Make WebRtcWebcamBrowserTest fail faster.

Make WebRtcWebcamBrowserTest fail faster.
When these tests fail, they time out after a long preriod of time instead of stopping when there's an error.

The specific error case I was hitting was that my camera didn't support the highest resolution. The error is to be expected given my configuration, but there's no need to wait for a timeout.

R=phoglund@chromium.org

Committed: https://crrev.com/e5b3f30d5b078a964e72f2fcec3a3f4fded79492
Cr-Commit-Position: refs/heads/master@{#327017}

[tommi (sloooow) - chröme, 2015-04-03 13:37:11]

Fix oopsy

[tommi (sloooow) - chröme, 2015-04-03 13:54:14]

.

[tommi (sloooow) - chröme, 2015-04-03 19:11:30]

Remove expectation from gUM calls and improve documentation

[phoglund_chromium, 2015-04-07 08:00:54]

https://codereview.chromium.org/1061473002/diff/60001/chrome/browser/media/we...
File chrome/browser/media/webrtc_browsertest_base.cc (right):

https://codereview.chromium.org/1061473002/diff/60001/chrome/browser/media/we...
chrome/browser/media/webrtc_browsertest_base.cc:119: return
test::PollingWaitUntil("obtainGetUserMediaResult()", kOkGotStream,
This is called in five places:
https://code.google.com/p/chromium/codesearch#chromium/src/chrome/browser/med...;
please add EXPECT_TRUE statements there so we don't just lose the failure.
Alternatively, EXPECT_TRUE like you do on line 344.

[phoglund_chromium, 2015-04-21 13:38:30]

Don't forget this one :) just had some minor comments.

[tommi (sloooow) - chröme, 2015-04-24 13:17:20]

Address comments

[tommi (sloooow) - chröme, 2015-04-24 13:17:43]

https://codereview.chromium.org/1061473002/diff/60001/chrome/browser/media/we...
File chrome/browser/media/webrtc_browsertest_base.cc (right):

https://codereview.chromium.org/1061473002/diff/60001/chrome/browser/media/we...
chrome/browser/media/webrtc_browsertest_base.cc:119: return
test::PollingWaitUntil("obtainGetUserMediaResult()", kOkGotStream,
On 2015/04/07 08:00:54, phoglund wrote:
> This is called in five places:
>
https://code.google.com/p/chromium/codesearch#chromium/src/chrome/browser/med...;
> please add EXPECT_TRUE statements there so we don't just lose the failure.
> Alternatively, EXPECT_TRUE like you do on line 344.

I made sure all callers to GetUserMediaWithSpecificConstraintsAndAccept and
GetUserMediaAndAccept, use the return value either to handle the error or use
EXPECT_TRUE().

The PollingWaitUntil function itself still returns false in cases where it fails
and that the failure is not the same failure as was expected  (2nd argument to
PollingWaitUntil), so I there shouldn't be a change in behavior as far as that
goes - but I could be missing something here.

[phoglund_chromium, 2015-04-24 14:04:28]

lgtm with last comment fixed

https://codereview.chromium.org/1061473002/diff/60001/chrome/browser/media/we...
File chrome/browser/media/webrtc_browsertest_base.cc (right):

https://codereview.chromium.org/1061473002/diff/60001/chrome/browser/media/we...
chrome/browser/media/webrtc_browsertest_base.cc:119: return
test::PollingWaitUntil("obtainGetUserMediaResult()", kOkGotStream,
On 2015/04/24 13:17:43, tommi wrote:
> On 2015/04/07 08:00:54, phoglund wrote:
> > This is called in five places:
> >
>
https://code.google.com/p/chromium/codesearch#chromium/src/chrome/browser/med...;
> > please add EXPECT_TRUE statements there so we don't just lose the failure.
> > Alternatively, EXPECT_TRUE like you do on line 344.
> 
> I made sure all callers to GetUserMediaWithSpecificConstraintsAndAccept and
> GetUserMediaAndAccept, use the return value either to handle the error or use
> EXPECT_TRUE().
> 
> The PollingWaitUntil function itself still returns false in cases where it
fails
> and that the failure is not the same failure as was expected  (2nd argument to
> PollingWaitUntil), so I there shouldn't be a change in behavior as far as that
> goes - but I could be missing something here.

Ah, I could have been more specific. You missed one on line 196 in this file; a
failure from there would fail the expect_true earlier but now does nothing since
it doesn't handle the return value. You seem to have covered the other places.

[tommi (sloooow) - chröme, 2015-04-26 10:52:25]

Fix issue in js code and improve logging of constraints

[tommi (sloooow) - chröme, 2015-04-26 10:55:02]

Hej Patrik - I've updated the code and I believe I found an issue with the js
code while doing so.  Can you do a sanity check?

https://codereview.chromium.org/1061473002/diff/100001/chrome/test/data/webrt...
File chrome/test/data/webrtc/getusermedia.js (right):

https://codereview.chromium.org/1061473002/diff/100001/chrome/test/data/webrt...
chrome/test/data/webrtc/getusermedia.js:79: gRequestWebcamAndMicrophoneResult =
'not-called-yet';
The problem we were hitting here is that when doing two gUM calls in a row, the
script would return the result from the first run for the second call.

[phoglund_chromium, 2015-04-27 07:46:48]

lgtm

https://codereview.chromium.org/1061473002/diff/100001/chrome/test/data/webrt...
File chrome/test/data/webrtc/getusermedia.js (right):

https://codereview.chromium.org/1061473002/diff/100001/chrome/test/data/webrt...
chrome/test/data/webrtc/getusermedia.js:79: gRequestWebcamAndMicrophoneResult =
'not-called-yet';
On 2015/04/26 10:55:02, tommi wrote:
> The problem we were hitting here is that when doing two gUM calls in a row,
the
> script would return the result from the first run for the second call.

Good catch. It's possible the re-set should actually be in doGetUserMedia, but I
think the calling code doesn't rely on getting the same answer several times.

[tommi (sloooow) - chröme, 2015-04-27 08:50:07]

The CQ bit was checked by tommi@chromium.org

[commit-bot: I haz the power, 2015-04-27 08:50:19]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1061473002/100001

[commit-bot: I haz the power, 2015-04-27 08:53:43]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2015-04-27 08:55:20]

Patchset 6 (id:??) landed as
https://crrev.com/e5b3f30d5b078a964e72f2fcec3a3f4fded79492
Cr-Commit-Position: refs/heads/master@{#327017}