[ 
https://issues.apache.org/jira/browse/SOLR-11490?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16209519#comment-16209519
 ] 

Alexandre Rafalovitch commented on SOLR-11490:
----------------------------------------------

Sure. Let's correct whatever is wrong. Maybe this patch is correct for most of 
entries but is wrong for some specific ones. Then, we can figure out the 
specific set and ensure it is treated right for both this and future additions.

So, let's take ArabicStemFilterFactory as an example you probably meant to use 
(ArabicTokenizer does not seem to exist). 

I see it in the release 1.4: 
https://github.com/apache/lucene-solr/blob/releases/solr/1.4.0/src/java/org/apache/solr/analysis/ArabicStemFilterFactory.java
Now, that's release predating Lucene and Solr merge (in 3.1), so perhaps this 
is the part we need to discuss to make it clearer?

I also see that it was moved to Lucene packages from Solr packages: 
https://github.com/apache/lucene-solr/blob/master/lucene/analysis/common/src/java/org/apache/lucene/analysis/ar/ArabicStemFilterFactory.java
 This happened with LUCENE-2510 for Lucene/Solr 4. But that's the package 
change and the functionality did not change. So, the user that needs to work 
with Arabic text still could.

I am not sure what happened in Lucene 2.9.0. I can see LUCENE-1460 which seems 
relevant, but the functionality in question seems to have been present both 
before and after.

What am I missing?

> Add @since javadoc tags to the interesting Solr/Lucene classes
> --------------------------------------------------------------
>
>                 Key: SOLR-11490
>                 URL: https://issues.apache.org/jira/browse/SOLR-11490
>             Project: Solr
>          Issue Type: Improvement
>      Security Level: Public(Default Security Level. Issues are Public) 
>            Reporter: Alexandre Rafalovitch
>            Assignee: Alexandre Rafalovitch
>            Priority: Minor
>
> As per the discussion on the dev list, it may be useful to add Javadoc since 
> tags to significant (or even all) Java files.
> For user-facing files (such as analyzers, URPs, stream evaluators, etc) it 
> would be useful when trying to identifying whether a particular class only 
> comes later than user's particular version.
> For other classes, it may be useful for historical reasons.



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

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

Reply via email to