[Android] Generate scripts at build time to run android tests.

[Android] Generate scripts at build time to run android tests.

Making changes so that at build time scripts will be generated that can be used to easily run tests.

BUG=490781
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Committed: https://crrev.com/56d80d7a4c1ff50bf21112486cc9b2308c7e7bf0
Cr-Commit-Position: refs/heads/master@{#332518}

[mikecase (-- gone --), 2015-05-01 00:53:27]

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

[mikecase (-- gone --), 2015-05-01 00:53:28]

If you could take a quick look over this and tell me if its what you envisioned
it would help me out.

Right now just figuring our where to put the gyp action is causing me the most
trouble. I ended up making a new test_runner.gypi file. You can include it in a
gtest or instrumentation test target (as long as you specify the test_type
variable) and it will make a little runner script for you in out_dir/bin.
Thoughts?

[mikecase (-- gone --), 2015-05-01 15:45:52]

#!/usr/bin/env python
#
By the way, here is an example of the (out/Debug/bin/run_base_unittests)
generated when you build base_unittests_apk.

# This file was generated by build/android/gyp/create_test_runner_script.py

import os
import sys

script_directory = os.path.dirname(__file__)

def ResolvePath(path):
  """Returns an absolute filepath given a path relative to this script.
  """
  return os.path.abspath(os.path.join(script_directory, path))

test_runner_path = ResolvePath('../../../build/android/test_runner.py')
test_runner_args = ['gtest', '--verbose', '--suite', 'base_unittests']
test_runner_path_args = {'--output-directory': '..', '--isolate-file-path':
'../../../base/base_unittests.isolate'}
for arg, path in test_runner_path_args.iteritems():
  test_runner_args.extend([arg, ResolvePath(path)])

test_runner_cmd = [test_runner_path] + test_runner_args + sys.argv[1:]
os.system(' '.join(test_runner_cmd))

[mikecase (-- gone --), 2015-05-01 15:48:18]

(Messed up formatting for previous comment.) Here is an example of the
(out/Debug/bin/run_base_unittests)
generated when you build base_unittests_apk.


#!/usr/bin/env python
#
# This file was generated by build/android/gyp/create_test_runner_script.py

import os
import sys

script_directory = os.path.dirname(__file__)

def ResolvePath(path):
  """Returns an absolute filepath given a path relative to this script.
  """
  return os.path.abspath(os.path.join(script_directory, path))

test_runner_path = ResolvePath('../../../build/android/test_runner.py')
test_runner_args = ['gtest', '--verbose', '--suite', 'base_unittests']
test_runner_path_args = {'--output-directory': '..', '--isolate-file-path':
'../../../base/base_unittests.isolate'}
for arg, path in test_runner_path_args.iteritems():
  test_runner_args.extend([arg, ResolvePath(path)])

test_runner_cmd = [test_runner_path] + test_runner_args + sys.argv[1:]
os.system(' '.join(test_runner_cmd))

[jbudorick, 2015-05-01 15:59:15]

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

[jbudorick, 2015-05-01 15:59:16]

+cjhopman: wdyt about this approach?

https://codereview.chromium.org/1120883002/diff/60001/build/android/gyp/creat...
File build/android/gyp/create_test_runner_script.py (right):

https://codereview.chromium.org/1120883002/diff/60001/build/android/gyp/creat...
build/android/gyp/create_test_runner_script.py:29: test_runner_path =
ResolvePath('{test_runner_path}')
All of this should be in a main().

https://codereview.chromium.org/1120883002/diff/60001/build/android/test_runn...
File build/android/test_runner.gypi (right):

https://codereview.chromium.org/1120883002/diff/60001/build/android/test_runn...
build/android/test_runner.gypi:8: # To use this, include this file in a gtest or
instrumentation test target.
This should document the required and optional variables. Other .gypis should
have reasonable examples.

https://codereview.chromium.org/1120883002/diff/60001/build/android/test_runn...
build/android/test_runner.gypi:37: 'test_runner_args': ['--verbose',
'--output-directory', '<(PRODUCT_DIR)'],
I'm not sure we want to include --verbose by default. I know the bots do, but
I'm not sure the devs necessarily do.

https://codereview.chromium.org/1120883002/diff/60001/build/android/test_runn...
build/android/test_runner.gypi:53: {
nit: this needs to be indented

[mckev, 2015-05-01 17:12:08]

mckev@amazon.com changed reviewers:
+ mckev@amazon.com

[mckev, 2015-05-01 17:12:09]

Just passing by - specifying the isolate file from GYP files will certainly be
of value to embedders as well, since the test_runner currently relies on
hard-coded isolate file paths.

Note that this would be valuable for instrumentation tests as well, since
isolates are also hard-coded in build/android/pylib/instrumentation/setup.py.

https://codereview.chromium.org/1120883002/diff/100001/base/base.gyp
File base/base.gyp (right):

https://codereview.chromium.org/1120883002/diff/100001/base/base.gyp#newcode1558
base/base.gyp:1558: 'isolate_file': 'base_unittests.isolate',
Should the hard-coded isolate file paths be removed from
build/android/pylib/gtest/setup.py once this change lands, and the bots are
updated to use these scripts?

[mikecase (-- gone --), 2015-05-01 17:27:05]

https://codereview.chromium.org/1120883002/diff/100001/base/base.gyp
File base/base.gyp (right):

https://codereview.chromium.org/1120883002/diff/100001/base/base.gyp#newcode1558
base/base.gyp:1558: 'isolate_file': 'base_unittests.isolate',
On 2015/05/01 17:12:09, mckev wrote:
> Should the hard-coded isolate file paths be removed from
> build/android/pylib/gtest/setup.py once this change lands, and the bots are
> updated to use these scripts?

I think that is the plan. Once all of the bots specify the isolate file through
the testrunner option (--isolate-file-path), we can remove the hardcoded paths
in b/a/pylib/gtest/setup.py and b/a/pylib/instrumentation/setup.py.

[cjhopman, 2015-05-01 17:39:00]

On 2015/05/01 15:59:16, jbudorick wrote:
> +cjhopman: wdyt about this approach?

I think this approach is great.

[cjhopman, 2015-05-01 17:40:11]

Needs GN, too.

https://codereview.chromium.org/1120883002/diff/60001/build/android/test_runn...
File build/android/test_runner.gypi (right):

https://codereview.chromium.org/1120883002/diff/60001/build/android/test_runn...
build/android/test_runner.gypi:37: 'test_runner_args': ['--verbose',
'--output-directory', '<(PRODUCT_DIR)'],
On 2015/05/01 15:59:16, jbudorick wrote:
> I'm not sure we want to include --verbose by default. I know the bots do, but
> I'm not sure the devs necessarily do.

Yeah, I'd drop --verbose from the defaults

[mikecase (-- gone --), 2015-05-13 02:09:33]

Updated CL addressing cjhopman's comments (added GN and got rid of verbose from
default test runner options). I also made changes so scripts to run
instrumentation tests are generated when you build with gyp (previously I was
only generating scripts when building gtests).

[jbudorick, 2015-05-15 13:19:57]

My only question is about instrumentation tests + gn.

https://codereview.chromium.org/1120883002/diff/200001/build/android/gyp/crea...
File build/android/gyp/create_test_runner_script.py (right):

https://codereview.chromium.org/1120883002/diff/200001/build/android/gyp/crea...
build/android/gyp/create_test_runner_script.py:16: script_template = """\
nit: should be SCRIPT_TEMPLATE (or _SCRIPT_TEMPLATE)

https://codereview.chromium.org/1120883002/diff/200001/build/android/gyp/crea...
build/android/gyp/create_test_runner_script.py:32: test_runner_path =
ResolvePath('{test_runner_path}')
put an if __name__ == '__main__': guard in here, please.

https://codereview.chromium.org/1120883002/diff/200001/build/config/android/r...
File build/config/android/rules.gni (right):

https://codereview.chromium.org/1120883002/diff/200001/build/config/android/r...
build/config/android/rules.gni:1561: template("instrumentation_test_apk") {
What was the reasoning behind adding this?

[mikecase (-- gone --), 2015-05-18 17:39:03]

https://codereview.chromium.org/1120883002/diff/200001/build/android/gyp/crea...
File build/android/gyp/create_test_runner_script.py (right):

https://codereview.chromium.org/1120883002/diff/200001/build/android/gyp/crea...
build/android/gyp/create_test_runner_script.py:16: script_template = """\
On 2015/05/15 13:19:56, jbudorick wrote:
> nit: should be SCRIPT_TEMPLATE (or _SCRIPT_TEMPLATE)

Done.

https://codereview.chromium.org/1120883002/diff/200001/build/android/gyp/crea...
build/android/gyp/create_test_runner_script.py:32: test_runner_path =
ResolvePath('{test_runner_path}')
On 2015/05/15 13:19:56, jbudorick wrote:
> put an if __name__ == '__main__': guard in here, please.

Done.

https://codereview.chromium.org/1120883002/diff/200001/build/config/android/r...
File build/config/android/rules.gni (right):

https://codereview.chromium.org/1120883002/diff/200001/build/config/android/r...
build/config/android/rules.gni:1561: template("instrumentation_test_apk") {
On 2015/05/15 13:19:57, jbudorick wrote:
> What was the reasoning behind adding this?

There wasn't a clear way to determine which tests were instrumentation tests
since regular apks, gtest apks, and instrumentation test apks all use the
android_apk GN template. Maybe I could do something like
if(defined(apk_under_test)) to determine if the test was an instrumentation
test. But, secondly, if you just build the test apk and try to run the tests the
test runner will fail saying that the forwarder2 and md5sum targets need to be
built. So, similar to what the unittests_apk template does, I added forwarder2
and md5sum targets to this new template's datadep dependencies. I also added the
driver_apk to the datadeps which is needed to run the tests on appurify.

[jbudorick, 2015-05-18 18:20:23]

https://codereview.chromium.org/1120883002/diff/200001/build/config/android/r...
File build/config/android/rules.gni (right):

https://codereview.chromium.org/1120883002/diff/200001/build/config/android/r...
build/config/android/rules.gni:1561: template("instrumentation_test_apk") {
On 2015/05/18 17:39:03, mikecase wrote:
> On 2015/05/15 13:19:57, jbudorick wrote:
> > What was the reasoning behind adding this?
> 
> There wasn't a clear way to determine which tests were instrumentation tests
> since regular apks, gtest apks, and instrumentation test apks all use the
> android_apk GN template. Maybe I could do something like
> if(defined(apk_under_test)) to determine if the test was an instrumentation
> test. But, secondly, if you just build the test apk and try to run the tests
the
> test runner will fail saying that the forwarder2 and md5sum targets need to be
> built. So, similar to what the unittests_apk template does, I added forwarder2
> and md5sum targets to this new template's datadep dependencies. I also added
the
> driver_apk to the datadeps which is needed to run the tests on appurify.

sgtm

https://codereview.chromium.org/1120883002/diff/220001/build/android/gyp/crea...
File build/android/gyp/create_test_runner_script.py (right):

https://codereview.chromium.org/1120883002/diff/220001/build/android/gyp/crea...
build/android/gyp/create_test_runner_script.py:54: 'the action\'s first
output.')
nit: double quotes around strings that contain single quotes.

https://codereview.chromium.org/1120883002/diff/220001/build/android/gyp/crea...
build/android/gyp/create_test_runner_script.py:62: def
RelativizePathToScript(path):
nit: I'm not sure if b/a/gyp/ follows this, but I've informally been saying
elsewhere in b/a/ that local functions should be lower_case_with_underscores.

https://codereview.chromium.org/1120883002/diff/220001/build/android/test_run...
File build/android/test_runner.gypi (right):

https://codereview.chromium.org/1120883002/diff/220001/build/android/test_run...
build/android/test_runner.gypi:53: {
nit: indent this block

[mikecase (-- gone --), 2015-05-18 18:43:46]

https://codereview.chromium.org/1120883002/diff/220001/build/android/gyp/crea...
File build/android/gyp/create_test_runner_script.py (right):

https://codereview.chromium.org/1120883002/diff/220001/build/android/gyp/crea...
build/android/gyp/create_test_runner_script.py:54: 'the action\'s first
output.')
On 2015/05/18 18:20:23, jbudorick wrote:
> nit: double quotes around strings that contain single quotes.

Done.

https://codereview.chromium.org/1120883002/diff/220001/build/android/gyp/crea...
build/android/gyp/create_test_runner_script.py:62: def
RelativizePathToScript(path):
On 2015/05/18 18:20:23, jbudorick wrote:
> nit: I'm not sure if b/a/gyp/ follows this, but I've informally been saying
> elsewhere in b/a/ that local functions should be lower_case_with_underscores.

Acknowledged. The functions in b/a/gyp are all UpperCamelCase so I will keep as
is.

https://codereview.chromium.org/1120883002/diff/220001/build/android/test_run...
File build/android/test_runner.gypi (right):

https://codereview.chromium.org/1120883002/diff/220001/build/android/test_run...
build/android/test_runner.gypi:53: {
On 2015/05/18 18:20:23, jbudorick wrote:
> nit: indent this block

Done.

[cjhopman, 2015-05-20 03:09:27]

lgtm

https://codereview.chromium.org/1120883002/diff/240001/build/config/android/i...
File build/config/android/internal_rules.gni (right):

https://codereview.chromium.org/1120883002/diff/240001/build/config/android/i...
build/config/android/internal_rules.gni:1283: test_name = invoker.test_apk
I feel like this should be based on the target name instead of the apk name...

For chrome_shell_test_apk I think I'd rather have the script be
bin/chrome_shell_test or bin/chrome_shell_test_apk than bin/ChromeShellTest.

This would make instrumentation/gtest more similar.

Don't really feel too strongly, though.

[jbudorick, 2015-05-20 03:10:17]

On 2015/05/20 03:09:27, cjhopman wrote:
> lgtm
> 
>
https://codereview.chromium.org/1120883002/diff/240001/build/config/android/i...
> File build/config/android/internal_rules.gni (right):
> 
>
https://codereview.chromium.org/1120883002/diff/240001/build/config/android/i...
> build/config/android/internal_rules.gni:1283: test_name = invoker.test_apk
> I feel like this should be based on the target name instead of the apk name...
> 
> For chrome_shell_test_apk I think I'd rather have the script be
> bin/chrome_shell_test or bin/chrome_shell_test_apk than bin/ChromeShellTest.
> 

yeah, +1 on this

> This would make instrumentation/gtest more similar.
> 
> Don't really feel too strongly, though.

[mikecase (-- gone --), 2015-05-20 21:22:01]

mikecase@chromium.org changed reviewers:
+ cbentzel@chromium.org, ddorwin@chromium.org, enne@chromium.org,
gbillock@chromium.org, sky@chromium.org, stanisc@chromium.org,
tedchoc@chromium.org, thestig@chromium.org

[mikecase (-- gone --), 2015-05-20 21:22:03]

Needed to change some BUILD.gn files in this CL (mostly to add isolate_files to
some of the test targets so the testrunner in src/build/android/test_runner.py
knows which isolate files to use). Adding...

thestig@ to review base/BUILD.gn, breakpad/BUILD.gn
enne@ to review cc/BUILD.gn
tedchoc@ to review chrome/android/BUILD.gn, content/shell/android/BUILD.gn
sky@ to review chrome/test/BUILD.gn, components/BUILD.gn, ui/base/BUILD.gn
ddorwin@ to review media/BUILD.gn, media/midi/BUILD.gn
cbentzel@ to review net/BUILD.gn
gbillock@ to review sql/BUILD.gn
stanisc@ to review sync/BUILD.gn

https://codereview.chromium.org/1120883002/diff/240001/build/config/android/i...
File build/config/android/internal_rules.gni (right):

https://codereview.chromium.org/1120883002/diff/240001/build/config/android/i...
build/config/android/internal_rules.gni:1283: test_name = invoker.test_apk
On 2015/05/20 03:09:27, cjhopman wrote:
> I feel like this should be based on the target name instead of the apk name...
> 
> For chrome_shell_test_apk I think I'd rather have the script be
> bin/chrome_shell_test or bin/chrome_shell_test_apk than bin/ChromeShellTest.
> 
> This would make instrumentation/gtest more similar.
> 
> Don't really feel too strongly, though.

Done.

[Ted C, 2015-05-20 21:27:19]

On 2015/05/20 21:22:03, mikecase wrote:
> Needed to change some BUILD.gn files in this CL (mostly to add isolate_files
to
> some of the test targets so the testrunner in src/build/android/test_runner.py
> knows which isolate files to use). Adding...
> 
> thestig@ to review base/BUILD.gn, breakpad/BUILD.gn
> enne@ to review cc/BUILD.gn
> tedchoc@ to review chrome/android/BUILD.gn, content/shell/android/BUILD.gn

lgtm

> sky@ to review chrome/test/BUILD.gn, components/BUILD.gn, ui/base/BUILD.gn
> ddorwin@ to review media/BUILD.gn, media/midi/BUILD.gn
> cbentzel@ to review net/BUILD.gn
> gbillock@ to review sql/BUILD.gn
> stanisc@ to review sync/BUILD.gn
> 
>
https://codereview.chromium.org/1120883002/diff/240001/build/config/android/i...
> File build/config/android/internal_rules.gni (right):
> 
>
https://codereview.chromium.org/1120883002/diff/240001/build/config/android/i...
> build/config/android/internal_rules.gni:1283: test_name = invoker.test_apk
> On 2015/05/20 03:09:27, cjhopman wrote:
> > I feel like this should be based on the target name instead of the apk
name...
> > 
> > For chrome_shell_test_apk I think I'd rather have the script be
> > bin/chrome_shell_test or bin/chrome_shell_test_apk than bin/ChromeShellTest.
> > 
> > This would make instrumentation/gtest more similar.
> > 
> > Don't really feel too strongly, though.
> 
> Done.

[cjhopman, 2015-05-20 21:30:45]

lgtm

[Lei Zhang, 2015-05-20 21:33:11]

https://codereview.chromium.org/1120883002/diff/260001/breakpad/BUILD.gn
File breakpad/BUILD.gn (right):

https://codereview.chromium.org/1120883002/diff/260001/breakpad/BUILD.gn#newc...
breakpad/BUILD.gn:621: isolate_file = "breakpad_unittests.isolate"
Should there be a corresponding change in breakpad.gyp?

[stanisc, 2015-05-21 01:21:59]

sync lgtm

[ddorwin, 2015-05-21 02:55:49]

https://codereview.chromium.org/1120883002/diff/260001/media/BUILD.gn
File media/BUILD.gn (right):

https://codereview.chromium.org/1120883002/diff/260001/media/BUILD.gn#newcode568
media/BUILD.gn:568: isolate_file = "media_unittests.isolate"
This applies to all platforms unlike the GYP change and the change below.

Shouldn't we have isolate files for all platforms? (If not, this is different
from the others.)

https://codereview.chromium.org/1120883002/diff/260001/media/midi/BUILD.gn
File media/midi/BUILD.gn (right):

https://codereview.chromium.org/1120883002/diff/260001/media/midi/BUILD.gn#ne...
media/midi/BUILD.gn:159: isolate_file = "midi_unittests.isolate"
The (existing) GYP file does not have this for Android and has a
'test_isolation_mode != "noop"' condition for other platforms. Should these be
equivalent?

[cbentzel, 2015-05-21 16:35:59]

I'll take a look at the net/ change, but one meta-question:

Is there any chance you could provide more detail and an associated bug with
this CL?

The description lets me know that compile-time scripts are being generated for
running tests, but it would be great if it went into a bit more detail about
which tests scripts are being generated for, potentially some of the motivation
for this, etc.

Thanks!

[mikecase (-- gone --), 2015-05-21 18:46:46]

Added a bug containing some details/motivation for creating these generated
scripts.
https://code.google.com/p/chromium/issues/detail?id=490781

https://codereview.chromium.org/1120883002/diff/260001/breakpad/BUILD.gn
File breakpad/BUILD.gn (right):

https://codereview.chromium.org/1120883002/diff/260001/breakpad/BUILD.gn#newc...
breakpad/BUILD.gn:621: isolate_file = "breakpad_unittests.isolate"
On 2015/05/20 at 21:33:11, Lei Zhang wrote:
> Should there be a corresponding change in breakpad.gyp?

Added. The way I currently have it setup, GYP only generates scripts to run
gtests that are wrapped in apks. This is because gtests that are wrapped in apks
all include a common gypi file (test_apk.gypi) which I was able to use to
generate the scripts. To generate scripts for gtests not wrapped in apks you
have to add a little more to their build files.

For example, the following will generate a script to run breakpad_unittests. I
added this to breakpad.gyp.

'conditions': [
  ['OS=="android"', {
    'variables': {
      'test_type': 'gtest',
      'isolate_file': ''breakpad_unittests.isolate',
    }
    'includes': ['build/android/test_runner.gypi'],
  }]
]

https://codereview.chromium.org/1120883002/diff/260001/media/BUILD.gn
File media/BUILD.gn (right):

https://codereview.chromium.org/1120883002/diff/260001/media/BUILD.gn#newcode568
media/BUILD.gn:568: isolate_file = "media_unittests.isolate"
On 2015/05/21 at 02:55:49, ddorwin wrote:
> This applies to all platforms unlike the GYP change and the change below.
> 
> Shouldn't we have isolate files for all platforms? (If not, this is different
from the others.)

Thanks for the catch. Since this variable wouldn't be used on anything but
Android (at least initially), GN will give an error when building on other
platforms (I got "ERROR Assignment had no effect." when trying to build on linux
just now). Moving into the if(is_android) block.

https://codereview.chromium.org/1120883002/diff/260001/media/midi/BUILD.gn
File media/midi/BUILD.gn (right):

https://codereview.chromium.org/1120883002/diff/260001/media/midi/BUILD.gn#ne...
media/midi/BUILD.gn:159: isolate_file = "midi_unittests.isolate"
On 2015/05/21 at 02:55:49, ddorwin wrote:
> The (existing) GYP file does not have this for Android and has a
'test_isolation_mode != "noop"' condition for other platforms. Should these be
equivalent?

No, our Android test runner uses isolate files to determine which files it needs
to push to the device (isolate_file is am argument for our testrunner). Other
platforms use the isolate files to actually run their tests which they use the
test_isolation_mode for.

[Lei Zhang, 2015-05-21 19:42:33]

base/ and breakpad/ lgtm

[cbentzel, 2015-05-21 19:55:14]

net/ LGTM

Thanks for adding the bug!

[cbentzel, 2015-05-21 19:55:16]

net/ LGTM

Thanks for adding the bug!

[ddorwin, 2015-05-21 20:21:36]

media/ LGTM with question

https://codereview.chromium.org/1120883002/diff/280001/media/midi/BUILD.gn
File media/midi/BUILD.gn (right):

https://codereview.chromium.org/1120883002/diff/280001/media/midi/BUILD.gn#ne...
media/midi/BUILD.gn:159: isolate_file = "midi_unittests.isolate"
Do you need to add this to media/midi/midi.gyp too?

[mikecase (-- gone --), 2015-05-22 17:17:11]

https://codereview.chromium.org/1120883002/diff/280001/media/midi/BUILD.gn
File media/midi/BUILD.gn (right):

https://codereview.chromium.org/1120883002/diff/280001/media/midi/BUILD.gn#ne...
media/midi/BUILD.gn:159: isolate_file = "midi_unittests.isolate"
On 2015/05/21 at 20:21:36, ddorwin wrote:
> Do you need to add this to media/midi/midi.gyp too?

Yep, done. Thanks for the catch.

[ddorwin, 2015-05-22 18:17:41]

lgtm

[jbudorick, 2015-05-28 04:12:23]

build/android/ lgtm

[mikecase (-- gone --), 2015-06-01 17:18:58]

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

[mikecase (-- gone --), 2015-06-01 17:18:59]

Just waiting on reviews from...

enne@ for cc/BUILD.gn
sky@ for ui/base/BUILD.gn
gbillock@ for sql/BUILD.gn

I missed an owner for components/BUILD.gn earlier, so adding erikwright@ to
review components/BUILD.gn

Please take a look at this CL when you get the chance. Thanks.

[enne (OOO), 2015-06-01 17:24:21]

cc lgtm

[sky, 2015-06-01 18:03:42]

LGTM

[mikecase (-- gone --), 2015-06-01 19:35:29]

mikecase@chromium.org changed reviewers:
- gbillock@chromium.org

[mikecase (-- gone --), 2015-06-01 19:35:30]

Looks like gbillock@ no longer works on Chrome, so erikwright could you look at
sql/BUILD.gn as well as components/BUILD.gn. Thanks

[erikwright (departed), 2015-06-02 18:26:39]

I don't really understand how isolate files work, but if someone who does has
looked at this the changes to components/ and sql GYP/GN files LGTM.

[mikecase (-- gone --), 2015-06-02 20:16:20]

The CQ bit was checked by mikecase@chromium.org

[mikecase (-- gone --), 2015-06-02 20:16:21]

The patchset sent to the CQ was uploaded after l-g-t-m from
stanisc@chromium.org, tedchoc@chromium.org, cjhopman@chromium.org,
cbentzel@chromium.org, thestig@chromium.org, sky@chromium.org,
erikwright@chromium.org, jbudorick@chromium.org, enne@chromium.org,
ddorwin@chromium.org
Link to the patchset: https://codereview.chromium.org/1120883002/#ps320001
(title: "Rebase.")

[commit-bot: I haz the power, 2015-06-02 20:17:05]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1120883002/320001

[commit-bot: I haz the power, 2015-06-02 20:33:38]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-06-02 20:33:40]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[mikecase (-- gone --), 2015-06-02 22:45:42]

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

[mikecase (-- gone --), 2015-06-02 22:45:44]

Adding boliu@chromium.org to review android_webview/android_webview_tests.gypi.

[boliu, 2015-06-02 22:52:27]

On 2015/06/02 22:45:44, mikecase wrote:
> Adding mailto:boliu@chromium.org to review
android_webview/android_webview_tests.gypi.

rs lgtm

[mikecase (-- gone --), 2015-06-02 22:53:14]

The CQ bit was checked by mikecase@chromium.org

[mikecase (-- gone --), 2015-06-02 22:53:14]

The patchset sent to the CQ was uploaded after l-g-t-m from
cbentzel@chromium.org, stanisc@chromium.org, sky@chromium.org,
thestig@chromium.org, erikwright@chromium.org, jbudorick@chromium.org,
enne@chromium.org, tedchoc@chromium.org, cjhopman@chromium.org,
ddorwin@chromium.org
Link to the patchset: https://codereview.chromium.org/1120883002/#ps340001
(title: "Renamed datadeps to data_deps to reflect recent change.")

[commit-bot: I haz the power, 2015-06-02 22:54:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1120883002/340001

[commit-bot: I haz the power, 2015-06-02 23:35:30]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-06-02 23:35:33]

Try jobs failed on following builders:
  linux_chromium_asan_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[mikecase (-- gone --), 2015-06-03 00:01:58]

The CQ bit was checked by mikecase@chromium.org

[commit-bot: I haz the power, 2015-06-03 00:03:50]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1120883002/340001

[commit-bot: I haz the power, 2015-06-03 00:57:35]

Committed patchset #18 (id:340001)

[commit-bot: I haz the power, 2015-06-03 00:58:21]

Patchset 18 (id:??) landed as
https://crrev.com/56d80d7a4c1ff50bf21112486cc9b2308c7e7bf0
Cr-Commit-Position: refs/heads/master@{#332518}