[
https://issues.apache.org/jira/browse/PIG-5223?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15968655#comment-15968655
]
Jin Sun commented on PIG-5223:
------------------------------
Hey [~daijy]
The second test failed should be easily fixed by changing "in" to "into". It is
just a syntax mistake.
However I want your opinion on the first one:
In this test case, limit is derived from a variable. Due to lazy evaluation,
LOLimit.mLimit is -1 until execution. When we merge LOLimit into LOSort, LOSort
will have limit = -1, which later gets converted to POSort with limit = -1.
This is how we lost limit and fails this test. I think we should disable
NestedLimitOptimizer when we find LOLimit.mLimit is -1. Please correct me if I
am wrong.
I just made a patch and seems like tests are passing.
Thanks,
Jin
> TestLimitVariable.testNestedLimitVariable1 and
> TestSecondarySortMR.testNestedLimitedSort failing
> -------------------------------------------------------------------------------------------------
>
> Key: PIG-5223
> URL: https://issues.apache.org/jira/browse/PIG-5223
> Project: Pig
> Issue Type: Bug
> Reporter: Koji Noguchi
>
> TestLimitVariable.testNestedLimitVariable1
> {quote}
> Comparing actual and expected results. expected:<\[(1,11), (2,3), (3,10),
> (6,15)]> but was:<\[(1,11), (2,3), (3,10), (4,11), (5,10), (6,15)]>
> {quote}
> TestSecondarySortMR.testNestedLimitedSort
> {quote}
> Error during parsing. <line 1, column 158> mismatched input 'in' expecting
> INTO
> {quote}
> Latter is probably a simple syntax error. Former looks serious.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)