[ 
https://issues.apache.org/jira/browse/HIVE-1359?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12871940#action_12871940
 ] 

John Sichi commented on HIVE-1359:
----------------------------------

Regarding proposed change #1, let's keep the test include/exclude 
specifications in a separate test-config.xml (imported from build.xml or 
whatever) to make maintenance easier.

Regarding proposed change #2, let's try to keep conf duplication to a minimum 
where possible.  In an HBase core-site.xml file, I saw this syntax used to pull 
in additional configuration overlay; not sure if that works in Hive conf files 
too:
{noformat}
<xi:include href="core-site-custom.xml" 
xmlns:xi="http://www.w3.org/2001/XInclude"; />
{noformat}

Regarding proposed change #3, the user should not need to do this explicitly 
when running ant test unless they want to override, right?  It should happen 
automatically based on the Hadoop version chosen.

Regarding proposed change #4, what we actually want is the ability to declare a 
subset of tests as needing minimr and run them that way by default with ant 
test (no -D required by the person running the tests, although internally we 
will set the property as part of invoking the test).  We can take this up 
separately in HIVE-117; just want to make the requirements clear here.


> Unit test should be shim-aware
> ------------------------------
>
>                 Key: HIVE-1359
>                 URL: https://issues.apache.org/jira/browse/HIVE-1359
>             Project: Hadoop Hive
>          Issue Type: New Feature
>            Reporter: Ning Zhang
>            Assignee: Ning Zhang
>         Attachments: unit_tests.txt
>
>
> Some features in Hive only works for certain Hadoop versions through shim. 
> However the unit test structure is not shim-aware in that there is only one 
> set of queries and expected outputs for all Hadoop versions. This may not be 
> sufficient when we will have different output for different Hadoop versions. 
> One example is CombineHiveInputFormat wich is only available from Hadoop 
> 0.20. The plan using CombineHiveInputFormat and HiveInputFormat may be 
> different. Another example is archival partitions (HAR) which is also only 
> available from 0.20. 

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

Reply via email to