[Cast] Clean-up RtpCastHeader and RtpParser, removing the last WebRTC dependency.

[Cast] Clean-up RtpCastHeader and RtpParser, removing the last WebRTC dependency.

Removed use of a WebRTC struct to hold the RTP-specific header values,
and only include those struct members needed by Cast in RtpCastHeader.
Also:

1. Cleaned up RtpParser, and fixed bugs relating to 8-->32 bit frame_id
   and reference_frame_id expansion.
2. Moved call to OnReceivedPayloadData() out of RtpParser::ParsePacket()
   and into RtpReceiver.
3. Massaged unit tests around RtpCastHeader changes.
4. Coding style violations fixed where code was touched.

Testing: All cast_unittests pass.  Also, manually casted from Chrome to cast_receiver_app to confirm everything works.

BUG=347391
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=266120

[miu, 2014-04-23 23:41:58]

hubbe: PTAL at everything.  Start with rtp_receiver_defines.h, then rtp_parser*,
then the rest is mostly variable renaming propagating throughout the code.

imcheng: PTAL at end2end_unittest.cc changes.  The logging test code works for
all frames (not just the first) now.

Miscellaneous notes follow:

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
File media/cast/rtp_receiver/rtp_parser/rtp_parser.cc (right):

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
media/cast/rtp_receiver/rtp_parser/rtp_parser.cc:28: bool
RtpParser::ParsePacket(const uint8* packet,
I consolidated all the ParseXXX() methods into one, and then documented the code
a bit more to reveal assumptions I had to painstakingly discover.  In addition,
I fixed the 8-->32 bit frame_id expansion logic, since the expansion is NOT
stateless.

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
File media/cast/rtp_receiver/rtp_parser/rtp_parser.h (left):

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
media/cast/rtp_receiver/rtp_parser/rtp_parser.h:14: struct RtpParserConfig {
I eleminated this, and replaced it with two arguments to the RtpParser ctor.

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
File media/cast/rtp_receiver/rtp_parser/rtp_parser_unittest.cc (right):

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
media/cast/rtp_receiver/rtp_parser/rtp_parser_unittest.cc:167:
TEST_F(RtpParserTest, ParseExpandingFrameIdTo32Bits) {
Two new tests to confirm the 8-->32 bit frame_id expansion logic works.

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
File media/cast/rtp_receiver/rtp_receiver_defines.h (left):

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
media/cast/rtp_receiver/rtp_receiver_defines.h:30: bool is_reference;  // Set to
true if the previous frame is not available,
I removed this because it is solely a detail of the
serialization/deserialization of packets.  I confirmed no downstream
implementation actually cares whether the reference_frame_id was explicitly
provided in the packet (as opposed to being set to a default value based on
frame_id).

https://codereview.chromium.org/250363002/diff/10001/media/cast/test/end2end_...
File media/cast/test/end2end_unittest.cc (left):

https://codereview.chromium.org/250363002/diff/10001/media/cast/test/end2end_...
media/cast/test/end2end_unittest.cc:992: TEST_F(End2EndTest,
ResetReferenceFrameId) {
I removed this test because it's not a good test: Not only does it violate the
abstraction layers in the code, but it's not functionally correct (it ignores
the fact that clearing the "reference frame id bit" reduces the size of the Cast
header in the packet by one byte).

https://codereview.chromium.org/250363002/diff/10001/media/cast/test/end2end_...
media/cast/test/end2end_unittest.cc:1266: // Verify that each frame have the
expected types of events logged.
imcheng: Looks like this isn't broken anymore.  In any case, the tests pass when
I iterate over the entire map now.  :)

[hubbe, 2014-04-24 00:31:12]

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
File media/cast/rtp_receiver/rtp_receiver.h (right):

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
media/cast/rtp_receiver/rtp_receiver.h:20: class RtpReceiver : public RtpParser
{
No need to inherit the parser anymore.
Please change to composition.
(Maybe add a todo to do the same sort of thing one step up, where
audio/video_receiver inherits RtpReceiver)

[miu, 2014-04-24 21:03:55]

PTAL.

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
File media/cast/rtp_receiver/rtp_receiver.h (right):

https://codereview.chromium.org/250363002/diff/10001/media/cast/rtp_receiver/...
media/cast/rtp_receiver/rtp_receiver.h:20: class RtpReceiver : public RtpParser
{
On 2014/04/24 00:31:13, hubbe wrote:
> No need to inherit the parser anymore.
> Please change to composition.
> (Maybe add a todo to do the same sort of thing one step up, where
> audio/video_receiver inherits RtpReceiver)

Done.

[hubbe, 2014-04-24 22:00:04]

lgtm

[imcheng, 2014-04-24 22:02:40]

lgtm

end2end_unittest.cc LGTM.

[miu, 2014-04-24 22:04:02]

The CQ bit was checked by miu@chromium.org

[commit-bot: I haz the power, 2014-04-24 22:05:20]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/miu@chromium.org/250363002/30001

[commit-bot: I haz the power, 2014-04-25 03:33:45]

Change committed as 266120