Re: [Gnash-commit] gnash ChangeLog server/button_character_instanc...

[Sandro Santilli]

On Sat, Apr 19, 2008 at 09:08:34AM +0200, Sandro Santilli wrote:
> On Sat, Apr 19, 2008 at 12:56:25AM +0000, Sandro Santilli wrote:
> > CVSROOT:    /sources/gnash
> > Module name:        gnash
> > Changes by: Sandro Santilli <strk>  08/04/19 00:56:24
> > 
> > Modified files:
> >     .              : ChangeLog 
> >     server         : button_character_instance.cpp 
> >                      button_character_instance.h 
> >     testsuite/misc-ming.all: ButtonEventsTest.c 
> > 
> > Log message:
> >             * server/button_character_instance.{cpp,h}:
> >               Redesign buttons' "DisplayList" doing proper
> >               load/unload and depth-shift. Execute statically-defined
> >               button actions by pushing them on stack (movie_root
> >               will take care of processing after event).
> 
> One thing to test is having an HIT character being a sprite with childs.
> The now-current model will NOT call ::stagePlacementCallback for the
> HIT character, to avoid triggering the onLoad event, or executing
> actions in it. This was based on tests in the ButtonEventsTest but
> is likely incomplete.
> 
> We will not execute tags of HIT characters at all, so either it's a shape
> or I think it will just never hit your mouse.

I made a quick test using ButtonEventTest.c and using sprites instead
of shapes for the HIT state. Indeed the button didn't receive events at
all, so it seems we're doing the right thing :)

--strk;