Erik,

A while back I was toying around with the idea of writing a tool that would use an IDL file from the COM object. The tool would then iterate doing a read access of the COM object properties listed in the IDL. Kinda like a "touch" utility that way the debugger would show all the properties to let you explore the object easier that digging through page after page of help file trying to find a property name. I was told by someone at Devcon, I don't remember who though, that it wasn't worth the effort. I kinda assumed this functionality was coming soon in the debugger...

The basis for the IDL tool would be like the tool I use to extract #define constants out of an COM interface.

>I've been playing around a bit, and observed the things that I liked better about the IE tool over what I had planned. Firstly, I had planned to show object hierarchy in a treeview control, but then show key properties in fields to the right of the treeview. The problem is, how to decide what properties are key. The IE tool handles this nicely- it shows all of them. So great, I thought, I'll do the same, just show the names and values for the properties of all objects in the treeview by displaying them in a grid: one record for each name/value pair. Looking at the JavaScript code in the IE tool, I learned that it uses the JS FOR..IN statement to iterate through all properties and members in any given object. That's cool, I can do the same thing in VFP with AMEMBERS(). But you probably know what I ran into next with this line of thought- VFP doesn't divulge PEMs of objects through the COM interface, only the ones that have been accessed, or under very particular (peculiar) conditions
>(Ed, Dave, remember that one with Nancy about a year ago?).
>
>So I'm stuck. I think that a tool that only allows access to a predetermined set of properties would be crippled, but I can't use VFP to extract the rest of them.
>
>Now, I suppose I could use VFP to insert script into the page, and have the script do the walking for me, and then check out the results with VFP, but that would be too hard. :-)
>
>Anybody have any ideas?