Restore the HTMLVideoElement-specific prefixed fullscreen API behind a flag

Restore the HTMLVideoElement-specific prefixed fullscreen API behind a flag

The removal has caused more trouble than expected on Android, because
usage is higher there and the UseCounter data from Windows was used in
the "Intent to Remove" thread.

Restore the API behind a flag so that it can more easily be turned on
and off. It's off by default, but likely it will have to be enabled on
at least the upcoming M35 branch.

BUG=346236
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=170303

[philipj_slow, 2014-03-27 09:59:30]

PTAL. The runtime flag was suggested by abarth in
https://code.google.com/p/chromium/issues/detail?id=356236#c8

The first PS is a plain revert with trivial conflicts resolved. The other
changes are separated out for easier review.

Note that "no userGestureRequiredForFullscreen" isn't just me being lazy, all
that code did was throw an exception. Even if userGestureRequiredForFullscreen()
returned false the user gesture check in
FullscreenElementStack::requestFullScreenForElement doesn't care about that, it
looks at the mediaFullscreenRequiresUserGesture() setting which remains
unchanged. It seems better to just drop the useless exception than to revive the
HTMLMediaElement state needed to support it.

I didn't bother testing webkitEnterFullScreen and webkitExitFullScreen (capital
'S') because they're just aliases and unlikely to break independently of what is
tested. If you want I can duplicate the test and search/replace.

[eseidel, 2014-03-27 16:54:17]

I like the idea of using a guard.  Difficult for me to tell if this is the
correct implementation or not. Do we have any testing we can also add back?

rslgtm.

https://codereview.chromium.org/213193010/diff/60001/Source/core/frame/UseCou...
File Source/core/frame/UseCounter.cpp (right):

https://codereview.chromium.org/213193010/diff/60001/Source/core/frame/UseCou...
Source/core/frame/UseCounter.cpp:665: return
"'HTMLVideoElement.webkitEnterFullscreen()' is deprecated. Please use
'Element.requestFullscreen()' and 'Element.webkitRequestFullscreen()' instead.";
Why even mention the webkit versions?

https://codereview.chromium.org/213193010/diff/60001/Source/core/html/HTMLMed...
File Source/core/html/HTMLMediaElement.cpp (right):

https://codereview.chromium.org/213193010/diff/60001/Source/core/html/HTMLMed...
Source/core/html/HTMLMediaElement.cpp:3197:
FullscreenElementStack::from(document()).webkitCancelFullScreen();
Do you mean to call the webkit version?

https://codereview.chromium.org/213193010/diff/60001/Source/core/html/HTMLVid...
File Source/core/html/HTMLVideoElement.h (right):

https://codereview.chromium.org/213193010/diff/60001/Source/core/html/HTMLVid...
Source/core/html/HTMLVideoElement.h:49: void
webkitEnterFullscreen(ExceptionState&);
Are these different from their non-prefix'd counterparts such that they need a
webkit name?

[philipj_slow, 2014-03-27 18:58:31]

Thanks for the feedback eseidel! I was also surprised when I realized that we
haven't unprefixed the fullscreen API, last I heard it was blocked on some
details that weren't in line with the spec. FWIW, no other engine (except
Presto, R.I.P.) has unprefixed either.

https://codereview.chromium.org/213193010/diff/60001/Source/core/frame/UseCou...
File Source/core/frame/UseCounter.cpp (right):

https://codereview.chromium.org/213193010/diff/60001/Source/core/frame/UseCou...
Source/core/frame/UseCounter.cpp:665: return
"'HTMLVideoElement.webkitEnterFullscreen()' is deprecated. Please use
'Element.requestFullscreen()' and 'Element.webkitRequestFullscreen()' instead.";
On 2014/03/27 16:54:18, eseidel wrote:
> Why even mention the webkit versions?

We don't support the unprefixed versions, so telling people to only use the
unprefixed version wouldn't be very nice.

https://codereview.chromium.org/213193010/diff/60001/Source/core/html/HTMLMed...
File Source/core/html/HTMLMediaElement.cpp (right):

https://codereview.chromium.org/213193010/diff/60001/Source/core/html/HTMLMed...
Source/core/html/HTMLMediaElement.cpp:3197:
FullscreenElementStack::from(document()).webkitCancelFullScreen();
On 2014/03/27 16:54:18, eseidel wrote:
> Do you mean to call the webkit version?

It's the only one there is :)

https://codereview.chromium.org/213193010/diff/60001/Source/core/html/HTMLVid...
File Source/core/html/HTMLVideoElement.h (right):

https://codereview.chromium.org/213193010/diff/60001/Source/core/html/HTMLVid...
Source/core/html/HTMLVideoElement.h:49: void
webkitEnterFullscreen(ExceptionState&);
On 2014/03/27 16:54:18, eseidel wrote:
> Are these different from their non-prefix'd counterparts such that they need a
> webkit name?

These don't have unprefixed counterparts and never will, since what ended up in
http://fullscreen.spec.whatwg.org/ is Element.requestFullscreen() and
Document.exitFullscreen().

[philipj_slow, 2014-03-27 19:02:46]

On 2014/03/27 16:54:17, eseidel wrote:
> I like the idea of using a guard.  Difficult for me to tell if this is the
> correct implementation or not. Do we have any testing we can also add back?

You can see the tests that were changed in
https://codereview.chromium.org/178003008 to get an idea of the original
coverage. It was pretty scattered, so I opted to put it all in a single test
instead. Overall coverage is better now, but I see now that maybe
full-screen-iframe-legacy.html would be a good idea to revive as well.

[philipj_slow, 2014-03-28 08:37:53]

The CQ bit was checked by philipj@opera.com

[commit-bot: I haz the power, 2014-03-28 08:37:53]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/philipj@opera.com/213193010/80001

[commit-bot: I haz the power, 2014-03-28 09:38:39]

Change committed as 170303