Add [TypeChecking=Unrestricted] to MIDIMessageEventInit

Add [TypeChecking=Unrestricted] to MIDIMessageEventInit

This will cause a TypeError to be thrown if 'receivedTime' is non-finite,
as is required (indirectly, by referencing Web IDL) by the specification.

BUG=354298
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=191007

[Jens Widell, 2015-02-27 11:07:42]

jl@opera.com changed reviewers:
+ toyoshim@chromium.org

[Jens Widell, 2015-02-27 11:07:43]

PTAL

[Takashi Toyoshima, 2015-02-27 11:55:24]

toyoshim@chromium.org changed reviewers:
+ kouhei@chromium.org

[Takashi Toyoshima, 2015-02-27 11:58:21]

+kouhei who may know better on Web IDL than me.

I don't understand the background of TypeChecking=Unrestricted, but I feel
MIDIEssageEvent should not allow to use infinite double for receivedTime. In
that case, what's the right action? Should we update something on Web MIDI API's
spec? Or just not adding this annotation?

[Jens Widell, 2015-02-27 12:01:43]

On 2015/02/27 11:58:21, Takashi Toyoshima (chromium) wrote:
> +kouhei who may know better on Web IDL than me.
> 
> I don't understand the background of TypeChecking=Unrestricted, but I feel
> MIDIEssageEvent should not allow to use infinite double for receivedTime. In
> that case, what's the right action? Should we update something on Web MIDI
API's
> spec? Or just not adding this annotation?

TypeChecking=Unrestricted is oddly named; the effect of it is that we enforce
the rule in Web IDL that "double" should reject non-finite values. That is, it
says "please generate code that follow the specification".

We're getting close to being able to drop it, so that "double" (and "float")
always work the way they're supposed to according to Web IDL.

The type "unrestricted double", which does not reject non-finite values, already
works as expected in our implementation, and is not affected by
[TypeChecking=Unrestricted].

[kouhei (in TOK), 2015-02-27 14:47:41]

lgtm

[Jens Widell, 2015-02-27 14:48:48]

The CQ bit was checked by jl@opera.com

[commit-bot: I haz the power, 2015-02-27 14:49:19]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/962163002/1

[commit-bot: I haz the power, 2015-02-27 16:00:17]

Committed patchset #1 (id:1) as
https://src.chromium.org/viewvc/blink?view=rev&revision=191007