On 04/23/09 16:23, Doug Leavitt wrote:
> As I understand it, if the package doesn't pass validation it is not
> eligible to be promoted, and voting is unnecessary.
> 
> If it has been validated then people can vote on it, and the
> submission can be rejected with a single -1 vote from anyone
> eligible to vote.
> 
> [jucr team, these are accurate statements, correct?]

Yes - All Correct.


- Dermot



> 
> Only packages with 2 +1 votes and no -1 votes, and a day of waiting,
> are eligible to be promoted to /contrib.
> 
> 
> If any approver notices a collision coming from a roboporter
> package, please give the submission a -1 vote and a comment
> what it collides with so we can correct the problem.
> 
> Thanks in advance,
> Doug.
> 
> 
> Alan Coopersmith wrote:
>> http://jucr.opensolaris.org/review/packages/351/ is the package in
>> question - I can't see how to reject it though, is that because it
>> failed to validate?
>>
>> (It's actually for an older bugfix release 2.3.4, instead of the 2.3.7
>>  we currently ship - probably because they took the SFE spec from when
>>  the bundled freetype was still at 2.1.x and SFE wanted a newer one, but
>>  since we upgraded the bundled one, SFE hasn't needed to keep their spec
>>  updated.)
>>
>>     -Alan Coopersmith-           alan.coopersmith at sun.com
>>      Sun Microsystems, Inc. - X Window System Engineering
>>
>> Doug Leavitt wrote:
>>> I'll second Christian's yes from my side of the field, and add that
>>> in theory all pkgfactory submissions should already be filtering
>>> for this type of error.  We currently have collision detection code
>>> in pkgfactory and we are actively checking everything we build against
>>> the base release we are building on (snv_101).
>>>
>>> That said, if something slips by our collision detector, please
>>> reject the package and let us know, so we can look at why it it was
>>> not caught.
>>>
>>> Thanks,
>>> Doug.
>>>
>>> Christian Kelly wrote:
>>>> Alan Coopersmith wrote:
>>>>> Jim Walker wrote:
>>>>>  
>>>>>> If you are listed here as a Package Advocate:
>>>>>> http://opensolaris.org/os/community/sw-porters/contributing/reporoles/ 
>>>>>>
>>>>>>     
>>>>> So one of the packages I see in the queue is SFEfreetype, which
>>>>> duplicates
>>>>> a library we've shipped in Solaris for many years and which is
>>>>> already in
>>>>> both /release and /dev - should we be rejecting such packages as
>>>>> duplication?
>>>>>
>>>>>   
>>>> Yes, we should be rejecting submissions like these, and there have
>>>> been a few we have rejected already.
>>>>
>>>> I'm working on some conflict detection code which will catch some of
>>>> these submissions and we're also implementing a search feature which
>>>> will help users to check whether they are duplicating work or not.
>>>>
>>>> _Christian
>>>> _______________________________________________
>>>> sw-porters-discuss mailing list
>>>> sw-porters-discuss at opensolaris.org
>>>> http://mail.opensolaris.org/mailman/listinfo/sw-porters-discuss
>>
>>
> _______________________________________________
> sw-porters-discuss mailing list
> sw-porters-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/sw-porters-discuss

Reply via email to