Closed Bug 1230672 Opened 9 years ago Closed 9 years ago

Add support for column layout on <fieldset>

Categories

(Core :: Layout, defect, P3)

defect

Tracking

()

RESOLVED FIXED
mozilla46
Tracking Status
firefox46 --- fixed

People

(Reporter: MatsPalmgren_bugz, Assigned: MatsPalmgren_bugz)

References

Details

Attachments

(4 files, 1 obsolete file)

      No description provided.
Depends on: 1230207
Attached patch wip (obsolete) — Splinter Review
As I was writing reftests for <fieldset> with column layout I found this
existing bug.  I'll fix it here since it's just a one-liner.
Attachment #8696061 - Attachment is obsolete: true
Attachment #8696290 - Flags: review?(bzbarsky)
Comment on attachment 8696290 [details] [diff] [review]
part 1 - Make '-moz-column-fill:auto' work also when 'overflow' isn't 'visible'.

r=me
Attachment #8696290 - Flags: review?(bzbarsky) → review+
Comment on attachment 8696292 [details] [diff] [review]
part 3 - Add support for multicol layout on <fieldset> elements.

r=me, though the code duplication with ConstructBlock is a bit unfortunate.
Attachment #8696292 - Flags: review?(bzbarsky) → review+
Backed out just the part 2 reftest changes in https://hg.mozilla.org/integration/mozilla-inbound/rev/eec861bce6d4, because we like to set you up to fail by not running either OS X 10.10 or WinXP tests on try unless you beg for them, and 10.10 didn't like the reftest change, https://treeherder.mozilla.org/logviewer.html#?job_id=18587815&repo=mozilla-inbound

Your trychooser patch's hint is out of date, what you want in the current era to get all desktop reftests is "reftest[Ubuntu,10.6,10.10,Windows XP,Windows 7,Windows 8]".
(In reply to Phil Ringnalda (:philor) from comment #10)
> Backed out just the part 2 reftest changes in

Thanks philor.

> we like to set you up to fail by not running either OS X 10.10 or WinXP
> tests on try unless you beg for them,

Yeah, I really wish our tools could be smarter here.  Like if the push
contains changes to a reftest, or changes code under layout/ or gfx/ then
"-p all" really means "-p every-single-platform-we-run-reftests-on".

> Your trychooser patch's hint is out of date, what you want in the current
> era to get all desktop reftests is "reftest[Ubuntu,10.6,10.10,Windows
> XP,Windows 7,Windows 8]".

What I usually want is all platforms we run reftests on, including
Android, B2G, e10s variations etc.  I don't have time keep track of
which platform/test-suite combinations we run because that changes
quite often, so I really wish there was new keyword for this,
or if Try itself could be smarter about "all" as I said above.
The reftest failures on OSX 10.10 are anti-aliasing differences.
Adding fuzzyness for victory here:
https://treeherder.mozilla.org/#/jobs?repo=try&revision=78dc347c00b0
Crap, I forgot about bug 1231995, until that's fixed you aren't actually going to get 10.10 reftest on try, just use m-i as your tryserver.
Thanks, I was a little puzzled why I didn't get any 10.10 runs there...
I retriggered a few manually, maybe that works.
Flags: in-testsuite+
Part 2 backed out in https://hg.mozilla.org/integration/mozilla-inbound/rev/f3131740bb5d because the reftest.list syntax is apparently wrong:

https://treeherder.mozilla.org/logviewer.html#?job_id=18613469&repo=mozilla-inbound
Flags: needinfo?(mats)
Flags: in-testsuite?
Flags: in-testsuite+
Looks green on 10.10 now.
Flags: needinfo?(mats)
Flags: in-testsuite?
Flags: in-testsuite+
No longer depends on: 1232853
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: