Map all ContextDestroyed custom element exceptions to InvalidStateError.

Map all ContextDestroyed custom element exceptions to InvalidStateError.

For the exceptions that document.registerElement() may throw due to the
context having gone away, map them all to InvalidStateError.

Specifically, map ContextDestroyedRegisteringDefinition to InvalidStateError
rather than NotSupportedError.

R=haraken@chromium.org,ager@chromium.org,dominicc@chromium.org
BUG=

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=177907

[sof, 2014-07-10 08:14:32]

Please take a look.

I think it makes sense to be consistent about what DOM exception to report for
these three exceptions. Not being so raises an issue with Oilpan as there the
timing about when Nodes and Documents being finalized will be different.

Actually, the spec

  http://w3c.github.io/webcomponents/spec/custom/#dfn-document-registerElement

says that NotSupportedError should be used here. If that's deemed preferable, we
could switch over the InvalidStateError uses here instead. I didn't do so as I
think InvalidStateError carries slightly more information.

[haraken, 2014-07-10 08:22:22]

LGTM, but dominicc@ should confirm.

[Mads Ager (chromium), 2014-07-10 08:27:52]

LGTM2, thanks for taking care of these and sorry for leaving the XYZ in the
status file instead of actually filing the bug. :-)

[dominicc (has gone to gerrit), 2014-07-11 06:00:23]

On 2014/07/10 08:27:52, Mads Ager (chromium) wrote:
> LGTM2, thanks for taking care of these and sorry for leaving the XYZ in the
> status file instead of actually filing the bug. :-)

LGTM.

I believe that the spec is not prescriptive because registering around
destruction time is a bit of a corner case. InvalidStateError seems reasonable.

[sof, 2014-07-11 06:05:24]

On 2014/07/11 06:00:23, dominicc wrote:
> On 2014/07/10 08:27:52, Mads Ager (chromium) wrote:
> > LGTM2, thanks for taking care of these and sorry for leaving the XYZ in the
> > status file instead of actually filing the bug. :-)
> 
> LGTM.
> 
> I believe that the spec is not prescriptive because registering around
> destruction time is a bit of a corner case. InvalidStateError seems
reasonable.

Good, thanks for the clarification.

[sof, 2014-07-11 06:05:29]

The CQ bit was checked by sigbjornf@opera.com

[commit-bot: I haz the power, 2014-07-11 06:06:18]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/sigbjornf@opera.com/375343003/1

[commit-bot: I haz the power, 2014-07-11 06:59:37]

Change committed as 177907