Use WeakPtr and do not use pointer equality in WebURLLoaderMockFactory

Use WeakPtr and do not use pointer equality in WebURLLoaderMockFactory

WebURLLoaderMockFactory determines whether WebURLLoaderMock is still pending
using WebURLLoaderMock's raw pointer value, but this is not correct when:
1. WebURLLoaderMock A is destructed and removed from |pending_loaders_|,
2. and then a new WebURLLoaderMock B is created at the same address as
   WebURLLoaderMock A, and added to |pending_loaders_|.
In this case, IsPending() returns true for WebURLLoaderMock A, but what is
actually pending is WebURLLoaderMock B.
This actually occurs when testing CORS redirects by
https://codereview.chromium.org/1257343003/ on mac.

This CL removes checks by WebURLLoaderMock pointer equality, and instead uses
WeakPtr and |WebURLLoaderMock::client_| to determine whether WebURLLoaderMock
is destructed or cancelled.

BUG=532364
Committed: https://crrev.com/30ab97a4b8acb0fa9752956e68c68a3d7d75ee81
Cr-Commit-Position: refs/heads/master@{#379152}

[hiroshige, 2016-02-26 23:05:20]

The CQ bit was checked by hiroshige@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-02-26 23:09:58]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1743783002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1743783002/1

[hiroshige, 2016-02-26 23:12:49]

Description was changed from

==========
Use WeakPtr and do not use pointer equality in WebURLLoaderMock

BUG=
==========

to

==========
Use WeakPtr and do not use pointer equality in WebURLLoaderMockFactory

WebURLLoaderMockFactory determines whether WebURLLoaderMock is still pending
using WebURLLoaderMock's raw pointer value, but this is not correct when:
1. WebURLLoaderMock A is destructed and removed from |pending_loaders_|,
2. and then a new WebURLLoaderMock B is created at the same address as
   WebURLLoaderMock A, and added to |pending_loaders_|.
In this case, IsPending() returns true for WebURLLoaderMock A.
This actually occurs when testing CORS redirects by
https://codereview.chromium.org/1257343003/ on mac.

This CL removes checks by WebURLLoaderMock pointer equality, and instead uses
WeakPtr and |WebURLLoaderMock::client_| to determine whether WebURLLoaderMock
is destructed or cancelled.

BUG=532364
==========

[hiroshige, 2016-02-26 23:20:48]

Description was changed from

==========
Use WeakPtr and do not use pointer equality in WebURLLoaderMockFactory

WebURLLoaderMockFactory determines whether WebURLLoaderMock is still pending
using WebURLLoaderMock's raw pointer value, but this is not correct when:
1. WebURLLoaderMock A is destructed and removed from |pending_loaders_|,
2. and then a new WebURLLoaderMock B is created at the same address as
   WebURLLoaderMock A, and added to |pending_loaders_|.
In this case, IsPending() returns true for WebURLLoaderMock A.
This actually occurs when testing CORS redirects by
https://codereview.chromium.org/1257343003/ on mac.

This CL removes checks by WebURLLoaderMock pointer equality, and instead uses
WeakPtr and |WebURLLoaderMock::client_| to determine whether WebURLLoaderMock
is destructed or cancelled.

BUG=532364
==========

to

==========
Use WeakPtr and do not use pointer equality in WebURLLoaderMockFactory

WebURLLoaderMockFactory determines whether WebURLLoaderMock is still pending
using WebURLLoaderMock's raw pointer value, but this is not correct when:
1. WebURLLoaderMock A is destructed and removed from |pending_loaders_|,
2. and then a new WebURLLoaderMock B is created at the same address as
   WebURLLoaderMock A, and added to |pending_loaders_|.
In this case, IsPending() returns true for WebURLLoaderMock A, but what is
actually pending is WebURLLoaderMock B.
This actually occurs when testing CORS redirects by
https://codereview.chromium.org/1257343003/ on mac.

This CL removes checks by WebURLLoaderMock pointer equality, and instead uses
WeakPtr and |WebURLLoaderMock::client_| to determine whether WebURLLoaderMock
is destructed or cancelled.

BUG=532364
==========

[hiroshige, 2016-02-26 23:23:54]

hiroshige@chromium.org changed reviewers:
+ jam@chromium.org

[hiroshige, 2016-02-29 16:33:35]

The CQ bit was checked by hiroshige@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-02-29 16:33:51]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1743783002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1743783002/20001

[hiroshige, 2016-02-29 16:35:01]

PTAL.

[jam, 2016-02-29 16:41:09]

lgtm

https://codereview.chromium.org/1743783002/diff/20001/content/test/weburl_loa...
File content/test/weburl_loader_mock.h (right):

https://codereview.chromium.org/1743783002/diff/20001/content/test/weburl_loa...
content/test/weburl_loader_mock.h:57: bool isCancelled() { return !client_; }
nit: these two should be is_deferred() and is_cancelled() per google style guide

[hiroshige, 2016-02-29 18:12:03]

The CQ bit was checked by hiroshige@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-02-29 18:12:36]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1743783002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1743783002/40001

[commit-bot: I haz the power, 2016-02-29 18:22:48]

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

[commit-bot: I haz the power, 2016-02-29 18:22:49]

Dry run: Try jobs failed on following builders:
  win_chromium_compile_dbg_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_comp...)
  win_chromium_x64_rel_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_x64_...)

[hiroshige, 2016-02-29 19:47:30]

The CQ bit was checked by hiroshige@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-02-29 19:48:34]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1743783002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1743783002/40001

[commit-bot: I haz the power, 2016-02-29 21:04:45]

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

[commit-bot: I haz the power, 2016-02-29 21:04:45]

Dry run: This issue passed the CQ dry run.

[hiroshige, 2016-02-29 21:21:55]

https://codereview.chromium.org/1743783002/diff/20001/content/test/weburl_loa...
File content/test/weburl_loader_mock.h (right):

https://codereview.chromium.org/1743783002/diff/20001/content/test/weburl_loa...
content/test/weburl_loader_mock.h:57: bool isCancelled() { return !client_; }
On 2016/02/29 16:41:09, jam wrote:
> nit: these two should be is_deferred() and is_cancelled() per google style
guide

Done.

[hiroshige, 2016-03-03 22:42:27]

The CQ bit was checked by hiroshige@chromium.org

[hiroshige, 2016-03-03 22:42:27]

The patchset sent to the CQ was uploaded after l-g-t-m from jam@chromium.org
Link to the patchset: https://codereview.chromium.org/1743783002/#ps60001
(title: "Rebase.")

[commit-bot: I haz the power, 2016-03-03 22:42:45]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1743783002/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1743783002/60001

[commit-bot: I haz the power, 2016-03-04 00:27:54]

Description was changed from

==========
Use WeakPtr and do not use pointer equality in WebURLLoaderMockFactory

WebURLLoaderMockFactory determines whether WebURLLoaderMock is still pending
using WebURLLoaderMock's raw pointer value, but this is not correct when:
1. WebURLLoaderMock A is destructed and removed from |pending_loaders_|,
2. and then a new WebURLLoaderMock B is created at the same address as
   WebURLLoaderMock A, and added to |pending_loaders_|.
In this case, IsPending() returns true for WebURLLoaderMock A, but what is
actually pending is WebURLLoaderMock B.
This actually occurs when testing CORS redirects by
https://codereview.chromium.org/1257343003/ on mac.

This CL removes checks by WebURLLoaderMock pointer equality, and instead uses
WeakPtr and |WebURLLoaderMock::client_| to determine whether WebURLLoaderMock
is destructed or cancelled.

BUG=532364
==========

to

==========
Use WeakPtr and do not use pointer equality in WebURLLoaderMockFactory

WebURLLoaderMockFactory determines whether WebURLLoaderMock is still pending
using WebURLLoaderMock's raw pointer value, but this is not correct when:
1. WebURLLoaderMock A is destructed and removed from |pending_loaders_|,
2. and then a new WebURLLoaderMock B is created at the same address as
   WebURLLoaderMock A, and added to |pending_loaders_|.
In this case, IsPending() returns true for WebURLLoaderMock A, but what is
actually pending is WebURLLoaderMock B.
This actually occurs when testing CORS redirects by
https://codereview.chromium.org/1257343003/ on mac.

This CL removes checks by WebURLLoaderMock pointer equality, and instead uses
WeakPtr and |WebURLLoaderMock::client_| to determine whether WebURLLoaderMock
is destructed or cancelled.

BUG=532364
==========

[commit-bot: I haz the power, 2016-03-04 00:27:55]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2016-03-04 00:29:05]

Description was changed from

==========
Use WeakPtr and do not use pointer equality in WebURLLoaderMockFactory

WebURLLoaderMockFactory determines whether WebURLLoaderMock is still pending
using WebURLLoaderMock's raw pointer value, but this is not correct when:
1. WebURLLoaderMock A is destructed and removed from |pending_loaders_|,
2. and then a new WebURLLoaderMock B is created at the same address as
   WebURLLoaderMock A, and added to |pending_loaders_|.
In this case, IsPending() returns true for WebURLLoaderMock A, but what is
actually pending is WebURLLoaderMock B.
This actually occurs when testing CORS redirects by
https://codereview.chromium.org/1257343003/ on mac.

This CL removes checks by WebURLLoaderMock pointer equality, and instead uses
WeakPtr and |WebURLLoaderMock::client_| to determine whether WebURLLoaderMock
is destructed or cancelled.

BUG=532364
==========

to

==========
Use WeakPtr and do not use pointer equality in WebURLLoaderMockFactory

WebURLLoaderMockFactory determines whether WebURLLoaderMock is still pending
using WebURLLoaderMock's raw pointer value, but this is not correct when:
1. WebURLLoaderMock A is destructed and removed from |pending_loaders_|,
2. and then a new WebURLLoaderMock B is created at the same address as
   WebURLLoaderMock A, and added to |pending_loaders_|.
In this case, IsPending() returns true for WebURLLoaderMock A, but what is
actually pending is WebURLLoaderMock B.
This actually occurs when testing CORS redirects by
https://codereview.chromium.org/1257343003/ on mac.

This CL removes checks by WebURLLoaderMock pointer equality, and instead uses
WeakPtr and |WebURLLoaderMock::client_| to determine whether WebURLLoaderMock
is destructed or cancelled.

BUG=532364

Committed: https://crrev.com/30ab97a4b8acb0fa9752956e68c68a3d7d75ee81
Cr-Commit-Position: refs/heads/master@{#379152}
==========

[commit-bot: I haz the power, 2016-03-04 00:29:06]

Patchset 4 (id:??) landed as
https://crrev.com/30ab97a4b8acb0fa9752956e68c68a3d7d75ee81
Cr-Commit-Position: refs/heads/master@{#379152}