Python MUD (PyMUD?)

[Chris Lawrence]

In article <slrn9j505e.ggc.scarblac at pino.selwerd.nl>, "Remco Gerlich"
<scarblac at pino.selwerd.nl> wrote:

> Jonathan Gardner <gardner at cardomain.com> wrote in comp.lang.python:
>> 2) Is there any mud hackers out there with advice or ideas? This thing
>> is as fresh as it gets, I want crazy and wacky ideas you've been
>> mulling about in your mind for the past 25 years, and I want this mud
>> to be truly unique.
> 
> I'm interested. Years ago I was a mud addict, always lpmuds, been admin
> on some. And I still think LPC is way cool.

Moi aussi.  I was a big LPmud guy.  I'd look back and say I should have
been paying more attention to my studies, but being a creator probably
kept me from getting into worse trouble escaping from everyday life ;-)

Indeed, one of the things that attracted me to Python was its similarity
to LPC, at least in the sorts of things you could do with it and the
facilities for mappings (dictionaries) and arrays.

> I've been thinking a bit about a mud in Python, the problems come when
> you run both the driver and code that was wrriten by wizards in the game
> (their areas) in the same interpreter. You need to shield the driver
> internals from the mud's objects. Although that's maybe possible with
> rexec and similar, it's still easy to do a DOS attack - you need only
> one drunken wizard to code an infinite loop, and your driver locks up. I
> haven't found a solution yet, haven't thought about it enough. It's very
> interesting :).

My thought at the time (when I was trying to wrap my brain around it)
was to set an alarm signal for +2 seconds; if the signal was raised,
I raised a Python exception.  Of course, there are still cases where the
exception might not happen (while 1: pass).

If I had the time and inclination again, I'd go back and hack on it some
more, especially now that I have a machine that could actually host a
MUD with a full-time T1 connection.  Definitely would be fun to play with
(did I just volunteer? ;-)


Chris