Duplicate column names in foreach do not throw parser error
-----------------------------------------------------------
Key: PIG-644
URL: https://issues.apache.org/jira/browse/PIG-644
Project: Pig
Issue Type: Bug
Components: impl
Affects Versions: types_branch
Reporter: Viraj Bhat
Fix For: types_branch
Consider the following Pig script where we generate column names b and b in the
FOREACH
{code}
DATA = LOAD 'blah.txt' as (a:long, b:long);
RESULT = FOREACH DATA GENERATE a, b, (b>20?b:0) as b;
DESCRIBE RESULT;
dump RESULT;
{code}
Pig runs the script successfully and does not complain of the duplicate column
names. I do not know if the new error handling framework will handle these
kinds of cases.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.