From this point forward we will keep the tree closed for _all_
checkins until Monday, at which point we will make a go/no go
decision on starting builds for beta 2.
-- Mike
> All remaining checkins appear to be complete, and we are now frozen
> for beta 2.
There had been talk of landing fixes for these bugs for beta 2:
* Bug 337418 - [@ PR_EnumerateHostEnt] flaky wifi connection or vpn or
laptop sleeping
* Bug 404112 - Crash with zero-sized fonts in some versions of libpango
[@ libpango-1.0.so.0.1001.1 - gfxPangoFont::GetMetrics]
Also, Marcia commented in bug 390568 saying that if that bug is indeed
back, it might be a beta 2 blocker.
~reed
--
Reed Loden - <re...@reedloden.com>
Existed in b1 so we shouldn't take this. Can wait for b3
> * Bug 404112 - Crash with zero-sized fonts in some versions of libpango
> [@ libpango-1.0.so.0.1001.1 - gfxPangoFont::GetMetrics]
If Mats gives the go-ahead *today* they can check this in.
Also I gave approval for:
https://bugzilla.mozilla.org/show_bug.cgi?id=407323
Web compat regression fix for an issue that happened just a few days ago
>
> Also, Marcia commented in bug 390568 saying that if that bug is indeed
> back, it might be a beta 2 blocker.
Nope - sucks if it is back but I wouldn't block on it.
If anyone lands the two approved fixes above today please do clobber
right after so we can get a new nightly spin. Also if everyone can make
sure to check for updates to dogfood the latest nightly that would help
a bunch.
Overall awesome work everyone - the new url bar perf fixes look awesome.
Beta2 should be great!
Schrep
Existed in b1 so we shouldn't take this. Can wait for b3
> * Bug 404112 - Crash with zero-sized fonts in some versions of libpango
> [@ libpango-1.0.so.0.1001.1 - gfxPangoFont::GetMetrics]
If Mats gives the go-ahead *today* they can check this in.
Also I gave approval for:
https://bugzilla.mozilla.org/show_bug.cgi?id=407323
Web compat regression fix for an issue that happened just a few days ago
>
> Also, Marcia commented in bug 390568 saying that if that bug is indeed
> back, it might be a beta 2 blocker.
Nope - sucks if it is back but I wouldn't block on it.
This looks like it is not something we should take for b2.
> Also I gave approval for:
>
> https://bugzilla.mozilla.org/show_bug.cgi?id=407323
Landed
New bug: https://bugzilla.mozilla.org/show_bug.cgi?id=406769
Fix approved - we should also get this one in today. Could use some
help getting it landed.
Anything else on folks radar? Once 406769 lands I think we are good to go!
Mike
This looks like it is not something we should take for b2.
> Also I gave approval for:
>
> https://bugzilla.mozilla.org/show_bug.cgi?id=407323
Landed
I've landed this on the trunk and triggered new nightlies.
Gavin
Fantastic- thanks Gavin.
Unless anyone has anything else - this is it for b2..
Mike
Fantastic- thanks Gavin.
Brendan has asked for M10 approval in a XS-JSON data leak bug:
https://bugzilla.mozilla.org/show_bug.cgi?id=376957
The patch is a one-line follow-up to a previous patch that landed on Dec
3 which didn't completely solve the problem, Brendan writes:
> Would like to get this into Fx3b2 to match the other changes, so we get cleaner
> yes/no test results and don't have to wait for b3 to find out
Patch is here:
https://bugzilla.mozilla.org/attachment.cgi?id=292361&action=edit
cheers,
mike
Mike
-----Original Message-----
From: Mike Beltzner [mailto:belt...@mozilla.com]
Sent: Monday, December 10, 2007 7:47 AM
To: Mike Schroepfer
Cc: dev-pl...@lists.mozilla.org
Subject: Re: Frozen for Beta 2
Mike Schroepfer wrote:
> Unless anyone has anything else - this is it for b2..
Brendan has asked for M10 approval in a XS-JSON data leak bug:
Done. The last commit is:
Gavin
Actually, for automation to work, we need to land
https://bugzilla.mozilla.org/show_bug.cgi?id=407670, create
shipped-locales file, first.
Patch coming up.
Axel
L10n tag date for Firefox 3 B2 should be 2007-12-10 08:30 PST.
Bonsai query since the l10n lock-down is
http://bonsai-l10n.mozilla.org/cvsquery.cgi?treeid=default&module=all&branch=HEAD&branchtype=match&dir=l10n&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=2007-12-10+00%3A00&maxdate=2007-12-10+08%3A30&cvsroot=%2Fl10n,
the last check-in is a whitespace fix for bookmarks.html for es-AR.
Localization count is now 29, pending QA, the exhibit keeping track is
on http://l10n.mozilla.org/~axel/Firefox3/Beta2/, my notes are on
http://wiki.mozilla.org/User:AxelHecht/Fx3B2_l10n_Log.
There are two localizations we shipped in B1 that are not in B2,
Georgian (ka) and Lithuanian (lt).
ka opted in, but has red builds
lt has red builds, but only one missing string, didn't opt in though.
That means, both had some kind of traction and could have likely shipped
with better planning and less back-and-forth for fx3b2 l10n.
New languages are gu-IN on windows, and ca, es-AR, eu, he, it, nb-NO,
pa-IN, pt-PT, ro, tr on all platforms.
Axel
And, actually, for the proper naming to work, we need to figure out how
we're going to get the patch in
https://bugzilla.mozilla.org/show_bug.cgi?id=407275 landed in a way that
works with build automation.
cheers,
mike
> Axel Hecht wrote:
>>> Done. The last commit is:
>>>
>>> http://bonsai.mozilla.org/cvsquery.cgi?
>>> date=explicit&mindate=2007-12-10+08%3A15&maxdate=2007-12-10+08%3A15
>>>
>>> Gavin
>>
>> Actually, for automation to work, we need to land
>> https://bugzilla.mozilla.org/show_bug.cgi?id=407670, create
>> shipped-locales file, first.
>
> And, actually, for the proper naming to work, we need to figure out
> how
> we're going to get the patch in
> https://bugzilla.mozilla.org/show_bug.cgi?id=407275 landed in a way
> that
> works with build automation.
I might be wrong, but I think the automation still uses a relbranch,
so they'll create the relbranch, land than, and then go. I hope.
-- Mike
Comments (specifically joduinn's comment #1) in that thread indicate
that you are, sadly, wrong, and that the automation includes the part
where they branch off of trunk. It's solvable, though, we just need to
chart a path and follow it (see comment #2 and #3).
cheers,
mike