2: bytearray}
> {code}
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
