ship the 'isolation' CSS property (enable the layout.css.isolation.enabled preference)

RESOLVED FIXED in mozilla36

Status

()

Core
CSS Parsing and Computation
RESOLVED FIXED
4 years ago
3 years ago

People

(Reporter: dbaron, Assigned: Rik Cabanier)

Tracking

(Blocks: 1 bug, {dev-doc-complete})

Trunk
mozilla36
dev-doc-complete
Points:
---
Dependency tree / graph
Bug Flags:
qe-verify -

Firefox Tracking Flags

(relnote-firefox 36+)

Details

Attachments

(1 attachment, 1 obsolete attachment)

(Reporter)

Description

4 years ago
We should ship the 'isolation' CSS property.  Bugs that need to be fixed before we do this should block this bug.

Initial implementation of the property was in bug 1077872.

(Is anything else needed here, or can we just do this?)
Flags: needinfo?(cabanier)
(Assignee)

Updated

4 years ago
Assignee: nobody → cabanier
Flags: needinfo?(cabanier)
(Assignee)

Comment 1

4 years ago
Created attachment 8514691 [details] [diff] [review]
Turn on 'isolation' by default
(Assignee)

Comment 2

4 years ago
Comment on attachment 8514691 [details] [diff] [review]
Turn on 'isolation' by default

Sorry, I thought it was turned on by default.

Try build: https://treeherder.mozilla.org/ui/#/jobs?repo=try&revision=727cf5fb74f8
Attachment #8514691 - Flags: review?(dbaron)
(Reporter)

Comment 3

4 years ago
Was there thread about shipping this on dev-platform?  If not, could you start one?

Are there any interop risks here in the details (e.g., interacting with painting order or stacking contexts), or is that mostly with mix-blend-mode?
Flags: needinfo?(cabanier)
(Reporter)

Comment 4

4 years ago
Oh, and what does test coverage look like?  (And are any of those tests in a W3C test suite?)
(Assignee)

Comment 5

4 years ago
(In reply to David Baron [:dbaron] (UTC-7) (needinfo? for questions) (away/busy Oct. 24-31) from comment #3)
> Was there thread about shipping this on dev-platform?  If not, could you
> start one?

No. Will do so asap.

> Are there any interop risks here in the details (e.g., interacting with
> painting order or stacking contexts), or is that mostly with mix-blend-mode?

This property just creates a stacking context. It should not have any side effects; the try bots should catch this.
Flags: needinfo?(cabanier)
(Assignee)

Comment 6

4 years ago
(In reply to David Baron [:dbaron] (UTC-7) (needinfo? for questions) (away/busy Oct. 24-31) from comment #3)
> Was there thread about shipping this on dev-platform?  If not, could you
> start one?
> 
> Are there any interop risks here in the details (e.g., interacting with
> painting order or stacking contexts), or is that mostly with mix-blend-mode?

(In reply to David Baron [:dbaron] (UTC-7) (needinfo? for questions) (away/busy Oct. 24-31) from comment #4)
> Oh, and what does test coverage look like?  (And are any of those tests in a
> W3C test suite?)

There are a couple of tests for isolation which Firefox is passing.
For the implementation report that I will generate for Peter, I might add a couple more test especially for SVG content and mixed SVG/HTML
(Assignee)

Comment 7

4 years ago
I'm getting some strange errors from the try run: 
https://treeherder.mozilla.org/ui/#/jobs?repo=try&revision=727cf5fb74f8

I'm not able to figure out why this is happening. Can you take a quick look?
Flags: needinfo?(dbaron)
(Reporter)

Comment 8

4 years ago
Most likely problem is that property_database.js uses 'isolated' instead of 'isolate'.  See bug 1077872 comment 6, etc.  (Could be others, though.)
Flags: needinfo?(dbaron)
(Assignee)

Updated

4 years ago
Depends on: 1092264
(Reporter)

Comment 10

4 years ago
(In reply to Rik Cabanier from comment #9)
> try build:
> https://treeherder.mozilla.org/ui/#/jobs?repo=try&revision=f4700ff0a084

You don't want the nsCSSPropList.h change in https://hg.mozilla.org/try/rev/5212846e4031 .  kIsolationKTable already has an eCSSKeyword_auto entry, and code (e.g., in nsRuleNode.cpp) is set up to expect auto to be an eCSSUnit_Enumerated value rather than an eCSSUnit_Auto value.
(Assignee)

Comment 11

4 years ago
Created attachment 8515192 [details] [diff] [review]
Turn on 'isolation' by default

Sorry. I didn't notice that line was still there.

try build (which includes fix for 1092264):
https://treeherder.mozilla.org/ui/#/jobs?repo=try&revision=ef08bc77c910
Attachment #8514691 - Attachment is obsolete: true
Attachment #8514691 - Flags: review?(dbaron)
Attachment #8515192 - Flags: review?(dbaron)
(Reporter)

Updated

4 years ago
Attachment #8515192 - Flags: review?(dbaron) → review+
https://hg.mozilla.org/mozilla-central/rev/18b705c478a0
Status: NEW → RESOLVED
Last Resolved: 4 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla36
Keywords: dev-doc-needed
Is this worth mentioning in the release notes?

Sebastian
Blocks: 913153
Release Note Request (optional, but appreciated)
[Why is this notable]: Activation of a new CSS property
[Suggested wording]: (I will suggest something later)
[Links (documentation, blog post, etc)]: https://developer.mozilla.org/en-US/docs/Web/CSS/isolation (still a WIP)
relnote-firefox: --- → ?
relnote-firefox: ? → 36+
Flags: qe-verify-
You need to log in before you can comment on or make changes to this bug.