[Android] Add --repeat for gtests and instrumentation tests.

[Android] Add --repeat for gtests and instrumentation tests.

BUG=285037, 341244
Committed: https://crrev.com/eb7ea71cab3a2511561a6336e00e39731472d616
Cr-Commit-Position: refs/heads/master@{#351085}

[jbudorick, 2015-09-28 13:14:19]

jbudorick@chromium.org changed reviewers:
+ mikecase@chromium.org

[jbudorick, 2015-09-28 13:14:31]

The CQ bit was checked by jbudorick@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-09-28 13:14:34]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1376483002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1376483002/20001

[mikecase (-- gone --), 2015-09-28 13:37:58]

Pretty much looks good to me, but have 1 comment. What will this be used for
btw.

Also, something related to this CL that I am trying to fix right now, is if you
repeat an test on an appurify bot, the tests results from the last test will
overwrite the rest (because the test results default to be saved in
<results_dir>/<suite>.json and they all have the same suite name)  :-/

https://codereview.chromium.org/1376483002/diff/20001/build/android/test_runn...
File build/android/test_runner.py (right):

https://codereview.chromium.org/1376483002/diff/20001/build/android/test_runn...
build/android/test_runner.py:955: else itertools.count())
From what I see online, itertools.count() will just keep counting forever.

When would someone set args.repeat to be < 0?

[jbudorick, 2015-09-28 13:40:16]

This will be used by devs, not bots.

https://codereview.chromium.org/1376483002/diff/20001/build/android/test_runn...
File build/android/test_runner.py (right):

https://codereview.chromium.org/1376483002/diff/20001/build/android/test_runn...
build/android/test_runner.py:955: else itertools.count())
On 2015/09/28 at 13:37:58, mikecase wrote:
> From what I see online, itertools.count() will just keep counting forever.
> 
> When would someone set args.repeat to be < 0?

For infinite repetition, of course:
https://code.google.com/p/googletest/wiki/AdvancedGuide#Repeating_the_Tests

[mikecase (-- gone --), 2015-09-28 14:00:12]

lgtm with nit

https://codereview.chromium.org/1376483002/diff/20001/build/android/test_runn...
File build/android/test_runner.py (right):

https://codereview.chromium.org/1376483002/diff/20001/build/android/test_runn...
build/android/test_runner.py:683: exit_code = 0
nit: This don't think this exit_code var is really used anymore in this
function. Its just set to 0 here and then used in the line

if test_exit_code and exit_code != constants.ERROR_EXIT_CODE:

which I guess is just really

if test_exit_code and constants.ERROR_EXIT_CODE != 0:




Previously, it looks like the python_tests would set the test_exit_code variable
and the java_tests would set the exit_code variable. hmm...

https://codereview.chromium.org/1376483002/diff/20001/build/android/test_runn...
build/android/test_runner.py:955: else itertools.count())
On 2015/09/28 at 13:40:16, jbudorick wrote:
> On 2015/09/28 at 13:37:58, mikecase wrote:
> > From what I see online, itertools.count() will just keep counting forever.
> > 
> > When would someone set args.repeat to be < 0?
> 
> For infinite repetition, of course:
https://code.google.com/p/googletest/wiki/AdvancedGuide#Repeating_the_Tests

Ah, this makes sense then.

[jbudorick, 2015-09-28 14:03:21]

https://codereview.chromium.org/1376483002/diff/20001/build/android/test_runn...
File build/android/test_runner.py (right):

https://codereview.chromium.org/1376483002/diff/20001/build/android/test_runn...
build/android/test_runner.py:683: exit_code = 0
On 2015/09/28 at 14:00:12, mikecase wrote:
> nit: This don't think this exit_code var is really used anymore in this
function. Its just set to 0 here and then used in the line
> 
> if test_exit_code and exit_code != constants.ERROR_EXIT_CODE:
> 
> which I guess is just really
> 
> if test_exit_code and constants.ERROR_EXIT_CODE != 0:

It gets modified in the body of that if statement, then returned at the end of
this function.

> 
> 
> 
> 
> Previously, it looks like the python_tests would set the test_exit_code
variable and the java_tests would set the exit_code variable. hmm...

the if statement should actually be after both the java tests and the python
tests

[jbudorick, 2015-09-28 14:28:05]

The CQ bit was checked by jbudorick@chromium.org

[jbudorick, 2015-09-28 14:28:05]

The patchset sent to the CQ was uploaded after l-g-t-m from
mikecase@chromium.org
Link to the patchset: https://codereview.chromium.org/1376483002/#ps40001
(title: " ")

[commit-bot: I haz the power, 2015-09-28 14:28:17]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1376483002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1376483002/40001

[commit-bot: I haz the power, 2015-09-28 16:40:30]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-09-28 16:42:38]

Patchset 3 (id:??) landed as
https://crrev.com/eb7ea71cab3a2511561a6336e00e39731472d616
Cr-Commit-Position: refs/heads/master@{#351085}