[
https://issues.apache.org/jira/browse/MAPREDUCE-3860?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13797413#comment-13797413
]
Andrey Klochkov commented on MAPREDUCE-3860:
--------------------------------------------
Ravi, thanks a lot for this summary. I should have written this myself when
submitting the patch. Now answering your questions.
The old tests were deleted by the following commit:
{code}
commit 00ac37838c4a55a2b855983e9730cbd26e6f3477
Author: Mahadev Konar <[email protected]>
Date: Sat Jan 21 01:15:24 2012 +0000
MAPREDUCE-3705. ant build fails on 0.23 branch. (Thomas Graves via mahadev)
git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/trunk@1234227
13f79535-47bb-0310-9956-ffa450edef68
{code}
So I got the tests from the preceding commit, namely:
{code}
commit 6b8a6a701a972f9528b9a2672401db51a31f52fb
Author: Mahadev Konar <[email protected]>
Date: Sat Jan 21 00:53:02 2012 +0000
MAPREDUCE-3549. write api documentation for web service apis for RM, NM,
mapreduce app master, and job history server (Thoma
git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/trunk@1234222
13f79535-47bb-0310-9956-ffa450edef68
{code}
The tests are under
{{hadoop-mapreduce-project/src/test/mapred/org/apache/hadoop/tools/rumen}}. The
data is under {{hadoop-mapreduce-project/src/test/tools/data/rumen}}.
The {{rumen-test-data.tar.gz}} file contains those data files which are in
binary form, so I couldn't put those into a patch file. The non-binary
(non-gzipped) files are in the patch file itself. So to put the changes
properly it's need to apply the patch and also to un-tar
{{rumen-test-data.tar.gz}}.
I did not create {{sample-conf.file.new.xml}} file, it existed in the old
tests. Also, see {{TestRumenAnonymization}}, {{TestRumenFolder}} there.
I did change {{job-tracker-logs-topology-output}} to make tests succeed. As I
understand this is caused by newer versions of Rumen doing "time adjustment",
so the expected data in {{job-tracker-logs-topology-output}} is not what's
being produced by Rumen currently. See {{Folder.adjustJobTimes}} method.
The change in {{WordList}} is actually a bug fix. I figured it doesn't make
much sense to file a separate Jira for that. Sometimes when WordList instance
is being deserialized from the disk, the "size" attribute is read after the
words themselves are read, and so when deserializing "size" the words list is
cleared (a bug in deserialization).
I did make many changes in {{TestRumenJobTraces}} and
{{TestRumenAnonymization}}, that was required due to changes in Hadoop itself.
> [Rumen] Bring back the removed Rumen unit tests
> -----------------------------------------------
>
> Key: MAPREDUCE-3860
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-3860
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: tools/rumen
> Reporter: Ravi Gummadi
> Assignee: Andrey Klochkov
> Attachments: MAPREDUCE-3860.patch, rumen-test-data.tar.gz
>
>
> MAPREDUCE-3582 did not move some of the Rumen unit tests to the new folder
> and then MAPREDUCE-3705 deleted those unit tests. These Rumen unit tests need
> to be brought back:
> TestZombieJob.java
> TestRumenJobTraces.java
> TestRumenFolder.java
> TestRumenAnonymization.java
> TestParsedLine.java
> TestConcurrentRead.java
--
This message was sent by Atlassian JIRA
(v6.1#6144)
