Add support for type checking of floating point attributes

Add support for type checking of floating point attributes

Just do the same thing for value passed to setter(if (!std::isfinite) ...)
as for method arguments.

Add examples in SourceBuffer for checking.

Followup to:
Add support for type checking of floating point arguments as [TypeChecking=Unrestricted]
https://codereview.chromium.org/259773008/

This does *not* add TypeChecking throughout SourceBuffer, only enough for testing; Fredrik is doing full type checking in:
Apply TypeChecking to MediaSource IDLs
https://codereview.chromium.org/266453005/

R=haraken
BUG=354298
TEST=http/tests/media/media-source/mediasource-timestamp-offset.html
TEST=http/tests/media/media-source/mediasource-appendwindow.html

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=173064

[Nils Barth (inactive), 2014-05-01 02:51:42]

Attributes too!
This finishes support for |unrestricted|;
now we just need it to be used everywhere!

[haraken, 2014-05-01 03:25:57]

https://codereview.chromium.org/265753003/diff/1/Source/modules/mediasource/S...
File Source/modules/mediasource/SourceBuffer.idl (right):

https://codereview.chromium.org/265753003/diff/1/Source/modules/mediasource/S...
Source/modules/mediasource/SourceBuffer.idl:60: [RaisesException=Setter,
TypeChecking=Unrestricted] attribute unrestricted double appendWindowEnd;

I'm a bit confused. What's the difference between:

- double x;
- unrestricted double x;
- [TypeChecking=Unrestricted] double x;
- [TypeChecking=Unrestricted] unrestricted double x;

?

[Nils Barth (inactive), 2014-05-01 03:34:25]

https://codereview.chromium.org/265753003/diff/1/Source/modules/mediasource/S...
File Source/modules/mediasource/SourceBuffer.idl (right):

https://codereview.chromium.org/265753003/diff/1/Source/modules/mediasource/S...
Source/modules/mediasource/SourceBuffer.idl:60: [RaisesException=Setter,
TypeChecking=Unrestricted] attribute unrestricted double appendWindowEnd;
On 2014/05/01 03:25:58, haraken wrote:
> What's the difference between:
> 
> - double x;
> - unrestricted double x;
> - [TypeChecking=Unrestricted] double x;
> - [TypeChecking=Unrestricted] unrestricted double x;

Behavior-wise, currently only:
[TypeChecking=Unrestricted] double x;
...has type-checking for non-finite values.
(For compatibility.)

However, [TypeChecking=Unrestricted] should go on the *interface*, in which case
attribute double x;
attribute unrestricted double x;
...*do* differ (attribute double x; is type-checked), and we have to get them
right.

The:
[TypeChecking=Unrestricted] attribute unrestricted double x;
...in this CL asserts that we've checked this attribute, so it'll be easy to
move the [TypeChecking=Unrestricted] to the interface (once methods are also
checked).

I'm just adding test cases here; Fredrik is doing full type-checking in:
Apply TypeChecking to MediaSource IDLs
https://codereview.chromium.org/266453005/

Make sense?

[haraken, 2014-05-01 03:38:23]

On 2014/05/01 03:34:25, Nils Barth wrote:
>
https://codereview.chromium.org/265753003/diff/1/Source/modules/mediasource/S...
> File Source/modules/mediasource/SourceBuffer.idl (right):
> 
>
https://codereview.chromium.org/265753003/diff/1/Source/modules/mediasource/S...
> Source/modules/mediasource/SourceBuffer.idl:60: [RaisesException=Setter,
> TypeChecking=Unrestricted] attribute unrestricted double appendWindowEnd;
> On 2014/05/01 03:25:58, haraken wrote:
> > What's the difference between:
> > 
> > - double x;
> > - unrestricted double x;
> > - [TypeChecking=Unrestricted] double x;
> > - [TypeChecking=Unrestricted] unrestricted double x;
> 
> Behavior-wise, currently only:
> [TypeChecking=Unrestricted] double x;
> ...has type-checking for non-finite values.
> (For compatibility.)
> 
> However, [TypeChecking=Unrestricted] should go on the *interface*, in which
case
> attribute double x;
> attribute unrestricted double x;
> ...*do* differ (attribute double x; is type-checked), and we have to get them
> right.
> 
> The:
> [TypeChecking=Unrestricted] attribute unrestricted double x;
> ...in this CL asserts that we've checked this attribute, so it'll be easy to
> move the [TypeChecking=Unrestricted] to the interface (once methods are also
> checked).
> 
> I'm just adding test cases here; Fredrik is doing full type-checking in:
> Apply TypeChecking to MediaSource IDLs
> https://codereview.chromium.org/266453005/
> 
> Make sense?

Makes sense, though I'd drop [TypeChecking=Unrestricted] from unrestricted
double attributes. It's confusing to have an IDL attribute that is no-op.

LGTM.

[Nils Barth (inactive), 2014-05-01 03:43:07]

On 2014/05/01 03:38:23, haraken wrote:
> Makes sense, though I'd drop [TypeChecking=Unrestricted] from unrestricted
> double attributes. It's confusing to have an IDL attribute that is no-op.

Got it, removed nop; will go away anyway with Fredrik's CL.
(Avoiding nops was why I asked people to not slap 'unrestricted' on everything
until we had support.)

[Nils Barth (inactive), 2014-05-01 03:43:12]

The CQ bit was checked by nbarth@chromium.org

[commit-bot: I haz the power, 2014-05-01 03:43:43]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nbarth@chromium.org/265753003/20001

[commit-bot: I haz the power, 2014-05-01 04:13:31]

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

[commit-bot: I haz the power, 2014-05-01 04:13:32]

Try jobs failed on following builders:
  mac_blink_rel on tryserver.blink

[Nils Barth (inactive), 2014-05-01 04:31:56]

The CQ bit was checked by nbarth@chromium.org

[commit-bot: I haz the power, 2014-05-01 04:32:09]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nbarth@chromium.org/265753003/20001

[commit-bot: I haz the power, 2014-05-01 05:13:09]

Change committed as 173064