hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Thomas Sandholm (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-5518) MRUnit unit test library
Date Tue, 07 Apr 2009 15:28:13 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-5518?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12696576#action_12696576
] 

Thomas Sandholm commented on HADOOP-5518:
-----------------------------------------

Seem like this commit broke the automated test-patch script. The eclipse classpath check always
fails (gives -1 as seen above for all patches).
The error can be reproduced by doing a diff on:

find build/ivy/lib/Hadoop/common/ lib/ src/test/lib/ -name '*.jar' |sort
and
sed -n 's@.*kind="lib".*path="\(.*jar\)".*@\1@p' < .eclipse.templates/.classpath | sort

I presume you just need to add the new junit version to the eclipse file 'eclipse.templates/.classpath'.
But I am not an eclipse user so I am not sure, I am just trying to submit a completely unrelated
patch.


> MRUnit unit test library
> ------------------------
>
>                 Key: HADOOP-5518
>                 URL: https://issues.apache.org/jira/browse/HADOOP-5518
>             Project: Hadoop Core
>          Issue Type: New Feature
>            Reporter: Aaron Kimball
>            Assignee: Aaron Kimball
>             Fix For: 0.21.0
>
>         Attachments: HADOOP-5518-2.patch, HADOOP-5518-3.patch, mrunit.patch, overview.html
>
>
> MRUnit is a tool to help authors of MapReduce programs write unit tests.
> Testing map() and reduce() methods requires some repeated work to mock the inputs and
outputs of a Mapper or Reducer class, and ensure that the correct values are emitted to the
OutputCollector based on inputs. Also, testing a mapper and reducer together requires running
them with the sorted ordering guarantees made by the shuffle process.
> This library provides the above functionality to authors of maps and reduces; it allows
you to test maps, reduces, and map-reduce pairs without needing to perform all the setup and
teardown work associated with running a job.
> I believe this tool may be useful to the broader Hadoop community, so I have cleaned
it up and would like to see it become a "contrib" module. My current environment is based
on Hadoop 0.18, so this is the format it expects to use. It does not have support for the
new Context-based interfaces for mappers/reducers.
> I have attached the overview.html file for its javadoc, which provides more synopsis
and an example of usage; I am also providing the current source code so that you can evaluate
its structure.
> Ideally with some feedback from the community this will move toward supporting the current
trunk interface soon.
> This currently works with JUnit 4; the supplied patch changes Ivy's libraries.properties
file to use JUnit 4.5. I'm marking HADOOP-4901 as a dependency for this reason.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message