In message <Pine.3.89.9406132013.E1906-0100000@ernie.acs.csulb.edu> you wrote:
>I'd really like to take a look at this code and the changes you made to
>work it into circle code... let me know when you feel its to the point
>where its functional and something you might actually let another person
>look over... :)
Well, as soon as I finish running spot tests on the thing I'll be putting
patches against vanilla Circle2.20 code onto ftp.cs.jhu.edu so that Jeremy
can get the code and hopefully massage it into 3.0 compatability.
I'm also planning on revamping the docs on how to use it a bit as they seem
to be somewhat MERC specific.
Most of what I had to change to get this working with Circle were:
-- Rewrite a lot of the mob_commands.c and mob_programs.c functions to
use circles structs.h. This was in fact the hardest part and involved
grubbing around in all the structures in structs.h and db.h
-- Figure out where to stick the calls for bribe, kill, etc triggers (this
is the part I still need to test and make sure I got it right)
-- Create some routines that Merc had but Circle didn't (generally easy)
As part of this mess, I replaced the calls to rand in dice() with the
use of the Mitchell-Moore random number simulator that Merc uses. It's
generally better than the OS random number generation code, and some of
the functions in mob_programs or mob_commands relied on the mitchell-moore
functions.
You should expect to see something consumable by the public in about a week.
--JT/Moonchilde
This archive was generated by hypermail 2b30 : 12/07/00 PST