Yes. You described the issue very precise. And it would be enough to catch presses occured between frames in a way like keyboard interface do. I'm not an expert in Scheme/Racket, especially in its objective part. But I see that native events are handled in canvas definition. Then they go to fluxus-input-callback. So I wonder is it possible to redefine it from my code?
2013/3/20 Kassen <[email protected]> > On Wed, Mar 20, 2013 at 04:39:09PM +0400, Ivan Youroff wrote: > > Hello there! > > I checked documentation and reference, and haven't found any natural > way > > to implement mouse event handler. Fluxus manual suggests to handle it > in > > every-frame loop using states of (mouse-button). I thinks it's > possible to > > get the switch moment keeping the state from prevous frame, but it > will > > definitely skip short touches (with apple's trackpad for example). > > Is there any simple way to write the callback for exact events like > > button-down/button-up? > > This sounds a bit similar to the problem I had with MIDI; quantising > all analysis of the clock to the frame-rate would give considerable > jitter when trying to figure out the BPM. I went into the C++ code > dealing with incoming MIDI and time-stamped the events there. I > imagine something like that will be needed here, it depends a bit on > what you need. I can see how if this would happen; > -frame tick > -button down > -button up > -frame tick > ....you'd miss the button event. A "button-down-since-last-frame" > would cure that. The keyboard interface has it, maybe the mouse lacks > it. I wonder whether that would be enough. It sounds a bit like what > you want would be more like linking buttons to functions that would > -at a event- be called with -say- the current mouse cursor location as > a parameter. That would be more powerful but considerably harder, I > don't think we have anything like that right now. Without something > like that I don't see how we'll keep awareness of the mouse cursor's > location at the moment of a button-down event; clearly that is likely > important information. > > Interesting issue. > > Yours, > Kas. > -- Kind regards, Ivan Youroff
