You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Shawn Heisey (JIRA)" <ji...@apache.org> on 2017/12/12 22:56:00 UTC

[jira] [Commented] (SOLR-11754) Remove AbstractSolrTestCase

    [ https://issues.apache.org/jira/browse/SOLR-11754?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16288416#comment-16288416 ] 

Shawn Heisey commented on SOLR-11754:
-------------------------------------

It certainly does seem that there's almost nothing there.  What little is there that has function can probably be added to SolrTestCaseJ4.  Looking over the code, I don't even see any evidence that the class does what the javadoc says it does, which is creating/destroying the index, and providing assertion methods.

I guess the only question really becomes whether or not the abstraction itself makes it easier for people to mentally grasp the test architecture.  I don't have a problem with shim classes that don't do very much, *if* their existence and the name they've been given helps with understanding the code.

I do see AbstractZkTestCase as well, with far fewer descendants, but that abstract class has a LOT more going on in it than the one you're proposing to remove.


> Remove AbstractSolrTestCase
> ---------------------------
>
>                 Key: SOLR-11754
>                 URL: https://issues.apache.org/jira/browse/SOLR-11754
>             Project: Solr
>          Issue Type: Test
>      Security Level: Public(Default Security Level. Issues are Public) 
>          Components: Tests
>            Reporter: David Smiley
>            Assignee: David Smiley
>
> I'm arguing AbstractSolrTestCase should be removed as it's obsoleted by SolrTestCaseJ4.
> In SOLR-3911 (back in 2012) Mark made it extend from SolrTestCaseJ4.  There is really very little in this test class.  Some of the methods here are duplicated by SolrTestCaseJ4 and thus are completely redundant (ignoreException, resetExceptionIgnores, getFile).  There haven't been any modifications to this class of substance since 2012 either.
> I think we can just outright remove it (no deprecation phase).  Anyone still using it can trivially switch.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org