Re: ScrollWheel, Xdps and NSEvent modification

[Richard Frith-Macdonald]

On Friday, March 1, 2002, at 08:52 AM, Pierre-Yves Rivaille wrote:

> > OK, I'll take the bait and say ...
> > Sounds like bloat - what was wrong with using the existing
> > otherEventWithType:... method?
> > That's what the method is there for (actually, I don't think you even 
> > need
> to
> > use that method).
>
> It was not possible to create an event with deltaX, deltaY and deltaZ 
> on MOSX, which is strange because you can create every other kind of 
> mouse event programmatically.
>
> > > 2. NSMouseMoved and NS*MouseDragged event do now use deltaX, deltaY 
> > > and
> > > deltaZ. It's not using eventNumber, clickCount and pressure anymore. 
> > > It's
> not
> > > raising an exception if you call those methods on those kind of 
> > > events, but
> > > it will soon (once i'm sure it does not break anyone's code). I doubt
> anyone
> > > use those features, but you never know.
>
> I forgot to say, that all these changes were MOSX-consistent, I did not 
> came with the silly idea to break things and create a deltaZ value :-). 
> The only additions I made was the new class method, which would, 
> anyhow, be needed as an internal method.
> The question should have been : do we make this class method public or 
> not.
>
> I don't really know why they did add this deltaZ stuff, I suppose the 
> only reason I see is that they had space to fill in the enum ...

Well, if it's MacOS-X bloat, I guess we need to add it :-)
I hadn't looked at the latest MacOS-X changes.

But we should not break the old code in order to add the new stuff - 
that's my main concern.

Basically, to implement the MacOS-X documentaed API we need to have 
these deltas AND eventNumber,
clickCount, pressure.  We can't remove some values to make way for the 
new, we have to increase
the size of the NSEvent object.

So ... the method you are proposing has too few arguments, it needs to 
be -

+mouseEventWithType:location:modifierFlags:timestamp:windowNumber:context:
eventNumber:clickCount:pressure:deltaX:deltaY:deltaZ:


> But there are mice with trackpoints instead of wheels (IBM did this at 
> one point on its desktop), I suppose this does explain why delta{X, Y, 
> Z} are float, no dual value thing.

IIRC, the trackpoint stuff is normally used for fine tuning of mouse 
movements,
so I'd have been inclined to implement it as such
ie mapping a trackpoint movement to a small mouse movement, and applying 
the current
trackpoint position in the backend as an offset to any mouse movement.
But if MacOS has added deltas, I guess we need to conform.

At some time in the future I expect to see true 3d pointer devices 
become available
(ie things where a movement in the third dimension is as natural and 
easy as a
movement in the normal plane - something I've not seen outside rather 
expensive
virtual reality systems) and I we obviously do need an API revision for 
that -
though perhaps the deltas extension would be sufficient for simple 
applications.