Re: [AD] Disagree with iPhone mouse handling

[ Thread Index | Date Index | More lists.liballeg.org/allegro-developers Archives ]


On Sat, 2010-11-06 at 10:54 -0600, Trent Gamblin wrote:
> On 2010-11-06, at 10:34 AM, Evert Glebbeek wrote:
> > What was the rationale behind doing it?
> 
> I guess the examples work by simulating a mouse, where there is a cursor,
> so they rely solely on mouse AXES events.
> 
> > Agreed, with the provision that it should behave the same across platforms as much as possible, so there may be situations where it's legitimate for Allegro to insert events of its own. Not saying this is one of them.
> 
> We've come to a good compromise I think. The axes events are still generated, but
> the MOUSE_UP event is generated after the axes event now instead of before. This
> allow both methods of mouse tracking to work.
> 

Yes, the idea was to have apps written only with mouse support in mind
work better. And splitting one event in two didn't seem to have any bad
side effects. The problem Trent saw was an actual bug. When a touch
occured, originally the events were:

DOWN, UP

The fake-axes-events change made that be:

DOWN, AXES, UP, AXES

But since there cannot be an AXES event when no touch is down the
correct way is:

DOWN, AXES, AXES, UP

I don't see a case where those two extra AXES events make a difference
anymore though. I remember it was in one or more of the examples when I
tried them all on my iPad a while back. Will have to try them again.

-- 
Elias Pschernig <elias.pschernig@xxxxxxxxxx>





Mail converted by MHonArc 2.6.19+ http://listengine.tuxfamily.org/