Distributions going smart & bugfixes?

[Axel Thimm]

On Mon, Jan 05, 2009 at 04:14:51PM +0100, Anders F Björklund wrote:
> Gustavo Niemeyer wrote:
>
>>> Depends on what release you want, "bugfix" was supposed to be
>>> the minor revision (e.g. 1.1.2) and "testing" the major (1.2)
>>
>> Anders, we discussed this a few times already, and we can schedule a
>> call if you still didn't get the idea behind it.  The way to get new
>> releases of Smart out isn't by creating standalone branches and
>> merging things on them.  We need to get branches or patches ready for
>> merges, then reviewing them, then merging them on trunk, and releasing
>> trunk as the next stable.
>
> I think I got the idea, I just thought it would be a shame to remove
> the branches that had already been created... Not creating any new ones
> like that, though. And I've tried to crossref/copy where applicable.
> As always, "trunk" (and releases) is the only official branch of Smart.
>
>> Merging several patches in the same branch makes the review process
>> hard since all the fixes get mixed on a single bucket, and isolating
>> which changes are tested and ok to merge and which ones need merging
>> is hard.
>
> The main reason for creating the branches was to merge the patches,
> some of them interfered with eachother which made the diffs obsolete.
>
>> That said, creating branches for merging several changes in and
>> testing things off is definitely nice.  It's just that we can't say
>> that a non-trunk branch *is* a release, since things can go in or not.
>
> Bad wording on my part, I meant to say "was intended for" not "is"...
> The idea was to help with the release, not create additional work :-)

Gustavo, I think the way the bugfixes branch is being "misused" is
critically good for smart.

There were several issues with smart on Fedora 10 most of which Anders
accumulated in the stable bugfixes branch (no features, just
bugfixes).

Waiting for a review and a release just make users run away from smart
since it doesn't work on their distribution.

Can we find a way in between? A branch were critical bugfixes are
allowed to be made on stable code (e.g. non trunk)? This service
is/was invaluable to us simple distribution supporters. Otherwise we
would need to wade through all the known bugs to see whether there is
already a patch attached or not.

Thanks!
-- 
Axel.Thimm at ATrpms.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <http://lists.labix.org/pipermail/smart-labix.org/attachments/20090105/f559e6fc/attachment-0003.pgp>