content: Split web_drag_dest_gtk.cc into chrome/ and content/ parts.

content: Split web_drag_dest_gtk.cc into chrome/ and content/ parts.

This patch doesn't move web_drag_dest_gtk.cc into content/, but does remove all
dependencies on chrome/ code.

- Split out a delegate interface and move it 
- cross platform code from chrome/.../bookmark_node_data.cc is moved to
  ui/.../bookmark_format.h
- gtk_util.cc has some layers split into gtk_screen_utils.cc.

BUG=none
TEST=none


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=105216

[Elliot Glaysher, 2011-10-11 00:42:34]

Split the actual moving of the file into another CL.

[jam, 2011-10-11 01:42:20]

http://codereview.chromium.org/8196001/diff/5001/chrome/browser/tab_contents/...
File chrome/browser/tab_contents/web_drag_dest_gtk.h (right):

http://codereview.chromium.org/8196001/diff/5001/chrome/browser/tab_contents/...
chrome/browser/tab_contents/web_drag_dest_gtk.h:21: class
WebDragBookmarkDelegate {
nit: this should be in its own file since current convention is to put each
class/enum etc in a separate file.

http://codereview.chromium.org/8196001/diff/5001/chrome/browser/tab_contents/...
File chrome/browser/tab_contents/web_drag_dest_gtk_unittest.cc (left):

http://codereview.chromium.org/8196001/diff/5001/chrome/browser/tab_contents/...
chrome/browser/tab_contents/web_drag_dest_gtk_unittest.cc:1: // Copyright (c)
2011 The Chromium Authors. All rights reserved.
do we not care to test this stuff anymore? seems bad to lose a test

http://codereview.chromium.org/8196001/diff/5001/content/browser/tab_contents...
File content/browser/tab_contents/gtk_drag_utils.h (right):

http://codereview.chromium.org/8196001/diff/5001/content/browser/tab_contents...
content/browser/tab_contents/gtk_drag_utils.h:1: // Copyright (c) 2011 The
Chromium Authors. All rights reserved.
nit: all the other files in this change that have "gtk" in their name put it at
the end, shouldn't this be consistent?

http://codereview.chromium.org/8196001/diff/5001/ui/base/dragdrop/bookmark_fo...
File ui/base/dragdrop/bookmark_format.h (right):

http://codereview.chromium.org/8196001/diff/5001/ui/base/dragdrop/bookmark_fo...
ui/base/dragdrop/bookmark_format.h:21: // Returns a format used across views for
writing bookmark data based on the
1) why is bookmark stuff in ui/base? is this for dragging and dropping
bookmarks? if so this is a chrome feature and i'm not sure why it's in ui/base
2) we don't want content knowing about views. will content include this file?

ben is more familiar with me about what goes in to ui/base, so adding him for
his opinion

[Elliot Glaysher, 2011-10-11 17:53:57]

http://codereview.chromium.org/8196001/diff/5001/chrome/browser/tab_contents/...
File chrome/browser/tab_contents/web_drag_dest_gtk_unittest.cc (left):

http://codereview.chromium.org/8196001/diff/5001/chrome/browser/tab_contents/...
chrome/browser/tab_contents/web_drag_dest_gtk_unittest.cc:1: // Copyright (c)
2011 The Chromium Authors. All rights reserved.
On 2011/10/11 01:42:21, John Abd-El-Malek wrote:
> do we not care to test this stuff anymore? seems bad to lose a test

This test checked for a case where the internals of WebDragDestGtk tried to
access a TabContentsWrapper when all it had was a TabContents. This test just
exercises a WebDragDestGtk without a TabContentsWrapper to make sure it doesn't
crash.

Now WebDragDestGtk doesn't have any dependency on TabContentsWrapper. This test
will never crash.

http://codereview.chromium.org/8196001/diff/5001/ui/base/dragdrop/bookmark_fo...
File ui/base/dragdrop/bookmark_format.h (right):

http://codereview.chromium.org/8196001/diff/5001/ui/base/dragdrop/bookmark_fo...
ui/base/dragdrop/bookmark_format.h:21: // Returns a format used across views for
writing bookmark data based on the
On 2011/10/11 01:42:21, John Abd-El-Malek wrote:
> 1) why is bookmark stuff in ui/base? is this for dragging and dropping
> bookmarks? if so this is a chrome feature and i'm not sure why it's in ui/base

This is for drag/dropping bookmarks. I think I know a way to fix this and keep
all the constraints on directories...

[Elliot Glaysher, 2011-10-11 21:58:13]

(This patch doesn't have trybots because they're down. I'll run them before I
commit though; want to at least get more comments.)

http://codereview.chromium.org/8196001/diff/5001/content/browser/tab_contents...
File content/browser/tab_contents/gtk_drag_utils.h (right):

http://codereview.chromium.org/8196001/diff/5001/content/browser/tab_contents...
content/browser/tab_contents/gtk_drag_utils.h:1: // Copyright (c) 2011 The
Chromium Authors. All rights reserved.
On 2011/10/11 01:42:21, John Abd-El-Malek wrote:
> nit: all the other files in this change that have "gtk" in their name put it
at
> the end, shouldn't this be consistent?

In most of the gtk directories, like ui/base/gtk/, files mostly start with
"gtk_". Still, moved this instance since it's in a shared directory.

[jam, 2011-10-12 19:53:15]

lgtm (although I just assumed you moved all the gtk code, I didn't review that
in detail)

[Elliot Glaysher, 2011-10-12 19:57:41]

On 2011/10/12 19:53:15, John Abd-El-Malek wrote:
> lgtm (although I just assumed you moved all the gtk code, I didn't review that
> in detail)

Correct. The only change was moving their location and namespace. Function
bodies weren't touched.

[commit-bot: I haz the power, 2011-10-12 20:31:40]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/erg@chromium.org/8196001/14001

[commit-bot: I haz the power, 2011-10-13 00:40:26]

Change committed as 105216