Reminder that freeze is tonight. At the Gecko meeting today the
following large changes yet to land were identified:
1) URLClassifier (DCamp)
2) doc.all changes (Sicking)
3) x-site XMLHttpRequest (Sicking)
4) ShowModalDialog (JST)
5) cross origin wrappers (Bkap/Brendan)
6) xml:id
7) oncopy/onpaste (JST)
8) xpath generator (Sicking)
We'd like to do landing slots for these (in the above order) so we
don't have a last-minute dash to the freeze. Each of the above
changes should land separately from the other (e.g. with a full tbox
cycle between them). Other smaller changes are still clear to land
up to the freeze tonight. After the freeze we'll continue landing
these big changes. So:
a) Can the above named folks coordinate their landings appropriately?
b) If you have a big change can you make sure to get on this list?
c) We are gecko feature frozen as of tonight - so if you have a
feature or any change that is exposed to content you should either get
on this list or defer to 1.9.1 or moz2.
Reply here if you have any thoughts or questions.
Best,
Schrep
-Olli
I have SVG lighting filters ready to land, so would like to get on the list.
> c) We are gecko feature frozen as of tonight - so if you have a
> feature or any change that is exposed to content you should either get
> on this list or defer to 1.9.1 or moz2.
We have a number of SVG feature items which will be ready in the 1.9
timeframe:
* SVG tile filter
* SVG displacement filter
* SVG image filter
* SVG text character positioning
* SVG fonts
k That's probably well contained so go ahead and land that.
> > c) We are gecko feature frozen as of tonight - so if you have a
> > feature or any change that is exposed to content you should either get
> > on this list or defer to 1.9.1 or moz2.
>
> We have a number of SVG feature items which will be ready in the 1.9
> timeframe:
>
> * SVG tile filter
> * SVG displacement filter
> * SVG image filter
> * SVG text character positioning
> * SVG fonts
Do you mean will be ready after today? Today is the cutoff for this
kind of stuff....
How about early AM PDT assuming the dust has settled?
Bug 4821 you bet. We need to whip up some UI for that as well :-)
Bug 17754 doesn't look ready to go (waiting on reviews)...
For what it's worth, this is a fairly self-contained change, that doesn't affect
web sites unless users make use of this functionality. Depending on what the
reviewers think, and depending on how usable it is, we should consider approving
it in the next beta.
-Boris
I figure that I'll pop on IRC tomorrow and find a timeslot based on
that. It'll probably be around 12:00pm EST.
--
Blake Kaplan
You mean, well contained besides touching DOMClassInfo and the style
system? I'll look for a relatively quiet time to land it tonight.
>>> c) We are gecko feature frozen as of tonight - so if you have a
>>> feature or any change that is exposed to content you should either get
>>> on this list or defer to 1.9.1 or moz2.
>> We have a number of SVG feature items which will be ready in the 1.9
>> timeframe:
>>
>> * SVG tile filter
>> * SVG displacement filter
>> * SVG image filter
>> * SVG text character positioning
>> * SVG fonts
>
> Do you mean will be ready after today? Today is the cutoff for this
> kind of stuff....
Yes, after today. The resources for SVG are low, so things are ready
when they're ready. The filters we want to add since it will complete
the specified SVG 1.1 filters. SVG fonts are a frequently requested
feature, and character positioning is of particular interest to a group
who is working on implementing it.
Patches are in various states of completion for four of these five
items, with the exception being the relatively straightforward
displacement filter.
These changes are isolated to SVG, and given the current schedule are
still acceptable from a risk/benefit point of view by the SVG module
owner (me).
> These changes are isolated to SVG, and given the current schedule are
> still acceptable from a risk/benefit point of view by the SVG module
> owner (me).
While I respect your opinion as a module owner, this is not likely to
be acceptable from a schedule or driver standpoint. We have been
quite open and clear about feature freeze for gecko being M7
(formerly beta 1), since at least March. This in the newsgroups, the
wiki schedule, and the weekly Gecko 1.9 meetings. Past M7, all
checkins for 1.9 to Gecko will require driver approval, and while in
past releases we have made exceptions to feature freezes for SVG, I
do not believe we will do so for 1.9.
Gecko 1.9 has been two years in the making, the longest cycle without
a release off the trunk in the project's history, and eventually we
have to close the door to new features.
-- Mike
The SVG tile filter that tor referred to is ready to land. Or at least
will be within the next hour or so.
Robert.
Rob
> On 26-Jul-07, at 1:48 AM, T Rowley wrote:
> > These changes are isolated to SVG, and given the current schedule
are
> > still acceptable from a risk/benefit point of view by the SVG
module
> > owner (me).
> While I respect your opinion as a module owner, this is not likely
to
> be acceptable from a schedule or driver standpoint. We have been
> quite open and clear about feature freeze for gecko being M7
> (formerly beta 1), since at least March. This in the newsgroups,
the
> wiki schedule, and the weekly Gecko 1.9 meetings. Past M7, all
> checkins for 1.9 to Gecko will require driver approval, and while
in
> past releases we have made exceptions to feature freezes for SVG, I
> do not believe we will do so for 1.9.
> Gecko 1.9 has been two years in the making, the longest cycle
without
> a release off the trunk in the project's history, and eventually we
> have to close the door to new features.
> -- Mike
The tile filter (https://bugzilla.mozilla.org/show_bug.cgi?id=373572)
is ready to land. Can I get permission to land it? I don't have IRC
access?
Robert.
No. The tree is closed for the 1.9a7 release, and only special exceptions
will be made. You can checkin the new tests when the tree reopens after the
1.9a7 release.
--BDS
Can I get a slot to check in bug 319455 today? I wasn't able to get it in yesterday, and it'd be
good to get the schema changes in, as well as fix the history table before the effects of bug 332748
(expand history to 180 days).
Right. Note to self - don't make assumptions without looking at the
patch as it makes me say dumb things :-(.
> Yes, after today. The resources for SVG are low, so things are ready
> when they're ready. The filters we want to add since it will complete
> the specified SVG 1.1 filters. SVG fonts are a frequently requested
> feature, and character positioning is of particular interest to a group
> who is working on implementing it.
>
> Patches are in various states of completion for four of these five
> items, with the exception being the relatively straightforward
> displacement filter.
>
> These changes are isolated to SVG, and given the current schedule are
> still acceptable from a risk/benefit point of view by the SVG module
> owner (me).
Yep - every module is under resourced and everyone is really excited by
the improvements in SVG. The Gecko feature freeze has been communicated
for a long time now and this is the first I've heard of these. We need
to start closing the door on changes for Gecko (it's been almost 2 years
now) so we have a shot at shipping sometime soon. Exceptions are
always heard by release drivers but the default answer on things will be
no. There will be a 1.9.1 shortly after 1.9 which would also be a good
place for this.
Best,
Mike
Yep - go for it.
Yes. Tests are always fair game!
Right. So Benjamin and I said opposite things. To be safe let's hold
these until the tree re-opens. There has been enough thrash over the
last day...
> b) If you have a big change can you make sure to get on this list?
It's not a big change, but can I get approval to land bug #385397:
"history menu is slow to open with a large history"
This morning, schrep asked in the bug: "Any chance we can get this into
a7?"
I'd like to get it in, especially now that we've bumped history
expiration to be 180 days by default (bug #332748)
-Seth
Based on past experience with Mozilla freezes, we expected it to be
somewhat soft for at least the first round. Nevertheless, development
continued with the July timeframe in mind. We're a week or two away
from having a feature set which satisfies us, not bad given typical
software scheduling.
> There will be a 1.9.1 shortly after 1.9 which would also be a good
> place for this.
What do you mean by "shortly"? A couple months, a year, by the end of
the decade?
When will MOZILLA_1_9 branch?
1.9.1 will look much like 1.8.1 - so 6-9 months after 1.9 ships.
Branching is not yet figured out. Since Moz2 is moving to Hg we might
try and do 1.9.1 off of Hg and thus not branch cvs. We gotta get
everyone (build, gecko team, etc) together to figure that out soon...