Re: Converting xconq to C++

[bboett at bboett dot dyndns dot org (Bruno Boettcher)]

On Fri, Mar 29, 2002 at 01:02:15PM +0100, Alan Schmitt wrote:
> What I'd really like to understand, if xconq is ported to another
> language, is what we gain from it. If xoncq fits nicely in an object
hmm seems i am too late to really give my opinion :D but here it goes
nevertheless...

as others stated OO is more fun to code and allows to strip down the
code into managable trunks.... furhter it allows to test and apply all
those nifty concepts we learned at university (even if it is now some
time ago :D)

as sayd i tryed my self a port to java some time ago, with some other
guy from this list, but it was a fiasko :D :D

from my time scedule i will only issue advice and not be able to let
those follow some real help, but anyways :D

> oriented programming style, why not. But choosing which object language
> to use depends on 3 things I can think of: efficiency (both of the
> resulting code and of the coding process), portability, and extra
> features (what librairies are available, what additional language
> constructs may be used).
agreed!

> language that could be used: objective caml (http://caml.inria.fr). In
yup nice! i like also eiffel a lot!

> why not rewrite xconq (it seems to be quite an undertaking, but why
> not), why not do it in a different programming style, but why c++ ?
the general idea is that the migration from C to C++ is really easy....

but, having myself participated in the migration of a C program into a
C++ program (don't know if anyone knows the circuit simulator acs...) it
can go down the path of the programmer hell.....

at last i rewrote a new simulator from scratch (even if largely
    profiting from the concepts i acquired by helping port acs to C++)

concerning xconq i think the porting will be simpler, since the C base
is way more sound than the acs base was ever, but since we are about
rewriting, why not try to change directly some paradigms in the game and
make immediately a move to a client server architecture?

i know it is a recurrent design thread :D and that the actual stage now
appears to come back to something usable, but i really think that a
client server design would be way more robust. 

Moreover, a client-server design is really easy to implement in C++,
there are even ready library classes and templats for such stuff....

so i would suggest another approach:

take the actual code base, and build up a new sourcebase, copying from
the old code base where appropriated, but building on an own scema....


-- 
ciao bboett
==============================================================
bboett@adlp.org <- UPDATE your address books please!!! CHANGE!
http://inforezo.u-strasbg.fr/~bboett
===============================================================
the total amount of intelligence on earth is constant.
human population is growing....