Content browser test that checks gum callback is fired after failure

Added content browser test that checks gum callbacks are fired after an
error callback has fired. Test timesout after the 2nd gUM request if
a callback is not fired.

BUG=420048
TEST=Try jobs + I ran the content_browsertest on 40.0.2189.0 (which contains the bug)
and Tot (where the CL has been reverted).

Committed: https://crrev.com/1c493cf5ec7f771d2222f9a453d6278916b9fc46
Cr-Commit-Position: refs/heads/master@{#299863}

[jansson, 2014-10-15 14:28:44]

jansson@chromium.org changed reviewers:
+ phoglund@chromium.org, xians@google.com

[jansson, 2014-10-15 14:28:45]

jansson@chromium.org changed required reviewers:
+ xians@google.com

[jansson, 2014-10-15 14:28:46]

Please perform a code review, thanks!

[no longer working on chromium, 2014-10-15 14:31:54]

xians@chromium.org changed reviewers:
+ xians@chromium.org

[no longer working on chromium, 2014-10-15 14:31:55]

lgtm with nits.

https://codereview.chromium.org/653533004/diff/1/content/browser/media/webrtc...
File content/browser/media/webrtc_getusermedia_browsertest.cc (right):

https://codereview.chromium.org/653533004/diff/1/content/browser/media/webrtc...
content/browser/media/webrtc_getusermedia_browsertest.cc:515: std::string
constraints1 =
nit const std::string

https://codereview.chromium.org/653533004/diff/1/content/browser/media/webrtc...
content/browser/media/webrtc_getusermedia_browsertest.cc:523: std::string
constraints2 =
ditto

[phoglund_chromium, 2014-10-15 18:03:09]

lgtm w/ nits

https://codereview.chromium.org/653533004/diff/1/content/browser/media/webrtc...
File content/browser/media/webrtc_getusermedia_browsertest.cc (right):

https://codereview.chromium.org/653533004/diff/1/content/browser/media/webrtc...
content/browser/media/webrtc_getusermedia_browsertest.cc:504: // This test will
make two getUserMedia requests, first one triggers error
Nit: phrase better; what about
"This test makes two getUserMedia requests, one with impossible constraints that
should trigger an error, and one with valid constraints. The test verifies
getUserMedia can succeed after being given impossible constraints."

https://codereview.chromium.org/653533004/diff/1/content/browser/media/webrtc...
content/browser/media/webrtc_getusermedia_browsertest.cc:515: std::string
constraints1 =
Nit: Calling it constraints1|2 is misleading, since it's really a getUserMedia
call populated with parameters (one of which is the constraints). A better name
would gum_with_impossible_constraints and gum_with_vga_constraints.

[jansson, 2014-10-16 07:05:05]

Patchset #3 (id:40001) has been deleted

[jansson, 2014-10-16 07:05:13]

Patchset #2 (id:20001) has been deleted

[jansson, 2014-10-16 07:08:32]

jansson@chromium.org changed required reviewers:
- xians@google.com

[jansson, 2014-10-16 07:08:33]

Waiting for try jobs to finish, fixed according to your comments.

https://codereview.chromium.org/653533004/diff/1/content/browser/media/webrtc...
File content/browser/media/webrtc_getusermedia_browsertest.cc (right):

https://codereview.chromium.org/653533004/diff/1/content/browser/media/webrtc...
content/browser/media/webrtc_getusermedia_browsertest.cc:504: // This test will
make two getUserMedia requests, first one triggers error
On 2014/10/15 18:03:09, phoglund wrote:
> Nit: phrase better; what about
> "This test makes two getUserMedia requests, one with impossible constraints
that
> should trigger an error, and one with valid constraints. The test verifies
> getUserMedia can succeed after being given impossible constraints."

Done.

https://codereview.chromium.org/653533004/diff/1/content/browser/media/webrtc...
content/browser/media/webrtc_getusermedia_browsertest.cc:515: std::string
constraints1 =
On 2014/10/15 14:31:54, xians1 wrote:
> nit const std::string 

Done.

https://codereview.chromium.org/653533004/diff/1/content/browser/media/webrtc...
content/browser/media/webrtc_getusermedia_browsertest.cc:515: std::string
constraints1 =
On 2014/10/15 18:03:09, phoglund wrote:
> Nit: Calling it constraints1|2 is misleading, since it's really a getUserMedia
> call populated with parameters (one of which is the constraints). A better
name
> would gum_with_impossible_constraints and gum_with_vga_constraints.

Done.

https://codereview.chromium.org/653533004/diff/1/content/browser/media/webrtc...
content/browser/media/webrtc_getusermedia_browsertest.cc:523: std::string
constraints2 =
On 2014/10/15 14:31:54, xians1 wrote:
> ditto

Done.

[jansson, 2014-10-16 07:26:28]

The CQ bit was checked by jansson@chromium.org

[commit-bot: I haz the power, 2014-10-16 07:27:40]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/653533004/60001

[commit-bot: I haz the power, 2014-10-16 07:33:44]

Committed patchset #2 (id:60001)

[commit-bot: I haz the power, 2014-10-16 07:34:51]

Patchset 2 (id:??) landed as
https://crrev.com/1c493cf5ec7f771d2222f9a453d6278916b9fc46
Cr-Commit-Position: refs/heads/master@{#299863}