Roll swarming_client @ 9cc42c3d4d.

Roll swarming_client @ 9cc42c3d4d.

This requires small changes in build/isolate.gypi, the isolate.py --variable
flag was split in three.

$ git log aab3a62beb..9cc42c3d4d --date=short --format="%ad %ae %s" | sed 's/@chromium\.org//'
2013-12-11 maruel Ensure variables are decoded to unicode.
2013-12-06 maruel Rename --command-variable to --extra-variables.
2013-12-06 maruel Add real variable support and split --variable in 3 different flags.
2013-12-05 vadimsh Add 'offset' parameter to StorageApi.fetch(...) method.
2013-12-05 maruel Change log message to be clearer.
2013-12-05 vadimsh Support custom HTTP headers in net.url_open.
2013-12-03 vadimsh Remove urllib2 support.
2013-12-03 vadimsh Wait for Storage's internal threads to finish before exiting.
2013-12-03 vadimsh Fix swarming_smoke_test.py and isolateserver_test.py.

R=vadimsh@chromium.org
BUG=324911
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=240155

[M-A Ruel, 2013-12-05 00:13:13]

I'll update this commit once https://codereview.appspot.com/37180043 is
committed. I just wanted to test it locally first. Seems to work. I'll update
this CL with the proper commit hash right after pushing the other one. I'd
prefer to keep the delay between both relatively short since the canary will be
broken in the meantime.

[M-A Ruel, 2013-12-06 14:20:23]

Now with real roll.

[M-A Ruel, 2013-12-06 15:22:54]

+Marcus for android change. I split the --variable flag depending on the type of
flags.

[M-A Ruel, 2013-12-06 15:33:32]

On 2013/12/06 15:22:54, M-A Ruel wrote:
> +Marcus for android change. I split the --variable flag depending on the type
of
> flags.

Hummm to fix mac_rel, I'll have to do another fix in swarming.client. :/ So this
roll will have to wait a bit.

[bulach, 2013-12-06 15:44:23]

lgtm for android, thanks! (btw, frankf@ is a better reviewer closer to your
timezone for build/android changes).

[Vadim Sh., 2013-12-06 21:54:41]

lgtm with nit

https://codereview.chromium.org/105933002/diff/80001/build/isolate.gypi
File build/isolate.gypi (left):

https://codereview.chromium.org/105933002/diff/80001/build/isolate.gypi#oldco...
build/isolate.gypi:46:
'<(DEPTH)/tools/swarming_client/googletest/run_test_cases.py',
Is there any bit of infrastructure that is still using run_test_cases.py?

Maybe just remove it from isolate client? :)

https://codereview.chromium.org/105933002/diff/80001/build/isolate.gypi
File build/isolate.gypi (right):

https://codereview.chromium.org/105933002/diff/80001/build/isolate.gypi#newco...
build/isolate.gypi:72: '--config-variable', 'OS=<(OS)',
nit: why not '--config-variable', 'OS', '<(OS)'? Looks like it's more preferred
form overall.

[M-A Ruel, 2013-12-06 22:01:45]

https://codereview.chromium.org/105933002/diff/80001/build/isolate.gypi
File build/isolate.gypi (left):

https://codereview.chromium.org/105933002/diff/80001/build/isolate.gypi#oldco...
build/isolate.gypi:46:
'<(DEPTH)/tools/swarming_client/googletest/run_test_cases.py',
On 2013/12/06 21:54:41, Vadim Sh. wrote:
> Is there any bit of infrastructure that is still using run_test_cases.py?
> 
> Maybe just remove it from isolate client? :)

I won't remove it from isolate client since it is still useful to bootstrap
.isolate files.

https://codereview.chromium.org/105933002/diff/80001/build/isolate.gypi
File build/isolate.gypi (right):

https://codereview.chromium.org/105933002/diff/80001/build/isolate.gypi#newco...
build/isolate.gypi:72: '--config-variable', 'OS=<(OS)',
On 2013/12/06 21:54:41, Vadim Sh. wrote:
> nit: why not '--config-variable', 'OS', '<(OS)'? Looks like it's more
preferred
> form overall.

I prefer not because of gyp's habit of stripping redundant arguments.

[Vadim Sh., 2013-12-11 02:18:50]

Is there anything that block this roll?

I'm mostly concerned about all red swarming canary...

[M-A Ruel, 2013-12-11 02:21:55]

On 2013/12/11 02:18:50, Vadim Sh. wrote:
> Is there anything that block this roll?
> 
> I'm mostly concerned about all red swarming canary...

It's the assert in android, I'll take a look as soon as I can.

[M-A Ruel, 2013-12-11 13:47:39]

On 2013/12/11 02:21:55, M-A Ruel wrote:
> On 2013/12/11 02:18:50, Vadim Sh. wrote:
> > Is there anything that block this roll?
> > 
> > I'm mostly concerned about all red swarming canary...
> 
> It's the assert in android, I'll take a look as soon as I can.

https://codereview.appspot.com/39100045 should be able to get this unblocked.

[commit-bot: I haz the power, 2013-12-11 14:42:00]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/maruel@chromium.org/105933002/100001

[commit-bot: I haz the power, 2013-12-11 14:48:47]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/maruel@chromium.org/105933002/100001

[commit-bot: I haz the power, 2013-12-11 19:44:33]

Change committed as 240155

[ckocagil, 2013-12-12 02:51:30]

A revert of this CL has been created in
https://codereview.chromium.org/111393006/ by ckocagil@chromium.org.

The reason for reverting is: tree breakage: "runisolatedtest.py:108 ERROR
Unexpected isolate version 1.2".