Start an empty interface OffScreenCanvas

Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is empty.

Also, a layout test is added to verify that the constructor
can be successfully called from JS, and verify the object's
width&height, together with verifying the constructor's behavior when
the argument is invalid.

BUG=563819
Committed: https://crrev.com/0f3eb18441ba1c98a8bb7ef8d482b22f682c6968
Cr-Commit-Position: refs/heads/master@{#365095}

[xidachen, 2015-11-30 21:41:03]

xidachen@chromium.org changed reviewers:
+ junov@chromium.org

[xidachen, 2015-11-30 21:41:04]

This is just a start, please take a look because there was some roadblock.

https://codereview.chromium.org/1488763002/diff/40001/third_party/WebKit/Sour...
File
third_party/WebKit/Source/modules/canvas2d/CanvasRenderingContext2DOffScreenCanvas.idl
(right):

https://codereview.chromium.org/1488763002/diff/40001/third_party/WebKit/Sour...
third_party/WebKit/Source/modules/canvas2d/CanvasRenderingContext2DOffScreenCanvas.idl:10:
readonly attribute (HTMLCanvasElement or OffScreenCanvas) canvas;
This one doesn't work at this moment because it conflict with the attribute
canvas in CRC2D.idl. We need to figure out a way to solve it.

[xidachen, 2015-12-01 15:02:28]

Description was changed from

==========
Start an emmpty interface OffScreenCanvas

This CL starts an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface is empty.
==========

to

==========
Start an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface is empty.

BUG=563819
==========

[xidachen, 2015-12-03 02:40:06]

Description was changed from

==========
Start an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface is empty.

BUG=563819
==========

to

==========
Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is
empty.

BUG=563819
==========

[xidachen, 2015-12-03 04:30:24]

Description was changed from

==========
Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is
empty.

BUG=563819
==========

to

==========
Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is
empty.

Also, a layout test is added to verify that the constructor
can be successfully called from JS.

BUG=563819
==========

[xidachen, 2015-12-03 04:31:01]

Hi Justin,

The latest patch contains empty interface&implementation, PTAL.

[Justin Novosad, 2015-12-03 04:42:58]

https://codereview.chromium.org/1488763002/diff/100001/third_party/WebKit/Lay...
File third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-constructor.html
(right):

https://codereview.chromium.org/1488763002/diff/100001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-constructor.html:4:
jsTestIsAsync = true;
No need for this test to be async

https://codereview.chromium.org/1488763002/diff/100001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.idl (right):

https://codereview.chromium.org/1488763002/diff/100001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.idl:5: //
https://wiki.whatwg.org/wiki/OffscreenCanvas
I am not sure we should link to this. It is not an actual spec, and will soon be
obsolete.

https://codereview.chromium.org/1488763002/diff/100001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.idl:11: ] interface
OffScreenCanvas {
initial interface should at least expose width and height attributes, then the
constructor test can verify that the size is equal to what was passed to the
constructor.

[xidachen, 2015-12-03 04:53:30]

Description was changed from

==========
Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is
empty.

Also, a layout test is added to verify that the constructor
can be successfully called from JS.

BUG=563819
==========

to

==========
Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is
empty.

Also, a layout test is added to verify that the constructor
can be successfully called from JS, and verify the object's
width&height.

BUG=563819
==========

[xidachen, 2015-12-03 04:54:47]

https://codereview.chromium.org/1488763002/diff/100001/third_party/WebKit/Lay...
File third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-constructor.html
(right):

https://codereview.chromium.org/1488763002/diff/100001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-constructor.html:4:
jsTestIsAsync = true;
On 2015/12/03 04:42:58, Justin Novosad wrote:
> No need for this test to be async

Done.

https://codereview.chromium.org/1488763002/diff/100001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.idl (right):

https://codereview.chromium.org/1488763002/diff/100001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.idl:5: //
https://wiki.whatwg.org/wiki/OffscreenCanvas
On 2015/12/03 04:42:58, Justin Novosad wrote:
> I am not sure we should link to this. It is not an actual spec, and will soon
be
> obsolete.

Done.

https://codereview.chromium.org/1488763002/diff/100001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.idl:11: ] interface
OffScreenCanvas {
On 2015/12/03 04:42:58, Justin Novosad wrote:
> initial interface should at least expose width and height attributes, then the
> constructor test can verify that the size is equal to what was passed to the
> constructor.

Done.

[Justin Novosad, 2015-12-03 05:07:59]

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Lay...
File third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-constructor.html
(right):

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-constructor.html:15:
finishJSTest();
get rid of this line.

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp (right):

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp:10:
OffScreenCanvas* OffScreenCanvas::create(unsigned long width, unsigned long
height)
no 'long'

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp:12: return new
OffScreenCanvas(IntSize(width, height));
another implicit cast with a risk of overflow.

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.h (right):

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.h:20: static
OffScreenCanvas* create(unsigned long width, unsigned long height);
'long' keyword is superfluous

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.h:23: unsigned long
width() const { return m_size.width(); }
same here

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.h:24: unsigned long
height() const { return m_size.height(); }
same here.

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.h:26: void
setWidth(unsigned long width) { m_size.setWidth(width); }
No 'long'. Implicit cast of unsigned to int may overflow, need to protect
against that, silently reject the new value if it can't be handled because it is
too large. this should also be tested.

[xidachen, 2015-12-03 14:16:17]

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Lay...
File third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-constructor.html
(right):

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-constructor.html:15:
finishJSTest();
On 2015/12/03 05:07:59, Justin Novosad wrote:
> get rid of this line.

Done.

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp (right):

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp:10:
OffScreenCanvas* OffScreenCanvas::create(unsigned long width, unsigned long
height)
On 2015/12/03 05:07:59, Justin Novosad wrote:
> no 'long'

Done.

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp:12: return new
OffScreenCanvas(IntSize(width, height));
On 2015/12/03 05:07:59, Justin Novosad wrote:
> another implicit cast with a risk of overflow.

I am not sure what's the best solution here. But right now I just set
width/height to be INT_MAX if it is larger than that.

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.h (right):

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.h:20: static
OffScreenCanvas* create(unsigned long width, unsigned long height);
On 2015/12/03 05:07:59, Justin Novosad wrote:
> 'long' keyword is superfluous

Done.

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.h:23: unsigned long
width() const { return m_size.width(); }
On 2015/12/03 05:07:59, Justin Novosad wrote:
> same here

Done.

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.h:24: unsigned long
height() const { return m_size.height(); }
On 2015/12/03 05:07:59, Justin Novosad wrote:
> same here.

Done.

https://codereview.chromium.org/1488763002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.h:26: void
setWidth(unsigned long width) { m_size.setWidth(width); }
On 2015/12/03 05:07:59, Justin Novosad wrote:
> No 'long'. Implicit cast of unsigned to int may overflow, need to protect
> against that, silently reject the new value if it can't be handled because it
is
> too large. this should also be tested.

Done.

[Justin Novosad, 2015-12-03 17:05:55]

https://codereview.chromium.org/1488763002/diff/140001/third_party/WebKit/Lay...
File
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html
(right):

https://codereview.chromium.org/1488763002/diff/140001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:9:
var setWidth = Math.pow(2, 31);
Explain that the overflow is an implementation-specific thing because blink uses
signed int internally.

https://codereview.chromium.org/1488763002/diff/140001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:11:
var canvas1 = new OffScreenCanvas(setWidth, setHeight);
You should also validate setting the size through the attributes, and you should
verify what happens (TypeError exceptions thrown) when passing a negative
number, null, an object.

https://codereview.chromium.org/1488763002/diff/140001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp (right):

https://codereview.chromium.org/1488763002/diff/140001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp:21: if (width >
INT_MAX)
use clampTo<int>, defined in wtf/MathExtras.h

[xidachen, 2015-12-03 18:41:41]

https://codereview.chromium.org/1488763002/diff/140001/third_party/WebKit/Lay...
File
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html
(right):

https://codereview.chromium.org/1488763002/diff/140001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:9:
var setWidth = Math.pow(2, 31);
On 2015/12/03 17:05:55, Justin Novosad wrote:
> Explain that the overflow is an implementation-specific thing because blink
uses
> signed int internally.

Done.

https://codereview.chromium.org/1488763002/diff/140001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:11:
var canvas1 = new OffScreenCanvas(setWidth, setHeight);
On 2015/12/03 17:05:54, Justin Novosad wrote:
> You should also validate setting the size through the attributes, and you
should
> verify what happens (TypeError exceptions thrown) when passing a negative
> number, null, an object.

The case of setting the size through the attributes is added to the
constructor.html, the other cases(invalid args) are added to this test.

https://codereview.chromium.org/1488763002/diff/140001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp (right):

https://codereview.chromium.org/1488763002/diff/140001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp:21: if (width >
INT_MAX)
On 2015/12/03 17:05:55, Justin Novosad wrote:
> use clampTo<int>, defined in wtf/MathExtras.h

Done.

[Justin Novosad, 2015-12-03 19:06:11]

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Lay...
File
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html
(right):

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:7:
// responses to the arguments that are larger than INT_MAX which would cause
responses -> responds

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:8:
// overflow.
Add: the current implementation is expected to clamp.

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:17:
shouldBe("canvas2.width", "setWidth-1");
This looks like a bug to me. The WebIDL spec says this should throw a TypeError:
http://www.w3.org/TR/WebIDL/#es-unsigned-long

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:28:
</script>
still missing test coverage for assigning invalid values to canvas.width,
canvas.height. (attribute setters)

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp (right):

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp:14: if (width >
INT_MAX)
This condition is not necessary. it is implied in clampTo. Same thing for the
code below.  INT_MAX should go away from this file.

[xidachen, 2015-12-03 20:42:22]

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Lay...
File
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html
(right):

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:7:
// responses to the arguments that are larger than INT_MAX which would cause
On 2015/12/03 19:06:11, Justin Novosad wrote:
> responses -> responds

Done.

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:8:
// overflow.
On 2015/12/03 19:06:11, Justin Novosad wrote:
> Add: the current implementation is expected to clamp.

Done.

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:17:
shouldBe("canvas2.width", "setWidth-1");
On 2015/12/03 19:06:11, Justin Novosad wrote:
> This looks like a bug to me. The WebIDL spec says this should throw a
TypeError:
> http://www.w3.org/TR/WebIDL/#es-unsigned-long

The WebIDL: http://www.w3.org/TR/WebIDL/#es-unsigned-long says throw TypeError
only when [EnforceRange] is specified. If EnforceRange is not used, then it will
do width = ToUint32(width), which is essentially modulo 2^32.

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:28:
</script>
On 2015/12/03 19:06:11, Justin Novosad wrote:
> still missing test coverage for assigning invalid values to canvas.width,
> canvas.height. (attribute setters)

Done.

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp (right):

https://codereview.chromium.org/1488763002/diff/160001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/html/canvas/OffScreenCanvas.cpp:14: if (width >
INT_MAX)
On 2015/12/03 19:06:11, Justin Novosad wrote:
> This condition is not necessary. it is implied in clampTo. Same thing for the
> code below.  INT_MAX should go away from this file.

Done.

[Justin Novosad, 2015-12-04 04:37:08]

On 2015/12/03 20:42:22, xidachen wrote:

> The WebIDL: http://www.w3.org/TR/WebIDL/#es-unsigned-long says throw TypeError
> only when [EnforceRange] is specified. If EnforceRange is not used, then it
will
> do width = ToUint32(width), which is essentially modulo 2^32.
> 

Ugh, I hate this, but you're correct. The clamp extended attribute would make
sense here, but we don't use it in other similar places in the spec. So be it.

[Justin Novosad, 2015-12-04 04:37:41]

lgtm with minor correction

https://codereview.chromium.org/1488763002/diff/180001/third_party/WebKit/Lay...
File
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html
(right):

https://codereview.chromium.org/1488763002/diff/180001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:42:
canvas3.width = -1;
Use the helper function "shouldThrow" instead of all these try/catch pairs

[xidachen, 2015-12-04 13:16:42]

On 2015/12/04 04:37:41, Justin Novosad wrote:
> lgtm with minor correction
> 
>
https://codereview.chromium.org/1488763002/diff/180001/third_party/WebKit/Lay...
> File
> third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html
> (right):
> 
>
https://codereview.chromium.org/1488763002/diff/180001/third_party/WebKit/Lay...
>
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:42:
> canvas3.width = -1;
> Use the helper function "shouldThrow" instead of all these try/catch pairs

I believe EnforceRange is the right thing to do here. In the future, we should
prevent anything like this: static create(IntSize&).

On the other hand, I actually tried to use "shouldThrow" when I was working on
the layout tests. I found that the "shouldThrow" is used to catch exceptions
that are thrown in C++ code. In particular, those methods that have
[RaiseException] in its definition. For the TypeError here which is thrown by
javascript, shouldThrow doesn't work, and gave me "CONSOLE ERROR: line 22:
Uncaught TypeError: Failed to construct 'OffScreenCanvas': Value is outside the
'unsigned long' value range.", and the entire test terminated from there.

[xidachen, 2015-12-04 13:18:49]

The CQ bit was checked by xidachen@chromium.org

[commit-bot: I haz the power, 2015-12-04 13:19:31]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1488763002/180001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1488763002/180001

[xidachen, 2015-12-04 14:11:35]

The CQ bit was unchecked by xidachen@chromium.org

[Justin Novosad, 2015-12-04 14:16:50]

On 2015/12/04 13:16:42, xidachen wrote:
> On 2015/12/04 04:37:41, Justin Novosad wrote:
> > lgtm with minor correction
> > 
> >
>
https://codereview.chromium.org/1488763002/diff/180001/third_party/WebKit/Lay...
> > File
> > third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html
> > (right):
> > 
> >
>
https://codereview.chromium.org/1488763002/diff/180001/third_party/WebKit/Lay...
> >
>
third_party/WebKit/LayoutTests/fast/canvas/OffScreenCanvas-invalid-args.html:42:
> > canvas3.width = -1;
> > Use the helper function "shouldThrow" instead of all these try/catch pairs
> 
> I believe EnforceRange is the right thing to do here. In the future, we should
> prevent anything like this: static create(IntSize&).
> 
> On the other hand, I actually tried to use "shouldThrow" when I was working on
> the layout tests. I found that the "shouldThrow" is used to catch exceptions
> that are thrown in C++ code. In particular, those methods that have
> [RaiseException] in its definition. For the TypeError here which is thrown by
> javascript, shouldThrow doesn't work, and gave me "CONSOLE ERROR: line 22:
> Uncaught TypeError: Failed to construct 'OffScreenCanvas': Value is outside
the
> 'unsigned long' value range.", and the entire test terminated from there.

shouldThrow should work. Perhaps you were not using it correctly. Did you forget
to but the expression in "quotes"?

[xidachen, 2015-12-04 15:09:35]

xidachen@chromium.org changed reviewers:
+ chrishtr@chromium.org

[Justin Novosad, 2015-12-05 15:10:58]

lgtm for realz

[xidachen, 2015-12-14 13:25:13]

Description was changed from

==========
Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is
empty.

Also, a layout test is added to verify that the constructor
can be successfully called from JS, and verify the object's
width&height.

BUG=563819
==========

to

==========
Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is
empty.

Also, a layout test is added to verify that the constructor
can be successfully called from JS, and verify the object's
width&height, together with verifying the constructor's behavior when
the argument is invalid.

BUG=563819
==========

[xidachen, 2015-12-14 14:10:02]

Hi Chris,

This is the CL that gets the OffScreenCanvas API started. PTAL.

[chrishtr, 2015-12-14 17:40:59]

The CQ bit was checked by chrishtr@chromium.org

[chrishtr, 2015-12-14 17:40:59]

lgtm

[chrishtr, 2015-12-14 17:40:59]

The patchset sent to the CQ was uploaded after l-g-t-m from junov@chromium.org
Link to the patchset: https://codereview.chromium.org/1488763002/#ps260001
(title: "update global-interface-listing-expected.txt")

[commit-bot: I haz the power, 2015-12-14 17:41:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1488763002/260001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1488763002/260001

[commit-bot: I haz the power, 2015-12-14 19:35:38]

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

[commit-bot: I haz the power, 2015-12-14 19:35:38]

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_...)

[xidachen, 2015-12-14 19:36:28]

The CQ bit was checked by xidachen@chromium.org

[commit-bot: I haz the power, 2015-12-14 19:38:29]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1488763002/260001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1488763002/260001

[commit-bot: I haz the power, 2015-12-14 20:57:59]

Description was changed from

==========
Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is
empty.

Also, a layout test is added to verify that the constructor
can be successfully called from JS, and verify the object's
width&height, together with verifying the constructor's behavior when
the argument is invalid.

BUG=563819
==========

to

==========
Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is
empty.

Also, a layout test is added to verify that the constructor
can be successfully called from JS, and verify the object's
width&height, together with verifying the constructor's behavior when
the argument is invalid.

BUG=563819
==========

[commit-bot: I haz the power, 2015-12-14 20:58:00]

Committed patchset #14 (id:260001)

[commit-bot: I haz the power, 2015-12-14 20:59:26]

Description was changed from

==========
Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is
empty.

Also, a layout test is added to verify that the constructor
can be successfully called from JS, and verify the object's
width&height, together with verifying the constructor's behavior when
the argument is invalid.

BUG=563819
==========

to

==========
Create an empty interface OffScreenCanvas

This CL creates an interface OffScreenCanvas.idl according to the proposal:
https://wiki.whatwg.org/wiki/OffscreenCanvas

At this moment, the interface only has a constructor, and its implementation is
empty.

Also, a layout test is added to verify that the constructor
can be successfully called from JS, and verify the object's
width&height, together with verifying the constructor's behavior when
the argument is invalid.

BUG=563819

Committed: https://crrev.com/0f3eb18441ba1c98a8bb7ef8d482b22f682c6968
Cr-Commit-Position: refs/heads/master@{#365095}
==========

[commit-bot: I haz the power, 2015-12-14 20:59:27]

Patchset 14 (id:??) landed as
https://crrev.com/0f3eb18441ba1c98a8bb7ef8d482b22f682c6968
Cr-Commit-Position: refs/heads/master@{#365095}