might have figured it out, could I get a verification on this

[Alex Rice]

On Feb 18, 2004, at 9:30 PM, Christopher Mitchell wrote:

> It seems that, and I guess in a way it makes sense, if there is no 
> over-riding handler in the children, the message path goes all the way 
> up until it finds it, then it runs it in that location.
>
> So I've over come this by putting a series of empty handlers in each 
> of the substacks.  It seems like this could be easier if there was a 
> way to specify that certain stacks in parents were not inherited... am 
> I missing this?

The message path is just always there- you can't turn it off for any 
stacks.

It's perfectly acceptable to put an empty handler if you want to trap a 
message and not have it reach the mainstack or whatever parent control.

There is an idiom that can be used instead, and I like this better 
(learned on this list perhaps from Ken Ray). In a mainstack you can 
put:

on preOpenStack
   if the owner of the target is me then
     -- handle mainstack message
   else
     -- handle substack message
   end if
end preOpenStack

Similarly, for a library stack "start using", you want to check the 
target of the message:

on libraryStack
   if the target <> me then exit libraryStack
  ...
end libraryStack

Check out Dar's excellent Message Mechanics primer to learn more about 
the message path.
<http://www.swcp.com/~dsc/revstacks.html>


--
Alex Rice | Mindlube Software | http://mindlube.com