Bug 49483 – Remove CSS Variable support

• Component: CSS
• Status: RESOLVED FIXED

[Alexey Proskuryakov]

It's been disabled for a long time, and doesn't seem to be gaining much traction.

[Alexey Proskuryakov]

Created attachment 73789 [details]
proposed patch

[Alexey Proskuryakov]

I'm not sure about some code in CSSParserValues and CSSPrimitiveValue which Hyatt added at the same time as CSS Variables. It seems to be still used (at least partially), but maybe it's too complicated for what we need now?

[Peter Beverloo]

You might want to take note of bug 46594, filed not too long ago, which talks about reviving the implementation.

[Dimitri Glazkov (Google)]

(In reply to comment #3)
> You might want to take note of bug 46594, filed not too long ago, which talks about reviving the implementation.

Yup, I am very much interested in keeping and improving CSS variables. From talking with dhyatt, he is too.

[Alexey Proskuryakov]

> Yup, I am very much interested in keeping and improving CSS variables. 

I unsuccessfully tried to reach you on IRC yesterday, and it's a mistake on my part that I didn't CC you. 

> From talking with dhyatt, he is too.

Hyatt gave me a green light to remove the code yesterday. The problem is that variables greatly complicate ownership model of CSSOM objects, and I'd like to get them out of the way while straightening it up.

Also, my understanding was that CSS variables did't have much chance of going anywhere beyond WebKit, according to stated positions of other vendors and CSS working group members. I haven't participated in any of these discussions though, so I may be way off.

[Dimitri Glazkov (Google)]

(In reply to comment #5)
> > Yup, I am very much interested in keeping and improving CSS variables. 
> 
> I unsuccessfully tried to reach you on IRC yesterday, and it's a mistake on my part that I didn't CC you. 
> 
> > From talking with dhyatt, he is too.
> 
> Hyatt gave me a green light to remove the code yesterday. The problem is that variables greatly complicate ownership model of CSSOM objects, and I'd like to get them out of the way while straightening it up.
> 
> Also, my understanding was that CSS variables did't have much chance of going anywhere beyond WebKit, according to stated positions of other vendors and CSS working group members. I haven't participated in any of these discussions though, so I may be way off.

I don't mind yanking the code for now, especially if that helps refactoring. But I still want to bring them back and will work hard to clear any non-technical hurdles around it. And to be clear, this vendor really wants it implemented! :)

[WebKit Commit Bot]

Comment on attachment 73789 [details]
proposed patch

Rejecting patch 73789 from commit-queue.

Failed to run "['./WebKitTools/Scripts/webkit-patch', '--status-host=queues.webkit.org', '--bot-id=abarth-cq-sl', 'land-attachment', '--force-clean', '--ignore-builders', '--non-interactive', '--parent-command=commit-queue', 73789]" exit_code: 2
Last 500 characters of output:
'LayoutTests/platform/mac/fast/css/variables/alternate-syntax/variable-iteration-test-expected.png'
patching file LayoutTests/platform/mac/fast/css/variables/alternate-syntax/variable-iteration-test-expected.txt
rm 'LayoutTests/platform/mac/fast/css/variables/alternate-syntax/variable-iteration-test-expected.txt'
patching file LayoutTests/platform/qt/Skipped

Failed to run "[u'/Users/abarth/git/webkit-queue/WebKitTools/Scripts/svn-apply', u'--reviewer', u'David Hyatt', u'--force']" exit_code: 1

Full output: http://queues.webkit.org/results/6083003

[Alexey Proskuryakov]

Committed <http://trac.webkit.org/changeset/72116>.

[Alexey Proskuryakov]

Windows build fix <http://trac.webkit.org/changeset/72121>.