Match CanvasGradient::addColorStop to the spec.

Match CanvasGradient::addColorStop to the spec.

The canvas spec [1] defines addColorStop without optional arguments. This patch
removes optional attribute from addColorStop interface.

[1] http://www.w3.org/TR/2dcontext2/

TEST=LayoutTests/canvas/philip/tests/2d.missingargs.html

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=152779

[dshwang, 2013-05-10 13:00:03]

This patch is the follow-up of https://codereview.chromium.org/14654002/

After this patch, Blink pass canvas/philip/tests/2d.missingargs.html

[dshwang, 2013-05-13 05:29:30]

Hi! could you review please?

[eseidel, 2013-05-13 06:45:13]

Does this match FF and IE?  Since now these are required arguments, if they
permit excluding these arguments then we may have to as well (to avoid throwing
javascxript exceptions and breaking sites).

[dshwang, 2013-05-13 13:49:24]

On 2013/05/13 06:45:13, Eric Seidel wrote:
> Does this match FF and IE?  Since now these are required arguments, if they
> permit excluding these arguments then we may have to as well (to avoid
throwing
> javascxript exceptions and breaking sites).

In conclusion, this change is compatible with other browsers: IE, FF, Opera.

In detail, when calling addColorStop with not enough arguments, browser must
emit TypeError exception.
1. Blink emits IndexSizeError/SyntaxError exceptions, currently.
2. IE behave correctly (emitting TypeError) . surprise.
3. FF emits [Exception... "Not enough arguments" nsresult: "0x80570001]
4. Opera emits WRONG_ARGUMENTS_ERR

This changes behavior slightly: emitting exception from
IndexSizeError/SyntaxError to TypeError. Emitting exception is not intended, but
piggybacked by chance.

[dshwang, 2013-05-17 13:54:20]

Hi! could you review please?

[Rik, 2013-05-17 16:19:07]

On 2013/05/17 13:54:20, dshwang wrote:
> Hi! could you review please?

There is a general trend to make the web facing APIs permissive. There is a
perception that it's better to keep an application going (with some rendering
issues) than to have it error out.
FWIW FF is also in the camp of making APIs not throw if possible.

[dshwang, 2013-05-17 17:05:55]

On 2013/05/17 16:19:07, Rik wrote:
> On 2013/05/17 13:54:20, dshwang wrote:
> > Hi! could you review please?
> 
> There is a general trend to make the web facing APIs permissive. There is a
> perception that it's better to keep an application going (with some rendering
> issues) than to have it error out.
> FWIW FF is also in the camp of making APIs not throw if possible.

Hi, Thx for review.

As I mentioned #msg5, I do not change it from not-throwing exception to throwing
exception. I DO change it from throwing wrong exception to throwing correct
exception.

Before/After this patch, Blink have been throwing exception. so IMO, it is very
hard to effect existing sites badly.

In detail, currently, Blink already throws IndexSizeError/SyntaxError exception,
but the spec said that Blink should throw TypeError exception. So I followed the
spec.

[dshwang, 2013-05-23 10:53:38]

ping

[dshwang, 2013-05-29 14:25:27]

Hi! could you review again plz?

[Stephen Chennney, 2013-06-18 13:26:11]

I can see Rik's point of view, but given the test exists already, and IE
behavior, and the fact that other browsers right now throw exceptions, I think
this should go in.

LGTM with the copyright nit.

https://codereview.chromium.org/14798006/diff/1/Source/core/html/canvas/Canva...
File Source/core/html/canvas/CanvasGradient.idl (right):

https://codereview.chromium.org/14798006/diff/1/Source/core/html/canvas/Canva...
Source/core/html/canvas/CanvasGradient.idl:2: * Copyright (C) 2006 Apple
Computer, Inc.  All rights reserved.
You should add another line here for this change.

[dshwang, 2013-06-19 14:39:03]

On 2013/06/18 13:26:11, Stephen Chenney wrote:
> LGTM with the copyright nit.

Thx for review as well as caring of license!

[commit-bot: I haz the power, 2013-06-19 14:39:15]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/dongseong.hwang@intel.com/14798006/13001

[commit-bot: I haz the power, 2013-06-19 14:48:44]

Step "update" is always a major failure.
Look at the try server FAQ for more details.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_layout...

[commit-bot: I haz the power, 2013-06-19 14:58:13]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/dongseong.hwang@intel.com/14798006/13001

[commit-bot: I haz the power, 2013-06-19 16:21:40]

Step "update" is always a major failure.
Look at the try server FAQ for more details.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=mac_layout...

[commit-bot: I haz the power, 2013-06-20 07:11:00]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/dongseong.hwang@intel.com/14798006/13001

[commit-bot: I haz the power, 2013-06-20 12:36:01]

Change committed as 152779