On Sun, Sep 2, 2012 at 12:19 AM, Benn Snyder <email@example.com>
I've made steady progress integrating my code into Jmol. I've split it
up into a few classes. They are
moleK - embeds a Jmol viewer for my testing purposes and as an example
JmolMultiTouchClientAdapter - adapter between OpenNI and Jmol
org.jmol.multitouch.HandTracker - encapsulates OpenNI hand tracking
org.jmol.multitouch.ImageCapture - encapsulates OpenNI image capture
org.jmol.multitouch.DtwGestureRecognizer - external gesture recognizer
1) In JmolOpenNIAdapter, I need to know the width and height of the Jmol
view. I can't seem to figure out how to get them.
2) In JmolOpenNIAdapter.doAction(), I make calls to a JmolViewer to
perform scripting operations. Is this an acceptable way to do it? Is
this the best way to do it? The good news is that doAction() is the
only place JmolOpenNIAdapter touches Jmol, so it shouldn't be too hard
to switch implementations.
I don't think so. You want to adapt org.jmol.multitouch.ActionManagerMT to your purposes. You can see there that there are options right now for Molecular Playground, tablets, and SparshUI. I think you would logically add another. Your client will only interact with ActionManagerMT.
As I continue to refine, I'm going to start looking at invoking
NITE-specific functionality, which can provide more gestures and more
sophisticated user tracking.