aura: Make 'ui' buildable without gtk.

aura: Make 'ui' buildable without gtk.

BUG=97131
TEST=ui_unittests links and runs with use_aura=1

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=102131

[sadrul, 2011-09-21 15:36:01]

The more significant changes:
 * Rename PlatformFontGtk to PlatformFontPango
 * Rename clipboard_linux to clipboard_gtk.
 * Add a new clipboard_aura
   - perhaps this would be used only on linux, and windows will continue to use
clipboard_win?
 * Stub out definitions of a number of Native types for AURA. This is done in a
way that this does not affect these definitions on OS_WIN
(native_widget_types.h).

[Ben Goodger (Google), 2011-09-21 16:08:39]

http://codereview.chromium.org/7977025/diff/51/ui/aura/demo/demo_main.cc
File ui/aura/demo/demo_main.cc (right):

http://codereview.chromium.org/7977025/diff/51/ui/aura/demo/demo_main.cc#newc...
ui/aura/demo/demo_main.cc:75: #if defined(TOOLKIT_USES_GTK)
I am confused by this... sense seems reversed??

[sky, 2011-09-21 16:11:09]

On Wed, Sep 21, 2011 at 8:36 AM,  <sadrul@chromium.org> wrote:
> Reviewers: Ben Goodger (Google), sky,
>
> Message:
> The more significant changes:
>  * Rename PlatformFontGtk to PlatformFontPango
>  * Rename clipboard_linux to clipboard_gtk.
>  * Add a new clipboard_aura
>   - perhaps this would be used only on linux, and windows will continue to
> use
> clipboard_win?

SGTM.

As Ben has looked at this patch I'm not going to review it.

  -Scott

>  * Stub out definitions of a number of Native types for AURA. This is done
> in a
> way that this does not affect these definitions on OS_WIN
> (native_widget_types.h).
>
>
> Description:
> aura: Make 'ui' buildable without gtk.
>
> BUG=97131
> TEST=ui_unittests links and runs with use_aura=1
>
>
> Please review this at http://codereview.chromium.org/7977025/
>
> SVN Base: svn://svn.chromium.org/chrome/trunk/src
>
> Affected files:
>  M build/linux/system.gyp
>  M chrome/browser/chromeos/login/language_switch_menu.cc
>  M ui/aura/demo/demo_main.cc
>  M ui/base/clipboard/clipboard.h
>  A ui/base/clipboard/clipboard_aura.cc
>  A + ui/base/clipboard/clipboard_gtk.cc
>  D ui/base/clipboard/clipboard_linux.cc
>  M ui/base/clipboard/clipboard_unittest.cc
>  M ui/base/dragdrop/os_exchange_data.h
>  M ui/base/l10n/l10n_util.cc
>  M ui/base/resource/resource_bundle.h
>  M ui/base/resource/resource_bundle_linux.cc
>  M ui/gfx/canvas_skia.h
>  M ui/gfx/canvas_skia_linux.cc
>  M ui/gfx/insets.h
>  M ui/gfx/native_widget_types.h
>  M ui/gfx/pango_util.cc
>  D ui/gfx/platform_font_gtk.h
>  D ui/gfx/platform_font_gtk.cc
>  A + ui/gfx/platform_font_pango.h
>  A + ui/gfx/platform_font_pango.cc
>  M ui/gfx/rect.h
>  M ui/gfx/rect.cc
>  M ui/ui.gyp
>  M ui/ui_unittests.gypi
>
>
>

[sadrul, 2011-09-21 16:14:18]

http://codereview.chromium.org/7977025/diff/51/ui/aura/demo/demo_main.cc
File ui/aura/demo/demo_main.cc (right):

http://codereview.chromium.org/7977025/diff/51/ui/aura/demo/demo_main.cc#newc...
ui/aura/demo/demo_main.cc:75: #if defined(TOOLKIT_USES_GTK)
On 2011/09/21 16:08:39, Ben Goodger (Google) wrote:
> I am confused by this... sense seems reversed??

This is ... a bit weird. When TOOLKIT_USES_GTK is set, the MPX can work both
with and without GTK events, but we don't want GTK events, so we disable them.
When TOOLKIT_USES_GTK is not set, the MPX doesn't know anything about GTK events
at all, so we don't need to explicitly disable gtk events here.

[Ben Goodger (Google), 2011-09-21 16:28:03]

On Wed, Sep 21, 2011 at 9:14 AM, <sadrul@chromium.org> wrote:

> This is ... a bit weird. When TOOLKIT_USES_GTK is set, the MPX can work
> both with and without GTK events, but we don't want GTK events, so we
> disable them. When TOOLKIT_USES_GTK is not set, the MPX doesn't know
> anything about GTK events at all, so we don't need to explicitly disable
> gtk events here.


But why would we be building this code with TOOLKIT_USES_GTK set at all?

-Ben

[sadrul, 2011-09-21 16:47:36]

On 2011/09/21 16:28:03, Ben Goodger (Google) wrote:
> On Wed, Sep 21, 2011 at 9:14 AM, <mailto:sadrul@chromium.org> wrote:
> 
> > This is ... a bit weird. When TOOLKIT_USES_GTK is set, the MPX can work
> > both with and without GTK events, but we don't want GTK events, so we
> > disable them. When TOOLKIT_USES_GTK is not set, the MPX doesn't know
> > anything about GTK events at all, so we don't need to explicitly disable
> > gtk events here.
> 
> 
> But why would we be building this code with TOOLKIT_USES_GTK set at all?

Right now, TOOLKIT_USES_GTK is still turned on for use_aura, because otherwise
aura/aura_shell doesn't compile. Originally, with
http://codereview.chromium.org/7904034/, I turned it off. But it made it
difficult for Oshima san and others to make progress, so I added
http://codereview.chromium.org/7978016 to turn the flag back on. After this CL
lands, with some additional changes, we can turn it back off, and this can be
removed then.

[Ben Goodger (Google), 2011-09-21 16:48:55]

ic. LGTM then.

On Wed, Sep 21, 2011 at 9:47 AM, <sadrul@chromium.org> wrote:

> On 2011/09/21 16:28:03, Ben Goodger (Google) wrote:
>
>  On Wed, Sep 21, 2011 at 9:14 AM, <mailto:sadrul@chromium.org> wrote:
>>
>
>  > This is ... a bit weird. When TOOLKIT_USES_GTK is set, the MPX can work
>> > both with and without GTK events, but we don't want GTK events, so we
>> > disable them. When TOOLKIT_USES_GTK is not set, the MPX doesn't know
>> > anything about GTK events at all, so we don't need to explicitly disable
>> > gtk events here.
>>
>
>
>  But why would we be building this code with TOOLKIT_USES_GTK set at all?
>>
>
> Right now, TOOLKIT_USES_GTK is still turned on for use_aura, because
> otherwise
> aura/aura_shell doesn't compile. Originally, with
>
http://codereview.chromium.**org/7904034/<http://codereview.chromium.org/7904...,
> I turned it off. But it made it
> difficult for Oshima san and others to make progress, so I added
>
http://codereview.chromium.**org/7978016<http://codereview.chromium.org/79780...
turn the flag back on. After this CL
> lands, with some additional changes, we can turn it back off, and this can
> be
> removed then.
>
>
>
http://codereview.chromium.**org/7977025/<http://codereview.chromium.org/7977...
>