Fix various MediaSource related crashes on Android.

Fix various MediaSource related crashes on Android.

This change fixes several issues that were causing crashes.
- MediaPlayerHostMsg_DemuxerReady_Params was not initializing the xxx_codec
  member variables so audio-only or video-only content would randomly be
  flagged as A/V content with a random codec ID.
- ChunkDemuxer::Stop() was not being called which could cause the delegate
  to be accessed after it was destroyed.
- MediaSourceDelegate was signalling that it had metadata before the 
  ChunkDemuxer was even initialized. This caused the HTMLMediaElement to
  signal that the load completed way too early and caused play() to be
  called when the delegate wasn't ready.

BUG=233420
TEST=webkitmediasource-play.html LayoutTest reliably runs w/o crashing now.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=201791

[qinmin, 2013-05-22 18:51:24]

lgtm%nits

https://codereview.chromium.org/15754004/diff/1/media/base/android/media_sour...
File media/base/android/media_source_player.cc (left):

https://codereview.chromium.org/15754004/diff/1/media/base/android/media_sour...
media/base/android/media_source_player.cc:211: OnMediaMetadataChanged(duration_,
width_, height_, false);
thanks for doing this, i was about to upload a CL doing the same thing

https://codereview.chromium.org/15754004/diff/1/webkit/media/android/media_so...
File webkit/media/android/media_source_delegate.cc (right):

https://codereview.chromium.org/15754004/diff/1/webkit/media/android/media_so...
webkit/media/android/media_source_delegate.cc:108: chunk_demuxer_->Stop(
Could there be any chance that message loop is gone first in fast shutdown case?
should we do sth in WillDestroyCurrentMessageLoop() in either WMPA or this
class?

https://codereview.chromium.org/15754004/diff/1/webkit/media/android/media_so...
File webkit/media/android/media_source_delegate.h (right):

https://codereview.chromium.org/15754004/diff/1/webkit/media/android/media_so...
webkit/media/android/media_source_delegate.h:46: inline void operator()(void* x)
const {
nit： s/x/media_source_delegate/

[acolwell GONE FROM CHROMIUM, 2013-05-22 19:12:09]

https://codereview.chromium.org/15754004/diff/1/webkit/media/android/media_so...
File webkit/media/android/media_source_delegate.cc (right):

https://codereview.chromium.org/15754004/diff/1/webkit/media/android/media_so...
webkit/media/android/media_source_delegate.cc:108: chunk_demuxer_->Stop(
On 2013/05/22 18:51:24, qinmin wrote:
> Could there be any chance that message loop is gone first in fast shutdown
case?
> should we do sth in WillDestroyCurrentMessageLoop() in either WMPA or this
> class?

Done. I added a media_source_delegate_.reset() to WMPA::Detach() since that
appears to be where all the other pointers are cleared.

https://codereview.chromium.org/15754004/diff/1/webkit/media/android/media_so...
File webkit/media/android/media_source_delegate.h (right):

https://codereview.chromium.org/15754004/diff/1/webkit/media/android/media_so...
webkit/media/android/media_source_delegate.h:46: inline void operator()(void* x)
const {
On 2013/05/22 18:51:24, qinmin wrote:
> nit： s/x/media_source_delegate/

Done.

[commit-bot: I haz the power, 2013-05-22 19:13:05]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/acolwell@chromium.org/15754004/5001

[commit-bot: I haz the power, 2013-05-22 21:07:45]

Sorry for I got bad news for ya.
Compile failed with a clobber build on android_dbg.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android_db...
Your code is likely broken or HEAD is junk. Please ensure your
code is not broken then alert the build sheriffs.
Look at the try server FAQ for more details.

[commit-bot: I haz the power, 2013-05-22 21:11:36]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/acolwell@chromium.org/15754004/26002

[commit-bot: I haz the power, 2013-05-23 14:27:20]

Change committed as 201791