Added tab-to-tab apprtc call.

Added tab-to-tab apprtc call.


BUG=
TEST=Ran test locally.
NOTRY=True

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=175503

[dennis_jeffrey, 2012-12-19 18:13:32]

One question.

https://codereview.chromium.org/11648013/diff/1/chrome/test/functional/webrtc...
File chrome/test/functional/webrtc_apprtc_call.py (right):

https://codereview.chromium.org/11648013/diff/1/chrome/test/functional/webrtc...
chrome/test/functional/webrtc_apprtc_call.py:38: random_call_id = 'pyauto%d' %
random.randint(0, 65536)
what benefit do we get by using a random int in the call_id here?  Could we just
use a constant value?

I'd just like to know whether it's worthwhile to add non-determinism to this
test, since (in general) non-determinism could potentially be a source of
flakiness.  If it serves a particular purpose here, it may help to add a comment
to the code explaining why.

[phoglund_chromium, 2013-01-02 10:18:55]

https://codereview.chromium.org/11648013/diff/1/chrome/test/functional/webrtc...
File chrome/test/functional/webrtc_apprtc_call.py (right):

https://codereview.chromium.org/11648013/diff/1/chrome/test/functional/webrtc...
chrome/test/functional/webrtc_apprtc_call.py:38: random_call_id = 'pyauto%d' %
random.randint(0, 65536)
On 2012/12/19 18:13:32, dennis_jeffrey wrote:
> what benefit do we get by using a random int in the call_id here?  Could we
just
> use a constant value?
> 
> I'd just like to know whether it's worthwhile to add non-determinism to this
> test, since (in general) non-determinism could potentially be a source of
> flakiness.  If it serves a particular purpose here, it may help to add a
comment
> to the code explaining why.
The ID is used to refer to a call session, which is why we use one here to get
the two tabs into the same call. The random ID is here since the apprtc service
is stateful - if we used the same ID I can imagine that we would get problems
with hung calls and similar issues. I'll add a comment about it.

[phoglund_chromium, 2013-01-07 12:17:13]

Hi!

Since Dennis is away in India, please have a look at this one.

[craigdh, 2013-01-07 17:39:07]

lgtm. PyAuto is deprecated, however, so there is a limited lifespan for any
library depending heavily on it.

[commit-bot: I haz the power, 2013-01-08 08:55:53]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/phoglund@chromium.org/11648013/5001

[commit-bot: I haz the power, 2013-01-08 08:56:06]

Change committed as 175503