Remove implicit conversions from scoped_refptr to T* in ui/

Remove implicit conversions from scoped_refptr to T* in ui/

This patch was generated by running the rewrite_scoped_refptr clang tool
on a Linux build.

BUG=110610
Committed: https://crrev.com/0efb94f2fd2e0ebc9ea572697e935d1021d0a5e3
Cr-Commit-Position: refs/heads/master@{#291764}

[dcheng, 2014-08-25 18:48:35]

dcheng@chromium.org changed reviewers:
+ danakj@chromium.org, sky@chromium.org

[dcheng, 2014-08-25 18:48:35]

+sky for primary OWNERS review
+danakj for a question about in_process_context_factory.cc:
The tool prefers to rewrite the return type of a function if it returns a
scoped_refptr<T> temporary. However,
InProcessContextFactory::GetCompositorMessageLoop() currently converts a
scoped_refptr<T> to a T* in the return. Since this is an overridden virtual,
that breaks the build. I've manually edited the generated patch to add an
explicit conversion to scoped_refptr<T>--please let me know if you'd like me to
change the return type of GetCompositorMessageLoop() instead.

[danakj, 2014-08-25 18:50:48]

https://codereview.chromium.org/496313004/diff/1/ui/compositor/test/in_proces...
File ui/compositor/test/in_process_context_factory.cc (right):

https://codereview.chromium.org/496313004/diff/1/ui/compositor/test/in_proces...
ui/compositor/test/in_process_context_factory.cc:102: return
compositor_thread_->message_loop_proxy().get();
If I understood your question right, I agree this is the right thing to do, we
should always .get() not implicitly cast to T*

[sky, 2014-08-25 19:47:31]

Doesn't scoped_ptr do some magic to allow it to be used in boolean expressions
but not implicitly converted to the underyling pointer? Could we do the same
here?

[danakj, 2014-08-25 19:49:06]

On 2014/08/25 19:47:31, sky wrote:
> Doesn't scoped_ptr do some magic to allow it to be used in boolean expressions
> but not implicitly converted to the underyling pointer? Could we do the same
> here?

Yes, but not until they remove the implicit T*. I believe they are planning to
go back and remove all the .get()s that are just for boolean cases after it is
removed. At least I read so in the bug for this removal.

[dcheng, 2014-08-25 19:49:22]

On 2014/08/25 19:47:31, sky wrote:
> Doesn't scoped_ptr do some magic to allow it to be used in boolean expressions
> but not implicitly converted to the underyling pointer? Could we do the same
> here?

I'm planning on adding that logic, but we can't do that until all the implicit
conversions are removed and disabled on a platform, since a bool conversion
would otherwise be ambiguous. I am currently planning a followup pass to rewrite
unnecessary .get() calls in boolean tests once the conversion operator is
disabled on each platform.

[danakj, 2014-08-25 19:56:07]

On Mon, Aug 25, 2014 at 3:49 PM, <dcheng@chromium.org> wrote:

> On 2014/08/25 19:47:31, sky wrote:
>
>> Doesn't scoped_ptr do some magic to allow it to be used in boolean
>> expressions
>> but not implicitly converted to the underyling pointer? Could we do the
>> same
>> here?
>>
>
> I'm planning on adding that logic, but we can't do that until all the
> implicit
> conversions are removed and disabled on a platform, since a bool conversion
> would otherwise be ambiguous. I am currently planning a followup pass to
> rewrite
> unnecessary .get() calls in boolean tests once the conversion operator is
> disabled on each platform.
>

Oh this is you doing that work. *100 cheers*

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[sky, 2014-08-25 20:00:25]

Ugh, what a hassle, but I get why you need to do it in this order. LGTM

[dcheng, 2014-08-25 21:07:47]

The CQ bit was checked by dcheng@chromium.org

[commit-bot: I haz the power, 2014-08-25 21:09:34]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dcheng@chromium.org/496313004/1

[commit-bot: I haz the power, 2014-08-25 21:50:16]

Committed patchset #1 (1) as 6e7ee14fa391f348cccbc7d096ebfc11fffc20a3

[commit-bot: I haz the power, 2014-09-10 02:37:42]

Patchset 1 (id:??) landed as
https://crrev.com/0efb94f2fd2e0ebc9ea572697e935d1021d0a5e3
Cr-Commit-Position: refs/heads/master@{#291764}