[ 
https://issues.apache.org/jira/browse/SOLR-3204?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13224424#comment-13224424
 ] 

Uwe Schindler edited comment on SOLR-3204 at 3/7/12 3:30 PM:
-------------------------------------------------------------

{quote}
bq. How we do this is a different issue: We can only rename all classes in that 
jar-file and republish it (but then we have to change our import statements in 
our code – I would prefer this), or we jarjar the whole Solr distribution after 
the complete build.

It's preferable to post process the Solr jar rather than renaming the packages 
imported in the source code. That's less intrusive and let you upgrade to the 
official fixed dependency more easily.
{quote}

That's our committers decision. The fact here is, that Solr or even Lucene is 
not just a WAR file, its consisting of lot's of unrelated JAR artifacts and all 
of them should live on its own, so a user can remove features he does not need. 
So I prefer to "transform" those "changed" artifacts separately to private 
"namespace" and simply change maybe 3 import statements (I can even use "sed" 
for that).
                
      was (Author: thetaphi):
    {quote}
{quote}
How we do this is a different issue: We can only rename all classes in that 
jar-file and republish it (but then we have to change our import statements in 
our code – I would prefer this), or we jarjar the whole Solr distribution after 
the complete build.
{quote}

It's preferable to post process the Solr jar rather than renaming the packages 
imported in the source code. That's less intrusive and let you upgrade to the 
official fixed dependency more easily.
{quote}

That's our committers decision. The fact here is, that Solr or even Lucene is 
not just a WAR file, its consisting of lot's of unrelated JAR artifacts and all 
of them should live on its own, so a user can remove features he does not need. 
So I prefer to "transform" those "changed" artifacts separately to private 
"namespace" and simply change maybe 3 import statements (I can even use "sed" 
for that).
                  
> solr-commons-csv must not use the org.apache.commons.csv package
> ----------------------------------------------------------------
>
>                 Key: SOLR-3204
>                 URL: https://issues.apache.org/jira/browse/SOLR-3204
>             Project: Solr
>          Issue Type: Bug
>          Components: Build
>    Affects Versions: 3.5
>            Reporter: Emmanuel Bourg
>            Priority: Blocker
>             Fix For: 3.6
>
>         Attachments: SOLR-3204.patch, solr-csv.patch
>
>
> The solr-commons-csv artifact reused the code from the Apache Commons CSV 
> project but the package wasn't changed to something else than 
> org.apache.commons.csv in the process. This creates a compatibility issue as 
> the Apache Commons team works toward an official release of Commons CSV. It 
> prevents Commons CSV from using its own org.apache.commons.csv package, or 
> forces the renaming of all the classes to avoid a classpath conflict.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: 
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira



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

Reply via email to