front/backscripts and code that disappears

[Shari]

I've run into something I ran into once before, and thought I knew 
the solution for it.  But I finally had to resort to the last resort 
to solve it, and am not happy with the solution.

Clicking on a menu item or on an object in the stack brings up an 
error message, can't find handler.  (All the code is in an external 
stack, not within the user stack).

Under certain conditions, the program (stack or standalone) can't see 
the code.  A front or backscript will exist and I can prove it 
exists, but calling to it fails.  As though there is a wall between 
the stack and the code.

I've tried calling directly to the handler at its source (and 
verified that firstHandler is running right up to the "send" part).

on firstHandler
    do this, that and the other
    send myHandler to cd fld "someField" of cd 1 of stack "someStack"
end firstHandler

And that fails as well.  It still produces a "can't find handler" 
error or some such thing.

It's as though it goes blind, and things are there but it doesn't see them.

There is a series of handlers that run, calling each other, opening 
the "ask" and "answer" stacks, opening text files and closing them, 
and the final handler sends you back to the place of origin, where 
the code that was there before is now broken.  You click on the stack 
and nothing works.

I've verified that:
The front and backscripts exist
The stack you are clicking at is the topstack

Even if Metacard inserted a frontscript of its own, it wouldn't have 
the handler names I use.  So they should still get passed along to my 
scripts.

If you click outside the window (in other words, leave Metacard and 
go to the Finder or some other program) and then click back into the 
window, everything is reset.

It's a puzzlement.

Shari




-- 
Mac and Windows shareware games
http://www.gypsyware.com