bindings: Uses V8's named property interceptor for Window.

bindings: Uses V8's named property interceptor for Window.

Other attributes have already been using V8's named property interceptor.
Window is only an exception.  Let's use it for Window, too.

This CL intentionally changes the look-up behavior.
Sample code:
  iframe = document.createElement('iframe');
  iframe.name = 'alert';
  document.body.appendChild(iframe);
Before:
  window.alert => iframe with the name 'alert'
After:
  window.alert => built-in alert function

According to the spec, (non-named, regular) attributes/methods must have priority over named properties.

See "The named property visibility algorithm" at
http://heycam.github.io/webidl/#indexed-and-named-properties
The order of look-up for non-[OverrideBuiltins] interfaces (including Window) must be:
  1. Own properties.
  2. Properties from the prototype chain.
  3. Named properties.

The Intent-to-Implement-and-Ship is here:
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/jlmsE3fDCNg

BUG=43394, 516274
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=201225

[Yuki, 2015-05-15 07:35:39]

yukishiino@chromium.org changed reviewers:
+ haraken@chromium.org

[Yuki, 2015-05-15 07:35:39]

Could you review this CL?

[haraken, 2015-05-15 12:03:39]

haraken@chromium.org changed reviewers:
+ jochen@chromium.org, verwaest@chromium.org

[haraken, 2015-05-15 12:03:40]

cc: Toon + Jochen

The security checks are complicated, and I might want to more deeply understand
the background of this CL. Let's chat offline on Monday :)

[Toon Verwaest, 2015-05-15 12:35:02]

I'm not familiar with IDL so I haven't looked at the CL that deeply. Note though
that FF doesn't leak iframe names in the parent iframe at all, so the example
isn't very helpful.

iframe = document.createElement('iframe');
iframe.name = "bla";
bla
-> ReferenceError: bla is not defined

[jochen (gone - plz use gerrit), 2015-05-15 15:37:16]

you have to document.body.appendChild(iframe) first I think, no?

In any case, I think we shouldn't do this right before the branch point, and at
least send an heads-up before landing this to blink-dev

code wise, the changes look good if we want this behavior

please also reference the spec+section from the CL description.

I don't think this is related to the crash bug btw

[Toon Verwaest, 2015-05-15 16:14:44]

Yeah, sorry, I forgot that line. But it doesn't matter: even if you add it they
don't pollute their parent global.

[Yuki, 2015-05-19 09:46:52]

Thanks for the review comments, and I'm sorry that I misunderstood FF's
behavior.  I updated the CL description and included the pointer to the spec.

We're considering to send a Intent-to-Ship to blink-dev for the behavioral
change, however, we may need more changes on Window regard to conformance to the
spec and security checks.  So we'll investigate what changes around Window (and
Workers) will be necessary before sending a Intent-to-Ship.

Let me suspend this CL for the time being.  Will resume once the plan gets
fixed.

[Yuki, 2015-08-25 06:56:03]

Finally I'm back to this CL.
Could you review this CL again?

[haraken, 2015-08-25 07:04:52]

LGTM

Add a link of the intent-to-ship to the CL description.

https://codereview.chromium.org/1121703002/diff/80001/LayoutTests/imported/we...
File
LayoutTests/imported/web-platform-tests/html/browsers/the-window-object/window-named-properties-expected.txt
(right):

https://codereview.chromium.org/1121703002/diff/80001/LayoutTests/imported/we...
LayoutTests/imported/web-platform-tests/html/browsers/the-window-object/window-named-properties-expected.txt:4:
FAIL constructor assert_equals: expected false but got true

Just to confirm: This FAIL line is expected, right?

[Yuki, 2015-08-25 08:03:52]

https://codereview.chromium.org/1121703002/diff/80001/LayoutTests/imported/we...
File
LayoutTests/imported/web-platform-tests/html/browsers/the-window-object/window-named-properties-expected.txt
(right):

https://codereview.chromium.org/1121703002/diff/80001/LayoutTests/imported/we...
LayoutTests/imported/web-platform-tests/html/browsers/the-window-object/window-named-properties-expected.txt:4:
FAIL constructor assert_equals: expected false but got true
On 2015/08/25 07:04:52, haraken wrote:
> 
> Just to confirm: This FAIL line is expected, right?

Yes, it's caused because we don't have WindowProperties object yet.

[Yuki, 2015-08-25 08:04:24]

The CQ bit was checked by yukishiino@chromium.org

[commit-bot: I haz the power, 2015-08-25 08:04:42]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1121703002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1121703002/80001

[commit-bot: I haz the power, 2015-08-25 08:58:33]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-08-25 08:58:34]

Try jobs failed on following builders:
  mac_chromium_rel_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[Yuki, 2015-08-25 09:11:06]

The CQ bit was checked by yukishiino@chromium.org

[commit-bot: I haz the power, 2015-08-25 09:11:20]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1121703002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1121703002/80001

[commit-bot: I haz the power, 2015-08-25 10:29:14]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-08-25 10:29:15]

Try jobs failed on following builders:
  mac_chromium_rel_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[Yuki, 2015-08-26 07:18:14]

The CQ bit was checked by yukishiino@chromium.org

[Yuki, 2015-08-26 07:18:14]

The patchset sent to the CQ was uploaded after l-g-t-m from haraken@chromium.org
Link to the patchset: https://codereview.chromium.org/1121703002/#ps100001
(title: "Synced.")

[commit-bot: I haz the power, 2015-08-26 07:18:17]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1121703002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1121703002/100001

[commit-bot: I haz the power, 2015-08-26 08:29:20]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-08-26 08:29:21]

Try jobs failed on following builders:
  win_chromium_rel_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[Yuki, 2015-08-26 12:32:13]

The CQ bit was checked by yukishiino@chromium.org

[commit-bot: I haz the power, 2015-08-26 12:32:24]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1121703002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1121703002/100001

[commit-bot: I haz the power, 2015-08-26 12:37:46]

Committed patchset #6 (id:100001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=201225