Update MouseEvent (initial pass).

Add new MouseEvent ctors; update WidgetWin & WindowWin usage.

BUG=72040
TEST=Manual mouse event testing.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=77221

[msw, 2011-03-02 23:57:56]

Hey Ben, this is a mess right now, but is it going in the right direction? Can I
trash a bunch of the WidgetWin::On* and consolidate message handling in
WidgetWin::Process* instead? (I'd save any On*'s that are actually being
overridden by WidgetWin subclasses...) I'd be happy to talk in person, sorry
that this is taking so long.
Mike

[Ben Goodger (Google), 2011-03-03 00:13:41]

In general very excited by this approach.

http://codereview.chromium.org/6591120/diff/2001/views/events/event.h
File views/events/event.h (right):

http://codereview.chromium.org/6591120/diff/2001/views/events/event.h#newcode345
views/events/event.h:345: //explicit DropTargetEvent(NativeEvent native_event);
poke

http://codereview.chromium.org/6591120/diff/2001/views/widget/widget_win.h
File views/widget/widget_win.h (right):

http://codereview.chromium.org/6591120/diff/2001/views/widget/widget_win.h#ne...
views/widget/widget_win.h:260: MESSAGE_HANDLER_EX(WM_LBUTTONDBLCLK,
OnLButtonDblClk)
There is definitely room to consolidate all of this with a mouse range handler
and a single mouse handler function instead of all these.

They're overridden in WindowWin as you discovered, but WindowWin could easily
override a single OnMouseMessage() function.

Look at ui/views/widget/native_widget_win.h,cc and copy liberally

[msw, 2011-03-07 19:37:55]

Please review this CL for landing; I feel this is a good intermediary step.
Updating further legacy ctor usage increases the change complexity.

[Ben Goodger (Google), 2011-03-07 22:08:42]

http://codereview.chromium.org/6591120/diff/12001/views/events/event.cc
File views/events/event.cc (right):

http://codereview.chromium.org/6591120/diff/12001/views/events/event.cc#newco...
views/events/event.cc:105: target),
indent this to under type.

http://codereview.chromium.org/6591120/diff/12001/views/events/event_gtk.cc
File views/events/event_gtk.cc (right):

http://codereview.chromium.org/6591120/diff/12001/views/events/event_gtk.cc#n...
views/events/event_gtk.cc:164: FromNativeEvent2 from_native)
I think you need a #if !defined(TOUCH_UI) around this ctor

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.cc
File views/widget/widget_win.cc (right):

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.cc#...
views/widget/widget_win.cc:912: GET_Y_LPARAM(l_param));
indent

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.cc#...
views/widget/widget_win.cc:939: GET_Y_LPARAM(l_param));
indent to under the first (

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.cc#...
views/widget/widget_win.cc:955: GET_Y_LPARAM(l_param));
indent

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.h
File views/widget/widget_win.h (right):

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.h#n...
views/widget/widget_win.h:504: LPARAM last_mouse_move_pos_;
Can you leave this as is - i.e. not a windows type. The WidgetGtk does something
very similar and if you see my ui/views/widget/ impl you can see I've
consolidated this down into the Widget.

[msw, 2011-03-07 23:57:21]

Please double-check my updated indentation.

http://codereview.chromium.org/6591120/diff/12001/views/events/event.cc
File views/events/event.cc (right):

http://codereview.chromium.org/6591120/diff/12001/views/events/event.cc#newco...
views/events/event.cc:105: target),
On 2011/03/07 22:08:42, Ben Goodger wrote:
> indent this to under type.

Done.

http://codereview.chromium.org/6591120/diff/12001/views/events/event_gtk.cc
File views/events/event_gtk.cc (right):

http://codereview.chromium.org/6591120/diff/12001/views/events/event_gtk.cc#n...
views/events/event_gtk.cc:164: FromNativeEvent2 from_native)
On 2011/03/07 22:08:42, Ben Goodger wrote:
> I think you need a #if !defined(TOUCH_UI) around this ctor

Done.

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.cc
File views/widget/widget_win.cc (right):

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.cc#...
views/widget/widget_win.cc:912: GET_Y_LPARAM(l_param));
On 2011/03/07 22:08:42, Ben Goodger wrote:
> indent

Done.

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.cc#...
views/widget/widget_win.cc:939: GET_Y_LPARAM(l_param));
On 2011/03/07 22:08:42, Ben Goodger wrote:
> indent to under the first (

Done.

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.cc#...
views/widget/widget_win.cc:955: GET_Y_LPARAM(l_param));
On 2011/03/07 22:08:42, Ben Goodger wrote:
> indent

Done.

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.h
File views/widget/widget_win.h (right):

http://codereview.chromium.org/6591120/diff/12001/views/widget/widget_win.h#n...
views/widget/widget_win.h:504: LPARAM last_mouse_move_pos_;
On 2011/03/07 22:08:42, Ben Goodger wrote:
> Can you leave this as is - i.e. not a windows type. The WidgetGtk does
something
> very similar and if you see my ui/views/widget/ impl you can see I've
> consolidated this down into the Widget.

Done.

[Ben Goodger (Google), 2011-03-08 00:53:41]

LGTM

[msw, 2011-03-08 01:24:21]

On 2011/03/08 00:53:41, Ben Goodger wrote:
> LGTM

FYI, Windows ran clean:
http://build.chromium.org/p/tryserver.chromium/builders/win/builds/19732
Also, browser_tests re-ran clean on linux with:
http://build.chromium.org/p/tryserver.chromium/builders/linux/builds/15721