Sync the Touch-related interfaces with the two specs

Sync the Touch-related interfaces with the two specs

Unfortunately there does not appear to be a single well-maintained spec
that deals with all of this, but these documents from the Touch Events
Community Group seem to be the most recently updated:
http://rawgit.com/w3c/touch-events/v1-errata/touchevents.html
http://rawgit.com/w3c/touch-events/master/touchevents.html

There are no changes to the generated code except for changes in order.

BUG=460722
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=191441

[philipj_slow, 2015-03-06 08:16:36]

philipj@opera.com changed reviewers:
+ rbyers@chromium.org

[philipj_slow, 2015-03-06 08:16:36]

PTAL. As you can tell I couldn't find a good spec when I added a link for
Document.createTouch, any chance of there eventually being a single spec for
implementors that ignores all history, kind of like
https://dom.spec.whatwg.org/?

[Rick Byers, 2015-03-06 13:56:55]

Thanks LGTM.

Yeah we don't currently have a WG working on a new official version of
TouchEvents (hence the errata / extensions note split), but I expect that will
change.

I filed crbug.com/464708 for the minor FIXMEs you added, and
https://github.com/w3c/touch-events/issues/10 to deprecate initTouchEvent in
favor of an event constructor (there's some history there explaining the state
we're in if you're interested).

https://codereview.chromium.org/982273002/diff/1/Source/core/dom/Document.idl
File Source/core/dom/Document.idl (right):

https://codereview.chromium.org/982273002/diff/1/Source/core/dom/Document.idl...
Source/core/dom/Document.idl:170: // FIXME: The webkit-prefixed arguments are
not in the spec.
If you want, you can remove the webkit prefixes on the argument names here since
they're in the TE extensions note and we're shipping them unprefixed (although
we still aren't shipping Touch.rotationAngle unprefixed, there's no reason not
to anymore).

[philipj_slow, 2015-03-06 15:33:59]

Thanks for the follow-up to fix these, I'll leave this as-is to not introduce
any behavior changes.

https://codereview.chromium.org/982273002/diff/1/Source/core/dom/Document.idl
File Source/core/dom/Document.idl (right):

https://codereview.chromium.org/982273002/diff/1/Source/core/dom/Document.idl...
Source/core/dom/Document.idl:170: // FIXME: The webkit-prefixed arguments are
not in the spec.
On 2015/03/06 13:56:55, Rick Byers wrote:
> If you want, you can remove the webkit prefixes on the argument names here
since
> they're in the TE extensions note and we're shipping them unprefixed (although
> we still aren't shipping Touch.rotationAngle unprefixed, there's no reason not
> to anymore).

Actually the extension spec doesn't mention these new createTouch arguments,
maybe it should?

[philipj_slow, 2015-03-06 15:34:03]

The CQ bit was checked by philipj@opera.com

[commit-bot: I haz the power, 2015-03-06 15:34:48]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/982273002/1

[commit-bot: I haz the power, 2015-03-06 15:37:49]

Committed patchset #1 (id:1) as
https://src.chromium.org/viewvc/blink?view=rev&revision=191441

[Rick Byers, 2015-03-06 15:53:14]

On 2015/03/06 15:33:59, philipj_UTC7 wrote:
> Thanks for the follow-up to fix these, I'll leave this as-is to not introduce
> any behavior changes.

Ok, updated crbug.com/460099 to include changing the arg names.  In what
scenarios might that be a "behavior change"?  Just when called from a language
with named parameters, right?  I think we really care only about JavaScript
bindings so this shouldn't be an issue (plus this API is quite obscure and
rarely used).

> 
> https://codereview.chromium.org/982273002/diff/1/Source/core/dom/Document.idl
> File Source/core/dom/Document.idl (right):
> 
>
https://codereview.chromium.org/982273002/diff/1/Source/core/dom/Document.idl...
> Source/core/dom/Document.idl:170: // FIXME: The webkit-prefixed arguments are
> not in the spec.
> On 2015/03/06 13:56:55, Rick Byers wrote:
> > If you want, you can remove the webkit prefixes on the argument names here
> since
> > they're in the TE extensions note and we're shipping them unprefixed
(although
> > we still aren't shipping Touch.rotationAngle unprefixed, there's no reason
not
> > to anymore).
> 
> Actually the extension spec doesn't mention these new createTouch arguments,
> maybe it should?

Yeah - filed https://github.com/w3c/touch-events/issues/12

[philipj_slow, 2015-03-06 17:15:17]

On 2015/03/06 15:53:14, Rick Byers wrote:
> On 2015/03/06 15:33:59, philipj_UTC7 wrote:
> > Thanks for the follow-up to fix these, I'll leave this as-is to not
introduce
> > any behavior changes.
> 
> Ok, updated crbug.com/460099 to include changing the arg names.  In what
> scenarios might that be a "behavior change"?  Just when called from a language
> with named parameters, right?  I think we really care only about JavaScript
> bindings so this shouldn't be an issue (plus this API is quite obscure and
> rarely used).

Yes I'm just confused, changing the argument names wouldn't be an observable
change.