[DGD] Re: Parse String and Virtual Objects

[S. Foley]

Regarding parse_string():

This function really isn't too hard to figure out, though the effort 
required is something north of 'none'.  I have no formal training in the 
computer sciences or any programming language and through a combination of 
the list archives, web searches on grammars, and trial and error use of the 
actual function, I was able to figure it out (and as bulbs go, I'm not 
particularly bright ;) ).  After making that kind of effort, you should be 
in a position to ask more specific questions about parse_string.  And, in 
general, the list is more responsive to specific questions, since they 
indicate that you've at least made some effort to understand things on your 
own, and beyond that, they don't require a veritable textbook to answer.

When I've posted specific questions regarding parse_string to this list, 
they have been answered quickly, completely, and politely.

---

Arnau,

By virtual objects, do you mean objects with objects names that do not 
correspond to any actual existing file?  If so, I believe the solution to 
your problem is much simpler.  The compile_object() help file states:

"Compile an object from a LPC file, specified by the argument string
with ".c" appended.  The returned object will have the argument string
as name."

So, you could do something like check for the existence of a file 
corresponding to the first argument passed into compile_object.  If a file 
didn't exist, you could pass the argument off to another daemon that could 
generate an appropriate object for the name specified and return it back 
into compile_object() (which would then do whatever it needed to with the 
object, before returning it).  I *think* this object could even be a clone.  
I'm not certain what impact this might have on your ability to upgrade 
objects, but I *think* an object registration daemon written with this 
behavior in mind could accomodate it.  This would obviate the need for your 
mapping, as well as the need to add in extra checks to call_other, et.al.

I haven't actually tried this, so if I'm wrong hopefully someone on the list 
will correct me.  And even if what I suggest is possible, perhaps someone 
could chime in on the drawbacks to this design.  I know a few concerns leap 
immediately to my mind.

--Steve F.

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

_________________________________________________________________
List config page:  http://list.imaginary.com/mailman/listinfo/dgd