Enable file path watcher for Android

Enable file path watcher for Android

Use the linux implementation for file path watcher because
Android supports 'inotify'. This fixes the failed case
UserStyleSheetWatcherTest.StyleLoad.

BUG=139412
TEST=run_tests.py -s unit_tests


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=154387

[yongsheng, 2012-08-27 12:40:39]

please help review it. Thanks.

[yongsheng, 2012-08-27 12:49:46]

On 2012/08/27 12:40:39, yongsheng wrote:
> please help review it. Thanks.

The fix for file_path_watcher_linux.cc is because of the below error on Android:
base/files/file_path_watcher_linux.cc:376:77: error: lvalue required as
increment operand

comps.begin() return a const result, which is not a lvalue.

[Mattias Nissler (ping if slow), 2012-08-27 12:53:02]

http://codereview.chromium.org/10874083/diff/1/base/base.gypi
File base/base.gypi (right):

http://codereview.chromium.org/10874083/diff/1/base/base.gypi#newcode586
base/base.gypi:586: ['include', '^files/file_path_watcher_linux\\.cc$'],
alphabetize

http://codereview.chromium.org/10874083/diff/1/base/files/file_path_watcher_l...
File base/files/file_path_watcher_linux.cc (right):

http://codereview.chromium.org/10874083/diff/1/base/files/file_path_watcher_l...
base/files/file_path_watcher_linux.cc:377: for (comp++; comp != comps.end();
comp++) {
Style guide says to use preincrement:
http://google-styleguide.googlecode.com/svn/trunk/cppguide.xml?showone=Preinc...

Also, you no longer need braces, as the loop body is only a single line now.

[yongsheng, 2012-08-28 03:09:18]

thanks for your review, Mattias.
Reviewers, any comments?

[nilesh, 2012-08-28 03:36:35]

LGTM

[Mattias Nissler (ping if slow), 2012-08-28 11:06:14]

LGTM

[yongsheng, 2012-08-29 01:29:03]

On 2012/08/28 11:06:14, Mattias Nissler wrote:
> LGTM
thanks.

[commit-bot: I haz the power, 2012-08-29 01:31:21]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/yongsheng.zhu@intel.com/10874083/5001

[commit-bot: I haz the power, 2012-08-29 01:31:25]

Presubmit check for 10874083-5001 failed and returned exit status 1.


Running presubmit commit checks ...

** Presubmit ERRORS **
Missing LGTM from an OWNER for files in these directories:
    base

Presubmit checks took 2.0s to calculate.

[yongsheng, 2012-08-29 01:33:35]

hi, brettw, need your review. Thanks.

[brettw, 2012-08-29 06:53:34]

overloaded, please find another reviewer

[yongsheng, 2012-08-29 06:55:10]

On 2012/08/29 06:53:34, brettw wrote:
> overloaded, please find another reviewer
hi, Jim, could you please help review it? Need the owner of 'base'

[yongsheng, 2012-08-30 01:20:10]

mark, could you please help review it? thanks.

[jar (doing other things), 2012-08-30 02:47:25]

I had a question below.... but the change LGTM.

http://codereview.chromium.org/10874083/diff/5001/base/files/file_path_watche...
File base/files/file_path_watcher_linux.cc (left):

http://codereview.chromium.org/10874083/diff/5001/base/files/file_path_watche...
base/files/file_path_watcher_linux.cc:376: for
(std::vector<FilePath::StringType>::const_iterator comp(++comps.begin());
I didn't especially like this pre-increment on the return from begin(), but it
wasn't clear why the new code was different.

Can you point out the change here?  ...was it just to make a compiler happy??

[Mark Mentovai, 2012-08-30 02:50:41]

http://codereview.chromium.org/10874083/diff/5001/base/files/file_path_watche...
File base/files/file_path_watcher_linux.cc (right):

http://codereview.chromium.org/10874083/diff/5001/base/files/file_path_watche...
base/files/file_path_watcher_linux.cc:376:
std::vector<FilePath::StringType>::const_iterator comp = comps.begin();
Why are you changing this?

[yongsheng, 2012-08-30 03:06:30]

thanks for your comments.

http://codereview.chromium.org/10874083/diff/5001/base/files/file_path_watche...
File base/files/file_path_watcher_linux.cc (right):

http://codereview.chromium.org/10874083/diff/5001/base/files/file_path_watche...
base/files/file_path_watcher_linux.cc:376:
std::vector<FilePath::StringType>::const_iterator comp = comps.begin();
The fix for file_path_watcher_linux.cc is because of the below error on Android:
base/files/file_path_watcher_linux.cc:376:77: error: lvalue required as
increment operand.
comps.begin() returns a rvalue.

[yongsheng, 2012-08-30 04:50:16]

I'll commit it if no other issues.

http://codereview.chromium.org/10874083/diff/5001/base/files/file_path_watche...
File base/files/file_path_watcher_linux.cc (left):

http://codereview.chromium.org/10874083/diff/5001/base/files/file_path_watche...
base/files/file_path_watcher_linux.cc:376: for
(std::vector<FilePath::StringType>::const_iterator comp(++comps.begin());
On 2012/08/30 02:47:25, jar wrote:
> I didn't especially like this pre-increment on the return from begin(), but it
> wasn't clear why the new code was different.
> 
> Can you point out the change here?  ...was it just to make a compiler happy??
yes, see the compiler error on Android.

[Mark Mentovai, 2012-08-30 12:24:57]

LGTM

[commit-bot: I haz the power, 2012-08-31 01:34:42]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/yongsheng.zhu@intel.com/10874083/5001

[commit-bot: I haz the power, 2012-08-31 05:27:17]

Change committed as 154387