Marshall Schor wrote:
> Michael Baessler wrote:
>> I have successfully verified/tested the memory fixes for release
>> uima-2.2.2-fp1-incubating.
>>
>> The release artifacts are available on
>>     
>>     people.a.o/~mbaessler/uima-2.2.2-fp1/02
>>
>>
>> For the fixpack we ship the uima-core.jar and a readme packaged as zip
>> (uima-2.2.2-fp1-incubating.zip).
>>
>> Please cast your vote:
>>
>> [  ] +1  uima-2.2.2-fp1-incubating release candidate 02 is ready and
>> can be released
>> [  ]  0  don't care
>> [  ] -1  don't release yet, I found some more issues
>>   
> +1
> 
> Observations - no blockers, but in my review I noticed the following:
> 
> 1) I see the hot fix packaging is just Zip?  No Tars?  This is probably
> OK in that the released artifact is just the Java JAR file which should
> be platform independent.
Yes, I thought a Zip should be sufficient since it is just used to package the 
hotfix readme with
the platform independent jar.

> 
> 2) The components being released do not link to the corresponding
> source(s).  Not a blocker, but if we redo, I think it would be good to
> identify in the readme where the source that corresponds to this is
> (what tag in SVN).
Good idea, will do if we have to redo.

> 
> 3) I would note on the post to [EMAIL PROTECTED] that the LICENSE,
> NOTICE, and DISCLAIMER files are inside the Jar, and that RAT was not
> run since there was only 1 file (the JAR), and the Readme.
Will do.

> 
> 4) The POM included in the Jar under the "maven" directory shows the
> parent of this artifact to be the pom org.apache.uima : uimaj :
> 2.2.2-01-incubating.   I think this means that when maven artifacts are
> uploaded, the uimaj POM needs to be "released" as well.  Because of
> this, I would suggest in the note for the [EMAIL PROTECTED] vote, to
> include a link to this in the SVN tag, along with a note that this would
> be uploaded to the incubator maven repository as part of the release.
Not sure if we really want to upload this hotfix also to the Maven repository.
I would say no, just upload real releases.

> 
> 5) I verified that the main uima documentation has been updated with the
> new performance tuning options.
> 
> Found one typo in the README-Hotfix1 (not a blocker, fixed in SVN):
> "There is a small number of performance tuning options" ->
> "There are a small number of performance tuning options"
Try to fix it if we have to redo.
> 
> 6) In a separate thread I propose some change to the documentation for
> JCAS_CACHE_ENABLED.  Current doc in the readme is ok for this release.
> 
> -Marshall

Thanks for looking into the details!

-- Michael

Reply via email to