Evan Prodromou wrote:
On Sat, 2005-12-03 at 13:41 -0800, Brion Vibber
wrote:
If it looks ok and doesn't break the actual
called hook functions (or if
it does, and we have some warning to fix them first), *then* go ahead
check it into REL1_4.
I've tested every single one of the hook events, both with and without
the Syslog sample extension. This was in the REL1_4 branch; I think it
probably makes better sense to test in the 1.4 branch than with the HEAD
branch (since that's what's actually going out), but I'll do that too.
Unless you're checking in something that will never ever be supported in
the future, or is a hack specifically for a problem in the 1.4 branch
that does not exist in the next version, it needs to go in HEAD.
If it doesn't go into HEAD, it will not be there in the next version. It
would be pretty lame if we fix a bunch of bugs in 1.4 and have them come
back in 1.5, wouldn't it? :)
I generally prefer to see things checked in on HEAD *first* and *then*
on REL1_4. If you prepare the code on both and check them in a few
minutes apart in the other order, that's fine too; I do that pretty
frequently. But if somebody makes a change out of the blue on the
release branch, and many hours layer there's no sign of it on the
development branch, big warning signs flash in my head and my
inclination is to pull it pending further work -- if it's not ready for
the development branch, it's probably not ready for production.
Keep in mind that we're pulling updates from REL1_4 (manually, but
frequently) on our live wikis. It's not a testing branch; it's in
production.
-- brion vibber (brion @
pobox.com)