I am about to check in a significant improvement to yesterday's
property-change listeners, once I've finished recompiling and testing
them (probably before 2130Z). The interface now looks like this:
class SGPropertyChangeListener
{
public:
virtual ~SGPropertyChangeListener ();
virtual void valueChanged (SGPropertyNode * node);
virtual void childAdded (SGPropertyNode * parent, SGPropertyNode * child);
virtual void childRemoved (SGPropertyNode * parent, SGPropertyNode * child);
protected:
friend class SGPropertyNode;
virtual void register_property (SGPropertyNode * node);
virtual void unregister_property (SGPropertyNode * node);
private:
vector<SGPropertyNode *> _properties;
};
There are separate notifications for a change value, and added child,
and a removed child, so that user code can take appropriate actions.
The protected methods are callbacks from
SGPropertyNode::addChangeListener to allow two-way pointers; the
destructor will remove all listener references so that there won't be
dangling pointers when the object that's listening disappears. It's
fairly clever, so I'm worried whether it will work.
SGPropertyNode now has three separate methods for firing change
events:
SGPropertyNode::fireValueChanged ()
SGPropertyNode::fireChildAdded (SGPropertyNode * child)
SGPropertyNode::fireChildRemoved (SGPropertyNode * child)
The child-added and child-removed events will *always* be fired
automatically. The value-changed events will be fired automatically
unless the property is tied, in which case the controlling code must
invoke it specifically (or not).
I think that this is enough for anything we'll need to do. I know
that this kind of thing can get a lot more complicated, but I'll
resist that until it's proven that we cannot do without it.
TODO:
1. Write good, up-to-date documentation and a programmer's tutorial
for the property manager.
2. Write a comprehensive test suite for the property manager, since
it's vital to FlightGear's good behaviour and is easy to test in batch
mode.
If there are any volunteers for either of these, I will be very
grateful; otherwise, I'll try to get to them as time is available.
Thanks, and all the best,
David
--
David Megginson, [EMAIL PROTECTED], http://www.megginson.com/
_______________________________________________
Flightgear-devel mailing list
[EMAIL PROTECTED]
http://mail.flightgear.org/mailman/listinfo/flightgear-devel
