Bug 98420

Summary: Unprefix the flexbox CSS properties
Product: WebKit Reporter: Ojan Vafai <ojan>
Component: CSSAssignee: Dean Jackson <dino>
Status: RESOLVED FIXED    
Severity: Normal CC: 7raivis, abarth, benjamin, commit-queue, dino, glenn, m.goleb+bugzilla, paulirish, phiw2, rik, simon.fraser, syoichi, tabatkins, tony, webkit-bug-importer
Priority: P2 Keywords: InRadar
Version: 528+ (Nightly build)   
Hardware: Unspecified   
OS: Unspecified   
Bug Depends on:    
Bug Blocks: 98419    
Attachments:
Description Flags
Patch none

Description Ojan Vafai 2012-10-04 09:53:32 PDT
At this point we're just waiting for another browser vendor to ship the new flexboxes so that we can do some basic compatibility testing. None of the remaining open bugs at https://bugs.webkit.org/showdependencytree.cgi?id=62048&hide_resolved=1 seem worth stalling unprefixing on.
Comment 1 Tony Chang 2012-10-04 10:12:36 PDT
It was just enabled in mozilla-inbound (https://wiki.mozilla.org/Tree_Rules/Inbound):
https://bugzilla.mozilla.org/show_bug.cgi?id=666041#c124
Comment 2 Glenn Adams 2012-10-04 18:26:01 PDT
has the CSS WG resolved to unprefix flexbox yet? if not, should we WK jump the gun on them?
Comment 3 Tab Atkins 2012-10-04 19:47:56 PDT
(In reply to comment #2)
> has the CSS WG resolved to unprefix flexbox yet? if not, should we WK jump the gun on them?

Flexbox is in CR, which by definition means that we're okay with unprefixing.
Comment 4 Glenn Adams 2012-10-04 22:15:59 PDT
(In reply to comment #3)
> (In reply to comment #2)
> > has the CSS WG resolved to unprefix flexbox yet? if not, should we WK jump the gun on them?
> 
> Flexbox is in CR, which by definition means that we're okay with unprefixing.

excellent
Comment 5 Paul Irish 2012-12-21 15:35:10 PST
Update: It's now landed in FF nightly builds: http://blog.dholbert.org/2012/12/css3-flexbox-enabled-in-nightlies-ready.html
Comment 6 Binyamin 2013-02-20 11:20:50 PST
Landed in FF19 https://bugzilla.mozilla.org/show_bug.cgi?id=801098
Comment 7 Michał Gołębiowski-Owczarek 2013-08-11 15:19:11 PDT
Both Firefox & Chrome have unprefixed Flexbox properties already, the same stands for the IE11 beta. Is there anything blocking unprefixing in WebKit?
Comment 8 Michał Gołębiowski-Owczarek 2013-08-11 15:23:05 PDT
Worth noting is that both Firefox & Chrome have unprefixed Flexbox in their stable versions not just nightlies. So it's 3 browsers in stable versions (Opera 15 included) and IE in a preview. WebKit is the last to keep prefixes.

Considering WebKit Flexbox implementation seems stable and is richer than Firefox one, I think it'd be nice to unprefix in time for Safari 6.1/7 (and maybe iOS7?), it would make web developers' lives a little bit easier.
Comment 9 Michał Gołębiowski-Owczarek 2013-08-11 15:33:43 PDT
Sorry, it seems I lied about Chrome having uprefixed flexbox in stable (v28), though it seems to have it in beta (v29) according to:
https://code.google.com/p/chromium/issues/detail?id=249111
and I can confirm it's present in Canary (v30)
Comment 10 Radar WebKit Bug Importer 2013-08-26 10:53:58 PDT
<rdar://problem/14835804>
Comment 11 Michał Gołębiowski-Owczarek 2014-09-01 10:53:55 PDT
What's the status of this bug? All other browsers have unprefixed Flexbox more than 10 months ago in their stable versions and in Safari it will happen not sooner than in a year; it's almost a two-years gap!

Is there anything blocking unprefixing?
Comment 12 Dean Jackson 2014-09-10 16:28:55 PDT
(In reply to comment #11)
> What's the status of this bug? All other browsers have unprefixed Flexbox more than 10 months ago in their stable versions and in Safari it will happen not sooner than in a year; it's almost a two-years gap!
> 
> Is there anything blocking unprefixing?

Nope. I'm taking it on now.
Comment 13 Dean Jackson 2014-09-12 12:05:10 PDT
Created attachment 238044 [details]
Patch
Comment 14 Benjamin Poulain 2014-09-12 12:36:11 PDT
(In reply to comment #12)
> (In reply to comment #11)
> > What's the status of this bug? All other browsers have unprefixed Flexbox more than 10 months ago in their stable versions and in Safari it will happen not sooner than in a year; it's almost a two-years gap!
> > 
> > Is there anything blocking unprefixing?
> 
> Nope. I'm taking it on now.

IMHO, we should also import all the flexbox tests from blink.
Comment 15 Tony Chang 2014-09-12 12:37:54 PDT
Comment on attachment 238044 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=238044&action=review

> Source/WebCore/ChangeLog:20
> +        - justify

justify-content

> Source/WebCore/css/CSSComputedStyleDeclaration.cpp:292
> +    CSSPropertyJustifySelf,

justify-self seems to be part of the CSS Alignment spec, not the flexbox spec.  Did you mean to include this change?  http://dev.w3.org/csswg/css-align/
Comment 16 Dean Jackson 2014-09-12 12:44:18 PDT
(In reply to comment #15)
> (From update of attachment 238044 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=238044&action=review
> 
> > Source/WebCore/ChangeLog:20
> > +        - justify
> 
> justify-content

Yes.

> 
> > Source/WebCore/css/CSSComputedStyleDeclaration.cpp:292
> > +    CSSPropertyJustifySelf,
> 
> justify-self seems to be part of the CSS Alignment spec, not the flexbox spec.  Did you mean to include this change?  http://dev.w3.org/csswg/css-align/

You're right, thanks!
Comment 17 WebKit Commit Bot 2014-09-12 13:46:19 PDT
Comment on attachment 238044 [details]
Patch

Clearing flags on attachment: 238044

Committed r173572: <http://trac.webkit.org/changeset/173572>
Comment 18 WebKit Commit Bot 2014-09-12 13:46:25 PDT
All reviewed patches have been landed.  Closing bug.
Comment 19 Dean Jackson 2014-09-12 14:43:56 PDT
Committed r173579: <http://trac.webkit.org/changeset/173579>