Correctly reflect the 'crossOrigin' attribute.

Correctly reflect the 'crossOrigin' attribute.

The CORS settings attribute

 http://www.whatwg.org/specs/web-apps/current-work/#cors-settings-attribute

when reflected as an IDL attribute, should map a bare/empty
'crossorigin' attribute to "anonymous". And if set to an invalid
value, to "anonymous" also.

So, for the reflected "crossOrigin" IDL attribute on <script>, <img>
and <video> (HTMLMediaElement, really), these attribute states should
be observable via its getter.

R=
BUG=332346
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=164980

[sof, 2014-01-09 15:38:50]

Please take a look when you next have a moment.

The treatment of crossOrigin here brings it in-line with the spec; none of the
other engines implement reflection for crossOrigin (or correctly.) I've no
reason to think that would be a compat problem.

[arv (Not doing code reviews), 2014-01-13 14:35:08]

LGTM

https://codereview.chromium.org/131483006/diff/30001/LayoutTests/fast/dom/HTM...
File LayoutTests/fast/dom/HTMLImageElement/image-crossOrigin-expected.txt
(right):

https://codereview.chromium.org/131483006/diff/30001/LayoutTests/fast/dom/HTM...
LayoutTests/fast/dom/HTMLImageElement/image-crossOrigin-expected.txt:9: PASS
img.removeAttribute('crossorigin'); img.crossOrigin is ""
I found this strange but I double checked with the spec and the above seems
correct.

https://codereview.chromium.org/131483006/diff/30001/LayoutTests/fast/dom/HTM...
LayoutTests/fast/dom/HTMLImageElement/image-crossOrigin-expected.txt:14: PASS
img.crossOrigin = 'foo'; img.getAttribute('crossorigin') is "foo"
This is also very strange. Maybe the spec should change to not change the value?

[sof, 2014-01-13 14:45:58]

https://codereview.chromium.org/131483006/diff/30001/LayoutTests/fast/dom/HTM...
File LayoutTests/fast/dom/HTMLImageElement/image-crossOrigin-expected.txt
(right):

https://codereview.chromium.org/131483006/diff/30001/LayoutTests/fast/dom/HTM...
LayoutTests/fast/dom/HTMLImageElement/image-crossOrigin-expected.txt:14: PASS
img.crossOrigin = 'foo'; img.getAttribute('crossorigin') is "foo"
On 2014/01/13 14:35:09, arv wrote:
> This is also very strange. Maybe the spec should change to not change the
value?

I do wonder why crossOrigin wasn't given an IDL enumerated type, which would
take care of this (..and more, it would throw.)

(FTR, the reason why there's no filtering of the value put is from

 http://www.whatwg.org/specs/web-apps/current-work/#limited-to-only-known-values

";and on setting, the content atribute must be set to the specified new value.")

[arv (Not doing code reviews), 2014-01-13 14:51:50]

On 2014/01/13 14:45:58, sof wrote:
>
https://codereview.chromium.org/131483006/diff/30001/LayoutTests/fast/dom/HTM...
> File LayoutTests/fast/dom/HTMLImageElement/image-crossOrigin-expected.txt
> (right):
> 
>
https://codereview.chromium.org/131483006/diff/30001/LayoutTests/fast/dom/HTM...
> LayoutTests/fast/dom/HTMLImageElement/image-crossOrigin-expected.txt:14: PASS
> img.crossOrigin = 'foo'; img.getAttribute('crossorigin') is "foo"
> On 2014/01/13 14:35:09, arv wrote:
> > This is also very strange. Maybe the spec should change to not change the
> value?
> 
> I do wonder why crossOrigin wasn't given an IDL enumerated type, which would
> take care of this (..and more, it would throw.)
> 
> (FTR, the reason why there's no filtering of the value put is from
> 
> 
http://www.whatwg.org/specs/web-apps/current-work/#limited-to-only-known-values
> 
> ";and on setting, the content atribute must be set to the specified new
value.")

Yeah, I had to reread that. I think this is a bad API design but at least this
patch makes us more compliant.

[sof, 2014-01-13 15:14:38]

On 2014/01/13 14:51:50, arv wrote:
> On 2014/01/13 14:45:58, sof wrote:
> >
>
https://codereview.chromium.org/131483006/diff/30001/LayoutTests/fast/dom/HTM...
> > File LayoutTests/fast/dom/HTMLImageElement/image-crossOrigin-expected.txt
> > (right):
> > 
> >
>
https://codereview.chromium.org/131483006/diff/30001/LayoutTests/fast/dom/HTM...
> > LayoutTests/fast/dom/HTMLImageElement/image-crossOrigin-expected.txt:14:
PASS
> > img.crossOrigin = 'foo'; img.getAttribute('crossorigin') is "foo"
> > On 2014/01/13 14:35:09, arv wrote:
> > > This is also very strange. Maybe the spec should change to not change the
> > value?
> > 
> > I do wonder why crossOrigin wasn't given an IDL enumerated type, which would
> > take care of this (..and more, it would throw.)
> > 
> > (FTR, the reason why there's no filtering of the value put is from
> > 
> > 
>
http://www.whatwg.org/specs/web-apps/current-work/#limited-to-only-known-values
> > 
> > ";and on setting, the content atribute must be set to the specified new
> value.")
> 
> Yeah, I had to reread that. I think this is a bad API design but at least this
> patch makes us more compliant.

I will follow up on this separately. Thanks for the careful review.

[commit-bot: I haz the power, 2014-01-13 15:15:27]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sigbjornf@opera.com/131483006/30001

[commit-bot: I haz the power, 2014-01-13 16:22:12]

Change committed as 164980