Base: Make FileProxy automaticaly close the file on a worker thread.

Base: Make FileProxy automaticaly close the file on a worker thread.

This CL removes the restriction that callers should call Close before
deleting the object if they want to make sure the file is not closed
on the current thread.

BUG=322664
TEST=base_unittests

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=263675

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=268348

[rvargas (doing something else), 2014-04-10 17:37:36]

PTAL

[willchan no longer on Chromium, 2014-04-10 20:27:47]

https://codereview.chromium.org/231703002/diff/1/base/files/file_proxy.cc
File base/files/file_proxy.cc (right):

https://codereview.chromium.org/231703002/diff/1/base/files/file_proxy.cc#new...
base/files/file_proxy.cc:43: scoped_refptr<TaskRunner> task_runner_;
Can this be private? Does it need to be protected?

https://codereview.chromium.org/231703002/diff/1/base/files/file_proxy.h
File base/files/file_proxy.h (right):

https://codereview.chromium.org/231703002/diff/1/base/files/file_proxy.h#newc...
base/files/file_proxy.h:131: TaskRunner* task_runner() { return
task_runner_.get(); }
I don't really care, but why not just access the task_runner_ directly?

[rvargas (doing something else), 2014-04-10 22:03:51]

Thanks

https://codereview.chromium.org/231703002/diff/1/base/files/file_proxy.cc
File base/files/file_proxy.cc (right):

https://codereview.chromium.org/231703002/diff/1/base/files/file_proxy.cc#new...
base/files/file_proxy.cc:43: scoped_refptr<TaskRunner> task_runner_;
On 2014/04/10 20:27:47, willchan wrote:
> Can this be private? Does it need to be protected?

yes. I made it private (along with proxy_)

https://codereview.chromium.org/231703002/diff/1/base/files/file_proxy.h
File base/files/file_proxy.h (right):

https://codereview.chromium.org/231703002/diff/1/base/files/file_proxy.h#newc...
base/files/file_proxy.h:131: TaskRunner* task_runner() { return
task_runner_.get(); }
On 2014/04/10 20:27:47, willchan wrote:
> I don't really care, but why not just access the task_runner_ directly?

I think it's a little cleaner if friends' access goes through methods.

[willchan no longer on Chromium, 2014-04-10 23:07:38]

lgtm

[rvargas (doing something else), 2014-04-14 14:33:16]

The CQ bit was checked by rvargas@chromium.org

[commit-bot: I haz the power, 2014-04-14 14:33:21]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rvargas@chromium.org/231703002/40001

[commit-bot: I haz the power, 2014-04-14 16:24:29]

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

[commit-bot: I haz the power, 2014-04-14 16:24:29]

Try jobs failed on following builders:
  tryserver.chromium on win_chromium_rel

[rvargas (doing something else), 2014-04-14 16:47:05]

The CQ bit was checked by rvargas@chromium.org

[commit-bot: I haz the power, 2014-04-14 16:47:26]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/rvargas@chromium.org/231703002/40001

[commit-bot: I haz the power, 2014-04-14 18:32:49]

Change committed as 263675

[leviw_travelin_and_unemployed, 2014-04-15 00:42:46]

A revert of this CL has been created in
https://codereview.chromium.org/238383003/ by leviw@chromium.org.

The reason for reverting is: Triggering AssertIOAllowed on a handful of Blink
LayoutTests..

[rvargas (doing something else), 2014-05-05 22:13:29]

On 2014/04/15 00:42:46, leviw wrote:
> A revert of this CL has been created in
> https://codereview.chromium.org/238383003/ by mailto:leviw@chromium.org.
> 
> The reason for reverting is: Triggering AssertIOAllowed on a handful of Blink
> LayoutTests..

The cause of that failure should be fixed by r267977.

[rvargas (doing something else), 2014-05-05 22:13:36]

The CQ bit was checked by rvargas@chromium.org

[commit-bot: I haz the power, 2014-05-05 22:15:10]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/rvargas@chromium.org/231703002/40001

[commit-bot: I haz the power, 2014-05-06 00:55:39]

Change committed as 268348