Resize MediaDocument playback to fit window size.

Simple CSS fix to prevent video, audio tags from outgrowing their containing windows. Added a new layout test to verify this behavior.

BUG=26848
TEST=Played video, audio, resized windows; ran layout tests.

[DaleCurtis, 2011-11-29 00:47:43]

PTAL and check I've got the right approach. I'll add a layout test and upstream
after local sign off.

[scherkus (not reviewing), 2011-11-29 01:19:29]

AFAIK you can do this via CSS on the user agent style sheet.

I tried this out and it seems to work fine:

--- a/Source/WebCore/css/mediaControlsChromium.css
+++ b/Source/WebCore/css/mediaControlsChromium.css
@@ -25,6 +25,11 @@
 
 /* Chromium default media controls */
 
+video:-webkit-full-page-media {
+    max-height: 100%;
+    max-width: 100%;
+}
+
 audio {
     width: 300px;
     height: 32px;
@@ -41,10 +46,6 @@ audio::-webkit-media-controls-panel,
video::-webkit-media-controls-panel {
     background-color: rgba(0, 0, 0, 0.6);
 }
 
-video:-webkit-full-page-media::-webkit-media-controls-panel {
-    bottom: 0px;
-}
-
 audio::-webkit-media-controls-mute-button,
video::-webkit-media-controls-mute-button {
     -webkit-appearance: media-mute-button;
     position: absolute;



Care to do some testing to see whether I'm missed something?

[DaleCurtis, 2011-11-30 01:33:31]

PTAL. Switched to CSS fix, added a layout test for <video>. Let me know if you
want one for <audio> too.

[scherkus (not reviewing), 2011-12-06 02:19:52]

cool!

you may need to skip this test under non-chromium platforms, for example:
http://trac.webkit.org/browser/trunk/LayoutTests/platform/mac/Skipped

http://codereview.chromium.org/8728018/diff/5001/LayoutTests/media/video-scal...
File LayoutTests/media/video-scales-in-media-document.html (right):

http://codereview.chromium.org/8728018/diff/5001/LayoutTests/media/video-scal...
LayoutTests/media/video-scales-in-media-document.html:14: var iframe =
document.getElementById("container");
instead of using IDs just use querySelector('iframe')

http://codereview.chromium.org/8728018/diff/5001/LayoutTests/media/video-scal...
LayoutTests/media/video-scales-in-media-document.html:16:
video.addEventListener("timeupdate", timeupdate);
use canplay instead of timeupdate

[DaleCurtis, 2011-12-06 21:13:44]

I'll let upstream decide which platforms get Skipped.

http://codereview.chromium.org/8728018/diff/5001/LayoutTests/media/video-scal...
File LayoutTests/media/video-scales-in-media-document.html (right):

http://codereview.chromium.org/8728018/diff/5001/LayoutTests/media/video-scal...
LayoutTests/media/video-scales-in-media-document.html:14: var iframe =
document.getElementById("container");
Interesting, I haven't used that before. By the numbers querySelector performs
much worse than getElementById:

http://jsperf.com/getelementbyid-vs-queryselector

Not that it matters for this test, but I'm curious to hear your thoughts. Done.

http://codereview.chromium.org/8728018/diff/5001/LayoutTests/media/video-scal...
LayoutTests/media/video-scales-in-media-document.html:16:
video.addEventListener("timeupdate", timeupdate);
On 2011/12/06 02:19:53, scherkus wrote:
> use canplay instead of timeupdate

Done.

[scherkus (not reviewing), 2011-12-08 06:43:51]

http://codereview.chromium.org/8728018/diff/5001/LayoutTests/media/video-scal...
File LayoutTests/media/video-scales-in-media-document.html (right):

http://codereview.chromium.org/8728018/diff/5001/LayoutTests/media/video-scal...
LayoutTests/media/video-scales-in-media-document.html:14: var iframe =
document.getElementById("container");
On 2011/12/06 21:13:45, DaleCurtis wrote:
> Interesting, I haven't used that before. By the numbers querySelector performs
> much worse than getElementById:
> 
> http://jsperf.com/getelementbyid-vs-queryselector
> 
> Not that it matters for this test, but I'm curious to hear your thoughts.
Done.

ids are unique in a document and as you might imagine could be implemented via
hash lookup

querySelector() requires a traversal of the DOM but I like using it for tests as
you don't need to muck around with ids and I've found it to be less brittle as
tests get updated

[scherkus (not reviewing), 2011-12-08 06:44:07]

also should we close this issue and work from the WK patch?

[DaleCurtis, 2011-12-08 20:53:33]

On 2011/12/08 06:44:07, scherkus wrote:
> also should we close this issue and work from the WK patch?

Yup, closing. https://bugs.webkit.org/show_bug.cgi?id=73948