Add support for DOM4's XMLDocument interface

Add support for DOM4's XMLDocument interface

Add support for DOM4's XMLDocument interface:
http://dom.spec.whatwg.org/#xmldocument

DOMImplementation.createDocument() now returns an XMLDocument instead of a
Document as per:
http://dom.spec.whatwg.org/#dom-domimplementation-createdocument

DOMParser.parseFromString() now returns an XMLDocument instead of a Document
when the input type is an xml type, as per:
http://domparsing.spec.whatwg.org/#dom-domparser-parsefromstring

This behavior is consistent with Firefox 26 and IE 11.
The compatibility risk is low since an XMLDocument is identical to a Document.

R=tkent, arv
BUG=238366
BUG=238372
TEST=fast/dom/DOMImplementation/createDocument-XMLDocument.html

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=165607

[eseidel, 2014-01-22 01:28:02]

We looked at doing this loooong ago:
https://bugs.webkit.org/show_bug.cgi?id=9063
https://bugs.webkit.org/show_bug.cgi?id=36480

There are probably more interesting ones searching for XMLDocument in
bugs.webkit.org.

I'm surprised we're adding this now.  I thought XMLDocument.load was on its way
out...

[Inactive, 2014-01-22 01:58:21]

On 2014/01/22 01:28:02, eseidel wrote:
> We looked at doing this loooong ago:
> https://bugs.webkit.org/show_bug.cgi?id=9063
> https://bugs.webkit.org/show_bug.cgi?id=36480
> 
> There are probably more interesting ones searching for XMLDocument in
> http://bugs.webkit.org.
> 
> I'm surprised we're adding this now.  I thought XMLDocument.load was on its
way
> out...

I know XMLDocument has been around for a long time but I believe the situation
has changed. XMLDocument is now part of DOM4 and is implemented by both Firefox
and IE11. The discussion about standardizing XMLDocument can be found here:
https://www.w3.org/Bugs/Public/show_bug.cgi?id=14037

The DOM Parsing spec refers to XMLDocument as well:
http://domparsing.spec.whatwg.org/#dom-domparser-parsefromstring

So does the HTML5 specification which now specifies XMLDocument.load():
http://www.whatwg.org/specs/web-apps/current-work/#xmldocument

[arv (Not doing code reviews), 2014-01-22 15:24:16]

LGTM

Thanks for working on cleaning up this mess.

https://codereview.chromium.org/144063004/diff/80001/LayoutTests/fast/dom/DOM...
File LayoutTests/fast/dom/DOMImplementation/createDocument-XMLDocument.html
(right):

https://codereview.chromium.org/144063004/diff/80001/LayoutTests/fast/dom/DOM...
LayoutTests/fast/dom/DOMImplementation/createDocument-XMLDocument.html:12:
shouldBe('document.implementation.createDocument("", "").__proto__',
'XMLDocument.prototype');
Can you also add a test that XMLDocument.prototype.__proto__ ===
Document.prototype?

https://codereview.chromium.org/144063004/diff/80001/Source/core/dom/XMLDocum...
File Source/core/dom/XMLDocument.h (right):

https://codereview.chromium.org/144063004/diff/80001/Source/core/dom/XMLDocum...
Source/core/dom/XMLDocument.h:41: static PassRefPtr<XMLDocument>
createXHTML(const DocumentInit& initializer = DocumentInit())
Is there a need for createSVG?

[Inactive, 2014-01-22 15:39:26]

https://codereview.chromium.org/144063004/diff/80001/LayoutTests/fast/dom/DOM...
File LayoutTests/fast/dom/DOMImplementation/createDocument-XMLDocument.html
(right):

https://codereview.chromium.org/144063004/diff/80001/LayoutTests/fast/dom/DOM...
LayoutTests/fast/dom/DOMImplementation/createDocument-XMLDocument.html:12:
shouldBe('document.implementation.createDocument("", "").__proto__',
'XMLDocument.prototype');
On 2014/01/22 15:24:19, arv wrote:
> Can you also add a test that XMLDocument.prototype.__proto__ ===
> Document.prototype?

Good idea, will do.

https://codereview.chromium.org/144063004/diff/80001/Source/core/dom/XMLDocum...
File Source/core/dom/XMLDocument.h (right):

https://codereview.chromium.org/144063004/diff/80001/Source/core/dom/XMLDocum...
Source/core/dom/XMLDocument.h:41: static PassRefPtr<XMLDocument>
createXHTML(const DocumentInit& initializer = DocumentInit())
On 2014/01/22 15:24:19, arv wrote:
> Is there a need for createSVG?

Not yet since we still have an SVGDocument subclass. To construct an SVG
document, you would call SVGDocument::create().

SVGDocument is gone from the IDL in SVG2 spec but I do not know yet how we want
it to be structured on native side. Right now, I tried to do as little change to
SVG as possible. The only thing I did is make SVGDocument inherit from
XMLDocument as createDocument() needs to return an XMLDocument (even for svg
namespace).

[Inactive, 2014-01-22 15:51:51]

https://codereview.chromium.org/144063004/diff/80001/LayoutTests/fast/dom/DOM...
File LayoutTests/fast/dom/DOMImplementation/createDocument-XMLDocument.html
(right):

https://codereview.chromium.org/144063004/diff/80001/LayoutTests/fast/dom/DOM...
LayoutTests/fast/dom/DOMImplementation/createDocument-XMLDocument.html:12:
shouldBe('document.implementation.createDocument("", "").__proto__',
'XMLDocument.prototype');
On 2014/01/22 15:24:19, arv wrote:
> Can you also add a test that XMLDocument.prototype.__proto__ ===
> Document.prototype?

Done.

[tkent, 2014-01-22 23:49:31]

public API changes LGTM.

[Inactive, 2014-01-22 23:59:15]

Eric, do you still have concerns or can I land this?

[eseidel, 2014-01-23 00:08:56]

lgtm

If arv is OK with it, I am.

[eseidel, 2014-01-23 00:10:20]

https://codereview.chromium.org/144063004/diff/190001/Source/core/dom/XMLDocu...
File Source/core/dom/XMLDocument.idl (right):

https://codereview.chromium.org/144063004/diff/190001/Source/core/dom/XMLDocu...
Source/core/dom/XMLDocument.idl:30: // boolean load(DOMString url);
I thought we had this already on Document?

[Inactive, 2014-01-23 00:24:56]

On 2014/01/23 00:10:20, eseidel wrote:
>
https://codereview.chromium.org/144063004/diff/190001/Source/core/dom/XMLDocu...
> File Source/core/dom/XMLDocument.idl (right):
> 
>
https://codereview.chromium.org/144063004/diff/190001/Source/core/dom/XMLDocu...
> Source/core/dom/XMLDocument.idl:30: // boolean load(DOMString url);
> I thought we had this already on Document?

I have just checked again, we don't have any load() method on Document.

[commit-bot: I haz the power, 2014-01-23 00:25:27]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ch.dumez@samsung.com/144063004/190001

[eseidel, 2014-01-23 00:27:53]

I'm not actually sure we want to ever expose XMLDocument.load().  It looks like
it's a synchronous API? Which seems like bad-news bears.

[arv (Not doing code reviews), 2014-01-23 00:45:27]

I think we had load before but removed it. We should not add it if we don't
have it. We should just get it removed from the spec instead.
On Jan 22, 2014 7:27 PM, <eseidel@chromium.org> wrote:

> I'm not actually sure we want to ever expose XMLDocument.load().  It looks
> like
> it's a synchronous API? Which seems like bad-news bears.
>
> https://codereview.chromium.org/144063004/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to blink-reviews+unsubscribe@chromium.org.

[Inactive, 2014-01-23 00:49:32]

On 2014/01/23 00:45:27, arv wrote:
> I think we had load before but removed it. We should not add it if we don't
> have it. We should just get it removed from the spec instead.

Ok, I will get rid of this FIXME before landing. We should probably close the
associated bug too.

[commit-bot: I haz the power, 2014-01-23 00:57:02]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/ch.dumez@samsung.com/144063004/310001

[commit-bot: I haz the power, 2014-01-23 04:36:16]

Change committed as 165607