You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mrunit.apache.org by "E. Sammer (JIRA)" <ji...@apache.org> on 2011/05/12 01:09:47 UTC
[jira] [Created] (MRUNIT-9) Fix failing testDoubleIteration test
Fix failing testDoubleIteration test
------------------------------------
Key: MRUNIT-9
URL: https://issues.apache.org/jira/browse/MRUNIT-9
Project: MRUnit
Issue Type: Bug
Affects Versions: 0.5.0
Reporter: E. Sammer
Assignee: E. Sammer
Fix For: 0.5.0
The testDoubleIteration() test in TestReduceDriver is failing due to an extra null record in position 0 (rather than 1).
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Resolved] (MRUNIT-9) Fix failing testDoubleIteration test
Posted by "E. Sammer (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MRUNIT-9?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
E. Sammer resolved MRUNIT-9.
----------------------------
Resolution: Fixed
It turns out that somewhere along the way, mrunit in trunk was missing a fix that returned a NullIterator (which never returns data) after the reducer has exhausted its input iterable. Maybe this fix was never ported to Hadoop trunk and only lives on one of the 0.20.x branches?
Anyway, fixed and pushed to trunk.
> Fix failing testDoubleIteration test
> ------------------------------------
>
> Key: MRUNIT-9
> URL: https://issues.apache.org/jira/browse/MRUNIT-9
> Project: MRUnit
> Issue Type: Bug
> Affects Versions: 0.5.0
> Reporter: E. Sammer
> Assignee: E. Sammer
> Fix For: 0.5.0
>
>
> The testDoubleIteration() test in TestReduceDriver is failing due to an extra null record in position 0 (rather than 1).
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MRUNIT-9) Fix failing testDoubleIteration test
Posted by "Aaron Kimball (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MRUNIT-9?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13032178#comment-13032178 ]
Aaron Kimball commented on MRUNIT-9:
------------------------------------
All these tests passed before.
> Fix failing testDoubleIteration test
> ------------------------------------
>
> Key: MRUNIT-9
> URL: https://issues.apache.org/jira/browse/MRUNIT-9
> Project: MRUnit
> Issue Type: Bug
> Affects Versions: 0.5.0
> Reporter: E. Sammer
> Assignee: E. Sammer
> Fix For: 0.5.0
>
>
> The testDoubleIteration() test in TestReduceDriver is failing due to an extra null record in position 0 (rather than 1).
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (MRUNIT-9) Fix failing testDoubleIteration test
Posted by "E. Sammer (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MRUNIT-9?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13032269#comment-13032269 ]
E. Sammer commented on MRUNIT-9:
--------------------------------
Not sure how this one worked before. The iterator() method on InspectableIterable always invokes its baseIter's iterator() method and creates a new instance of InspectableIterator. If the for (x : i) construct invokes iterator() on i, you're always going to get a reset iterator and the failure I'm seeing[1].
[1] My description of the error is wrong. The actual failure is actually that the iterator doesn't reset and this code:
for (int v : i) {
count += v;
}
for (int v : i) {
count += v;
}
yields count == 8 if i contains 4 1s rather than 4.
> Fix failing testDoubleIteration test
> ------------------------------------
>
> Key: MRUNIT-9
> URL: https://issues.apache.org/jira/browse/MRUNIT-9
> Project: MRUnit
> Issue Type: Bug
> Affects Versions: 0.5.0
> Reporter: E. Sammer
> Assignee: E. Sammer
> Fix For: 0.5.0
>
>
> The testDoubleIteration() test in TestReduceDriver is failing due to an extra null record in position 0 (rather than 1).
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira