Danek - if you want to back this out then no issue.

I see J has a backout patch available. I'm not at a build machine at 
present, but can be if need be. My tests on the original patch had 
included the PM and UM not the CLI. If you or J can back it out, we can 
then resolve the issues with Fujiwara over the next few days to 
everyones satisfaction going forward. Apologies for the foul up.

JR

Danek Duvall wrote:
> On Wed, Oct 29, 2008 at 10:58:37AM -0700, Stephen Hahn wrote:
>
>   
>>   It's clear that the test suite was not run prior to integration.  If
>>   there's a simple fix to this, it would be good to see it in the next
>>   couple of hours; otherwise, this change needs to be delta'ed out.
>>   Volunteers?
>>     
>
> I took a look to see what the fix would be, as I'm not inclined to just
> back it out.
>
> The immediate problem, of course, is that this call should be to
> textdomain(), not domain().  But even with that fixed, the fact that the
> call to gettext.install() was removed from client.py means that all the
> modules which were expecting _() to be in their builtin namespace (which is
> one of the things that gettext.install() does) no longer have it.
>
> Making the stack traces go away and getting the test suite passing again
> isn't all that much more difficult, but there are a couple of ways forward,
> and it's not clear to me which it is.  If Fujiwara can clearly answer the
> questions I put out last night, that'll help guide us, but that's not
> likely to happen for several hours yet.  (One of the downsides of having
> the three major parties in this all separated by eight hours.)
>
> Danek
> _______________________________________________
> pkg-discuss mailing list
> [email protected]
> http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
>   

_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss

Reply via email to