Cast: Added a EncodingEventSubscriber to cast sender app.

Cast: Added a EncodingEventSubscriber to cast sender app.

- Logs encoded uncompressed raw events for the first X seconds to a file path Y, where X and Y
  can be specified by the user.
- Also outputs some basic stats.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=251972

[imcheng, 2014-02-13 00:56:20]

Note on DEPS: Since I am adding a dependency on protobufs here, I am thinking
the serialization / compression components will also live in media/cast. On the
other hand, chrome/renderer/media will handle uploading.


The logic to serialize is taken from a simple program I've written
(https://codereview.chromium.org/161593002/). The other CL also contains
compression, decompression, and deserialize.

[Alpha Left Google, 2014-02-13 20:30:58]

https://codereview.chromium.org/162103002/diff/170001/media/cast/DEPS
File media/cast/DEPS (right):

https://codereview.chromium.org/162103002/diff/170001/media/cast/DEPS#newcode8
media/cast/DEPS:8: "+google/protobuf",
nit: order.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender.cc
File media/cast/test/sender.cc (right):

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:103: "/tmp/raw_events.log", INT_MIN, INT_MAX);
Save the log to current dir is better.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:342: // TODO(imcheng): Extract this function to a
class.
No need for this TODO.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:353: // Frame events - write size first, then write
entries
nit: period at the end of sentence.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:359: // Write size of the proto, then write the proto
nit: period at the end of sentence.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:370: // Write packet events
nit: period at the end of sentence.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:375: // Write size of the proto, then write the proto
nit: period at the end of sentence.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:387: // Write generic events
nit: period at the end of sentence.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:391: // Write size of the proto, then write the proto
nit: period at the end of sentence.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:416: // TODO(imcheng): Consider sending the maps to
another thread for processing
There's no need for this. This is a test app any way.

[imcheng, 2014-02-13 21:49:26]

Adding darin@ for DEPS. The only DEPS added are +google/protobuf and
+third_party/protobuf/src/google/protobuf.

https://codereview.chromium.org/162103002/diff/170001/media/cast/DEPS
File media/cast/DEPS (right):

https://codereview.chromium.org/162103002/diff/170001/media/cast/DEPS#newcode8
media/cast/DEPS:8: "+google/protobuf",
On 2014/02/13 20:30:58, Alpha wrote:
> nit: order.

Done. Now presubmit lists missing LGTM from 6 dependencies, but looks like I
will need LGTM from one of the top level owners either way.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender.cc
File media/cast/test/sender.cc (right):

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:103: "/tmp/raw_events.log", INT_MIN, INT_MAX);
On 2014/02/13 20:30:58, Alpha wrote:
> Save the log to current dir is better.

Done.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:342: // TODO(imcheng): Extract this function to a
class.
On 2014/02/13 20:30:58, Alpha wrote:
> No need for this TODO.

I want to do extract this to a class though so that it can be used by cast
extension code as well.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:353: // Frame events - write size first, then write
entries
On 2014/02/13 20:30:58, Alpha wrote:
> nit: period at the end of sentence.

Done.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:359: // Write size of the proto, then write the proto
On 2014/02/13 20:30:58, Alpha wrote:
> nit: period at the end of sentence.

Done.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:370: // Write packet events
On 2014/02/13 20:30:58, Alpha wrote:
> nit: period at the end of sentence.

Done.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:375: // Write size of the proto, then write the proto
On 2014/02/13 20:30:58, Alpha wrote:
> nit: period at the end of sentence.

Done.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:387: // Write generic events
On 2014/02/13 20:30:58, Alpha wrote:
> nit: period at the end of sentence.

Done.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:391: // Write size of the proto, then write the proto
On 2014/02/13 20:30:58, Alpha wrote:
> nit: period at the end of sentence.

Done.

https://codereview.chromium.org/162103002/diff/170001/media/cast/test/sender....
media/cast/test/sender.cc:416: // TODO(imcheng): Consider sending the maps to
another thread for processing
On 2014/02/13 20:30:58, Alpha wrote:
> There's no need for this. This is a test app any way.

Done.

[imcheng, 2014-02-18 20:39:25]

-darin from reviewer list, I changed the code so that it no longer requires the
use of google::protobuf classes.

[Alpha Left Google, 2014-02-18 21:52:01]

LGTM.

[imcheng, 2014-02-19 01:10:11]

The CQ bit was checked by imcheng@chromium.org

[commit-bot: I haz the power, 2014-02-19 01:10:51]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/imcheng@chromium.org/162103002/540001

[commit-bot: I haz the power, 2014-02-19 06:35:41]

Change committed as 251972

[Lei Zhang, 2014-02-19 06:51:18]

https://codereview.chromium.org/162103002/diff/540001/media/cast/test/sender.cc
File media/cast/test/sender.cc (right):

https://codereview.chromium.org/162103002/diff/540001/media/cast/test/sender....
media/cast/test/sender.cc:405: printf("Video frame map size: %lu\n",
frame_events.size());
Why are we doing printf() and not using LOG() ?

For printf(), you need the PRIuS macro from base/format_macros.h.

[Lei Zhang, 2014-02-19 06:51:49]

On 2014/02/19 06:51:18, Lei Zhang wrote:
>
https://codereview.chromium.org/162103002/diff/540001/media/cast/test/sender.cc
> File media/cast/test/sender.cc (right):
> 
>
https://codereview.chromium.org/162103002/diff/540001/media/cast/test/sender....
> media/cast/test/sender.cc:405: printf("Video frame map size: %lu\n",
> frame_events.size());
> Why are we doing printf() and not using LOG() ?
> 
> For printf(), you need the PRIuS macro from base/format_macros.h.

(Resend)

[Lei Zhang, 2014-02-19 06:52:20]

A revert of this CL has been created in
https://codereview.chromium.org/170283009/ by thestig@chromium.org.

The reason for reverting is: Broke 32 bit bots, printf() usage is non-portable..