Re: C++ (project proposal)

From: Daniel Houghton (dasher@TEZ.NET)
Date: 03/10/98


Chris Jacobson wrote:

> On 3/8/98 3:26 PM, Daniel W. Burke (dwb@IX.NETCOM.COM) stated:
>
> I have converted the majority of my code to C++, and am continuing to do
> so.  The flexibility it offers is extreme - constructors and destructors
> replacing those messy CREATE() macros with setup and free_whatever()
> functions, and polymorphism (several functions of same name, same
> function, but taking different arguments) also cuts down on the number of
> different function names needed.
>
> I highly recommend it, even if just for readability.  I am also the only
> one who reads my code, but I gained a keener understanding of CircleMUD
> by having to look at ALL of the code (over the course of time) during my
> conversions.  Even the extra readability it provides just me speeds my
> development.  Plus the classes and other C++ features are just so
> wonderful, you'll find yourself having a lot of fun in the process of
> conversion, and it will make future development much easier.

Given that I am working with a code base that incorporates large portions of
the DeathGate code,
I am already quite interested in a more object-oriented approach.

I think we should start an open "Circle++" project to create a version of the
Circle code that takes
an object-oriented approach to the characters and objects, incorporates fast
hash tables
and intelligent strings, templates, and so forth.  Those of us who want to
experiment with an
object-oriented approach could pool our talent, and our developments, if they
prove solid and
useful, could find their way to the stock CircleMUD base in future revisions.

Yes, there's a lot to be said for the efficiency of C.  I won't advocate a 100%
rewrite, as
there are sections of the CircleMUD code that ought to be mean and lean, and
work just
fine as they are.  But other sections of the code could benefit greatly (I'm
hoping that this
proposed variant isn't sounding too mongrel).  I don't want to be a zealot for
any given
methodology; I'd just like to see some genuinely useful improvements to the
solid, proven
base Circle.

Essentially, for the cost of a little extra overhead, we could have a code base
that features
the one thing that is most important to me in much of my coding:  easy re-use
and expandability.
With as little time as I have to spend on coding, I like being able to develop
the features my
players request as rapidly as possible, without lots of bugs (don't we all!).

Now, I'm just curious to hear what the more experienced on this list think of
this concept. :)

        Daniel Houghton AKA Garadon


     +------------------------------------------------------------+
     | Ensure that you have read the CircleMUD Mailing List FAQ:  |
     | http://democracy.queensu.ca/~fletcher/Circle/list-faq.html |
     +------------------------------------------------------------+



This archive was generated by hypermail 2b30 : 12/15/00 PST