Replace Pipeline::SetNetworkActivity() with BufferedDataSource -> WebMediaPlayerImpl callback.

Replace Pipeline::SetNetworkActivity() with BufferedDataSource -> WebMediaPlayerImpl callback.

This removes a bunch of plumbing that's completely unnecessary.  Concretely:
Before: BufferedResourceLoader -> BufferedDataSource -> DemuxerHost -> DataSourceHost -> Pipeline -> WebMediaPlayerImpl
After: BufferedResourceLoader -> BufferedDataSource -> WebMediaPlayerImpl


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=141619

[Ami GONE FROM CHROMIUM, 2012-06-10 21:43:46]

https://chromiumcodereview.appspot.com/10535101/diff/1/media/base/pipeline.cc
File media/base/pipeline.cc (left):

https://chromiumcodereview.appspot.com/10535101/diff/1/media/base/pipeline.cc...
media/base/pipeline.cc:486: media_log_->AddEvent(
Oops, forgot to move this to WMPI.  New upload has it.

[scherkus (not reviewing), 2012-06-12 02:55:04]

LGTM w/ nits

me like

I swore we had a bug for this filed but I guess not!

https://chromiumcodereview.appspot.com/10535101/diff/3001/media/tools/player_...
File media/tools/player_wtl/movie.cc (left):

https://chromiumcodereview.appspot.com/10535101/diff/3001/media/tools/player_...
media/tools/player_wtl/movie.cc:96: media::NetworkEventCB(),
my gosh the amount of code we have to support this is crazy

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/buffer...
File webkit/media/buffered_data_source.cc (right):

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/buffer...
webkit/media/buffered_data_source.cc:553: if (!downloading_cb_.is_null())
all clients provide a downloading_cb_ and docs in .h don't state what DownloadCB
does and whether it's required/optional/etc

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/buffer...
File webkit/media/buffered_data_source_unittest.cc (right):

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/buffer...
webkit/media/buffered_data_source_unittest.cc:41:
base::Bind(&MockBufferedDataSource::set_loading,
I think you need a downloading_ variable instead of loading_.

loading_ is used for simulating the correct WebURLLoaderClient methods in tests
(i.e., calling didFail() if you navigate away from the page) and not for
detecting deferred/undeferred

I'm actually kinda surprised no tests broke! :|

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/webmed...
File webkit/media/webmediaplayer_impl.cc (right):

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/webmed...
webkit/media/webmediaplayer_impl.cc:462: DCHECK_EQ(main_loop_,
MessageLoop::current());
left over debugging code?

it's fine to leave it in if you want but shouldn't readyState() also have the
same dcheck?

[Ami GONE FROM CHROMIUM, 2012-06-12 03:42:43]

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/buffer...
File webkit/media/buffered_data_source.cc (right):

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/buffer...
webkit/media/buffered_data_source.cc:553: if (!downloading_cb_.is_null())
On 2012/06/12 02:55:04, scherkus wrote:
> all clients provide a downloading_cb_ and docs in .h don't state what
DownloadCB
> does and whether it's required/optional/etc

Done.

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/buffer...
File webkit/media/buffered_data_source_unittest.cc (right):

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/buffer...
webkit/media/buffered_data_source_unittest.cc:41:
base::Bind(&MockBufferedDataSource::set_loading,
On 2012/06/12 02:55:04, scherkus wrote:
> I think you need a downloading_ variable instead of loading_.
> 
> loading_ is used for simulating the correct WebURLLoaderClient methods in
tests
> (i.e., calling didFail() if you navigate away from the page) and not for
> detecting deferred/undeferred
> 
> I'm actually kinda surprised no tests broke! :|

Done.

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/webmed...
File webkit/media/webmediaplayer_impl.cc (right):

https://chromiumcodereview.appspot.com/10535101/diff/3001/webkit/media/webmed...
webkit/media/webmediaplayer_impl.cc:462: DCHECK_EQ(main_loop_,
MessageLoop::current());
On 2012/06/12 02:55:04, scherkus wrote:
> left over debugging code?

No, copying from dataRate() above, I think.

> it's fine to leave it in if you want but shouldn't readyState() also have the
> same dcheck?

Sure.  Done.

[commit-bot: I haz the power, 2012-06-12 03:44:22]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/fischman@chromium.org/10535101/6002

[commit-bot: I haz the power, 2012-06-12 05:01:58]

Change committed as 141619