Fix smelly scoped_ptr code in print_web_view_helper_win.

Fix smelly scoped_ptr code in print_web_view_helper_win.

A series of changes resulted in the need to store an owned pointer in a different scoped_ptr temporarily.  This change changes the ownership model of the called method to remove the hack.

BUG=NONE
TEST=NONE


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=95954

[Lei Zhang, 2011-08-06 01:38:34]

build bot failed?

http://codereview.chromium.org/7491074/diff/1/chrome/renderer/print_web_view_...
File chrome/renderer/print_web_view_helper_win.cc (right):

http://codereview.chromium.org/7491074/diff/1/chrome/renderer/print_web_view_...
chrome/renderer/print_web_view_helper_win.cc:138: // Release since
|print_preview_context_| is the real owner.
remove comment

[vandebo (ex-Chrome), 2011-08-08 21:19:14]

On 2011/08/06 01:38:34, Lei Zhang wrote:
> build bot failed?

Done.

http://codereview.chromium.org/7491074/diff/1/chrome/renderer/print_web_view_...
> File chrome/renderer/print_web_view_helper_win.cc (right):
> 
>
http://codereview.chromium.org/7491074/diff/1/chrome/renderer/print_web_view_...
> chrome/renderer/print_web_view_helper_win.cc:138: // Release since
> |print_preview_context_| is the real owner.
> remove comment

Done.

[Lei Zhang, 2011-08-08 22:08:59]

http://codereview.chromium.org/7491074/diff/5001/chrome/renderer/print_web_vi...
File chrome/renderer/print_web_view_helper_win.cc (right):

http://codereview.chromium.org/7491074/diff/5001/chrome/renderer/print_web_vi...
chrome/renderer/print_web_view_helper_win.cc:131: // In the preview flow,
RenderPage will never return a new metafile.
Should we add a DCHECK() to make sure this stays true?

[vandebo (ex-Chrome), 2011-08-08 22:11:29]

http://codereview.chromium.org/7491074/diff/5001/chrome/renderer/print_web_vi...
File chrome/renderer/print_web_view_helper_win.cc (right):

http://codereview.chromium.org/7491074/diff/5001/chrome/renderer/print_web_vi...
chrome/renderer/print_web_view_helper_win.cc:131: // In the preview flow,
RenderPage will never return a new metafile.
On 2011/08/08 22:08:59, Lei Zhang wrote:
> Should we add a DCHECK() to make sure this stays true?

Can we do that without introducing a warning?  The variable will only be used in
a DCHECK and therefore won't be used in Release mode.

[Lei Zhang, 2011-08-08 23:04:11]

On 2011/08/08 22:11:29, vandebo wrote:
>
http://codereview.chromium.org/7491074/diff/5001/chrome/renderer/print_web_vi...
> File chrome/renderer/print_web_view_helper_win.cc (right):
> 
>
http://codereview.chromium.org/7491074/diff/5001/chrome/renderer/print_web_vi...
> chrome/renderer/print_web_view_helper_win.cc:131: // In the preview flow,
> RenderPage will never return a new metafile.
> On 2011/08/08 22:08:59, Lei Zhang wrote:
> > Should we add a DCHECK() to make sure this stays true?
> 
> Can we do that without introducing a warning?  The variable will only be used
in
> a DCHECK and therefore won't be used in Release mode.

I think we can. DCHECK has some magic to keep the compiler happy. At least for
me on Linux. You can add a DCHECK and run a try job against the win_rel try bot
and see what happens.

[vandebo (ex-Chrome), 2011-08-09 00:22:47]

http://codereview.chromium.org/7491074/diff/5001/chrome/renderer/print_web_vi...
File chrome/renderer/print_web_view_helper_win.cc (right):

http://codereview.chromium.org/7491074/diff/5001/chrome/renderer/print_web_vi...
chrome/renderer/print_web_view_helper_win.cc:131: // In the preview flow,
RenderPage will never return a new metafile.
On 2011/08/08 22:08:59, Lei Zhang wrote:
> Should we add a DCHECK() to make sure this stays true?

Done.

[Lei Zhang, 2011-08-09 00:24:46]

LGTM

[commit-bot: I haz the power, 2011-08-09 06:30:11]

Change committed as 95954