Fixed quality tests after PyAuto test purge.

Fixed quality tests after PyAuto test purge.

All test resources and reference files have been moved to be handled by
the webrtc.DEPS solution. The audio test used to find its files in the
pyauto_private folder, which has been removed. Instead I've uploaded
the files to the webrtc-resources bundle and put .sha1 files in the
webrtc test data folder, which are processed by the webrtc.DEPS
solution.

The downloading of the files is only triggered if you have the WebRTC
solution, so this should not burden people who do not want to run the
tests.

The reason for moving the files is that the tests are more healthy
if they don't have to look for resources outside the chrome/test/data
folder. I'm able to straighten out some hacks as a result.

Also fixing an issue where https://codereview.chromium.org/222873002
removed the web socket handler for the video quality test.

Depends on https://codereview.chromium.org/225263018.

BUG=360033, 359579
NOTRY=true

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=262398

[phoglund_chromium, 2014-04-07 12:12:55]

tommi: owner stamp
kjellander: main review

[phoglund_chromium, 2014-04-07 12:17:33]

On 2014/04/07 12:12:55, phoglund wrote:
> tommi: owner stamp
> kjellander: main review

Sorry hit send a bit early; I will adapt the audio test to use some more common
code in webrtc_browsertest_common.cc before I'm done.

[phoglund_chromium, 2014-04-07 12:31:13]

ok, go for it!

[phoglund_chromium, 2014-04-07 14:51:53]

On 2014/04/07 12:31:13, phoglund wrote:
> ok, go for it!

The patch is ok now, please review at your leisure.

[kjellander_chromium, 2014-04-07 14:59:57]

lgtm with a few comments.
Have you been able to test this on all platforms?

Also to avoid re-syncing the resources over and over: I'd expect at least a
.gitignore change in here as well (I assume svn:ignore changes needs to be done
in a separate manual commit).

https://codereview.chromium.org/227633002/diff/10018/chrome/browser/media/chr...
File chrome/browser/media/chrome_webrtc_video_quality_browsertest.cc (right):

https://codereview.chromium.org/227633002/diff/10018/chrome/browser/media/chr...
chrome/browser/media/chrome_webrtc_video_quality_browsertest.cc:82: // the video
from the reference files located int GetReferenceFilesDir().
int -> in

https://codereview.chromium.org/227633002/diff/10018/chrome/browser/media/web...
File chrome/browser/media/webrtc_browsertest_common.cc (left):

https://codereview.chromium.org/227633002/diff/10018/chrome/browser/media/web...
chrome/browser/media/webrtc_browsertest_common.cc:53: path.erase(found,
src_token_length);
Very nice to get rid of this hacky code!

https://codereview.chromium.org/227633002/diff/10018/chrome/test/data/webrtc/...
File chrome/test/data/webrtc/video_extraction.js (right):

https://codereview.chromium.org/227633002/diff/10018/chrome/test/data/webrtc/...
chrome/test/data/webrtc/video_extraction.js:193: return;
Just curious, why is this needed?

https://codereview.chromium.org/227633002/diff/10018/chrome/test/data/webrtc/...
File chrome/test/data/webrtc/wsh/webrtc_write_wsh.py (right):

https://codereview.chromium.org/227633002/diff/10018/chrome/test/data/webrtc/...
chrome/test/data/webrtc/wsh/webrtc_write_wsh.py:1: # Copyright (c) 2012 The
Chromium Authors. All rights reserved.
I assume this is an exact copy of the file that was removed?

https://codereview.chromium.org/227633002/diff/10018/chrome/test/data/webrtc/...
chrome/test/data/webrtc/wsh/webrtc_write_wsh.py:18: _WORKING_DIR =
os.path.join(os.environ[_HOME_ENV_NAME], 'webrtc_video_quality')
It would be great if you could create a temporary working dir to use instead,
but that can be done in a separate CL.

[phoglund_chromium, 2014-04-08 09:29:00]

https://codereview.chromium.org/227633002/diff/10018/chrome/browser/media/chr...
File chrome/browser/media/chrome_webrtc_video_quality_browsertest.cc (right):

https://codereview.chromium.org/227633002/diff/10018/chrome/browser/media/chr...
chrome/browser/media/chrome_webrtc_video_quality_browsertest.cc:82: // the video
from the reference files located int GetReferenceFilesDir().
On 2014/04/07 14:59:58, Henrik Kjellander wrote:
> int -> in

Done.

https://codereview.chromium.org/227633002/diff/10018/chrome/test/data/webrtc/...
File chrome/test/data/webrtc/video_extraction.js (right):

https://codereview.chromium.org/227633002/diff/10018/chrome/test/data/webrtc/...
chrome/test/data/webrtc/video_extraction.js:193: return;
On 2014/04/07 14:59:58, Henrik Kjellander wrote:
> Just curious, why is this needed?
The point of this code is "if the websocket is not open, call sendFrames 100 ms
later and hope it is open then". That's not what it does though; it schedules
sendFrames in 100 ms and proceeds to execute the code that is dependent on the
websocket being open. So the send on line 199 is just going to fail. This
doesn't make any sense which is why I added the return. This bug has probably
been here since the websocket has never failed to get up earlier.

https://codereview.chromium.org/227633002/diff/10018/chrome/test/data/webrtc/...
File chrome/test/data/webrtc/wsh/webrtc_write_wsh.py (right):

https://codereview.chromium.org/227633002/diff/10018/chrome/test/data/webrtc/...
chrome/test/data/webrtc/wsh/webrtc_write_wsh.py:1: # Copyright (c) 2012 The
Chromium Authors. All rights reserved.
On 2014/04/07 14:59:58, Henrik Kjellander wrote:
> I assume this is an exact copy of the file that was removed?
Yes.

https://codereview.chromium.org/227633002/diff/10018/chrome/test/data/webrtc/...
chrome/test/data/webrtc/wsh/webrtc_write_wsh.py:18: _WORKING_DIR =
os.path.join(os.environ[_HOME_ENV_NAME], 'webrtc_video_quality')
On 2014/04/07 14:59:58, Henrik Kjellander wrote:
> It would be great if you could create a temporary working dir to use instead,
> but that can be done in a separate CL.

I completely agree.

[tommi (sloooow) - chröme, 2014-04-08 10:18:44]

rslgtm

[phoglund_chromium, 2014-04-08 12:02:02]

The CQ bit was checked by phoglund@chromium.org

[commit-bot: I haz the power, 2014-04-08 12:02:10]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/phoglund@chromium.org/227633002/70001

[phoglund_chromium, 2014-04-08 12:02:14]

The CQ bit was unchecked by phoglund@chromium.org

[phoglund_chromium, 2014-04-08 13:41:33]

The CQ bit was checked by phoglund@chromium.org

[commit-bot: I haz the power, 2014-04-08 13:41:43]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/phoglund@chromium.org/227633002/70001

[commit-bot: I haz the power, 2014-04-08 14:20:42]

Change committed as 262398