Use OAuth2AccessTokenFetcher implementation on Android.

Use OAuth2AccessTokenFetcher implementation on Android.  This cleans up the way that AO2TS overrides functionality from the base class to differentiate the way that access tokens are minted, and makes the overriding mechanism for android the same as for iOS.  Its possible to reimplement OAuth2AccessTokenFetcher now that Mihai extended for use in bling.

BUG=390353
Committed: https://crrev.com/69d3373d602b2ccbc19e6c5f4b87c802c02f0116
Cr-Commit-Position: refs/heads/master@{#292205}

[Roger Tawa OOO till Jul 10th, 2014-06-30 20:55:14]

Hi Alan,

While investigating different things in AO2TS, I noticed that the way it
override O2TS behaviour is different from iOS.  I changed the code to match iOS
more closely.

This should be a noop, however it will allow sharing of requests to the android
OS for various access tokens requests with the same scope.  Basically the same
sharing as done on desktop and iOS today.

[acleung1, 2014-06-30 22:53:25]

I don't really follow on this fixes 385828. Can you elaborate?

https://codereview.chromium.org/345703003/diff/80001/chrome/browser/signin/an...
File chrome/browser/signin/android_profile_oauth2_token_service.cc (right):

https://codereview.chromium.org/345703003/diff/80001/chrome/browser/signin/an...
chrome/browser/signin/android_profile_oauth2_token_service.cc:27: std::string
CombineScopes(const std::vector<std::string>& scopes) {
Since we no longer need this in the outer namespace, we can limit this
function's visibility by moving it inside of AndroidAccessTokenFetcher instead.

https://codereview.chromium.org/345703003/diff/80001/chrome/browser/signin/an...
chrome/browser/signin/android_profile_oauth2_token_service.cc:60: virtual void
CancelRequest() OVERRIDE;
Why is this called? When the profile went away before the request completed? I
couldn't find a caller in a code search.

[Roger Tawa OOO till Jul 10th, 2014-07-01 01:38:44]

I used 385828 since I found this issue while investigating there, but agree does
not make sense to lump this change under that bug.  Opened a new bug.

https://codereview.chromium.org/345703003/diff/80001/chrome/browser/signin/an...
File chrome/browser/signin/android_profile_oauth2_token_service.cc (right):

https://codereview.chromium.org/345703003/diff/80001/chrome/browser/signin/an...
chrome/browser/signin/android_profile_oauth2_token_service.cc:27: std::string
CombineScopes(const std::vector<std::string>& scopes) {
On 2014/06/30 22:53:25, acleung1 wrote:
> Since we no longer need this in the outer namespace, we can limit this
> function's visibility by moving it inside of AndroidAccessTokenFetcher
instead.

Done.

https://codereview.chromium.org/345703003/diff/80001/chrome/browser/signin/an...
chrome/browser/signin/android_profile_oauth2_token_service.cc:60: virtual void
CancelRequest() OVERRIDE;
On 2014/06/30 22:53:25, acleung1 wrote:
> Why is this called? When the profile went away before the request completed? I
> couldn't find a caller in a code search.

This is the implementation of a pure virtual method from the base class.  Having
said that, O2TS does not seem to call this method.  Maybe it needs to be called
from OAuth2TokenService::Fetcher::Cancel().  I'll take a look.

[msarda, 2014-07-01 11:36:48]

LGTM

https://codereview.chromium.org/345703003/diff/100001/chrome/browser/signin/a...
File chrome/browser/signin/android_profile_oauth2_token_service.cc (right):

https://codereview.chromium.org/345703003/diff/100001/chrome/browser/signin/a...
chrome/browser/signin/android_profile_oauth2_token_service.cc:214:
DCHECK(content::BrowserThread::CurrentlyOn(content::BrowserThread::UI));
I do not see the rest of the methods DCHECKing the fact that they are called on
the UI thread. Consider removing this DCHECK.

[acleung1, 2014-07-01 20:27:25]

lgtm

[msarda, 2014-07-04 13:27:49]

lgtm

[Roger Tawa OOO till Jul 10th, 2014-08-27 15:35:42]

The CQ bit was checked by rogerta@chromium.org

[commit-bot: I haz the power, 2014-08-27 15:36:30]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/rogerta@chromium.org/345703003/320001

[commit-bot: I haz the power, 2014-08-27 16:55:28]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  mac_gpu on tryserver.chromium.gpu
(http://build.chromium.org/p/tryserver.chromium.gpu/builders/mac_gpu/builds/47025)
  win_chromium_compile_dbg on tryserver.chromium.win
(http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_comp...)

[commit-bot: I haz the power, 2014-08-27 17:15:12]

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

[commit-bot: I haz the power, 2014-08-27 17:15:13]

Try jobs failed on following builders:
  mac_gpu_retina_triggered_tests on tryserver.chromium.gpu
(http://build.chromium.org/p/tryserver.chromium.gpu/builders/mac_gpu_retina_tr...)

[Roger Tawa OOO till Jul 10th, 2014-08-27 18:55:59]

The CQ bit was checked by rogerta@chromium.org

[commit-bot: I haz the power, 2014-08-27 18:57:37]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/rogerta@chromium.org/345703003/320001

[commit-bot: I haz the power, 2014-08-27 19:30:42]

Committed patchset #11 (id:320001) as f70a1c04442448fd6fa4cfd982c6dd7119b36515

[commit-bot: I haz the power, 2014-09-10 02:53:26]

Patchset 11 (id:??) landed as
https://crrev.com/69d3373d602b2ccbc19e6c5f4b87c802c02f0116
Cr-Commit-Position: refs/heads/master@{#292205}