From: Squeak <>
Date: Fri, 21 May 1999 20:36:17 -0400 (EDT)

I wrote a silly little c++ interface to gmp. It does all the operator
overloads that are explicitly provided by gmp. (in other words each method
is under 3 lines of code).

It works for me, and makes writing some code alot nicer.

I don't think it should slow down code much (all the methods are purely
inlined calls to gmp functions). Especially if you turn off Runtime Type
T* Information (--f-nortti or somehting) and virtual functions yada yada.
I don't know beans about c++ (this was a neat learning experience), but
I've read quite a few tutorials now, and I think I've not done any of the
standard mistakes. I'm probably wrong tho.

Feel free to bugtest it if you want, or use it under the same license as
libgmp. Does this already exist somewhere btw? I would think someone
would have done it. I don't think it'd be useful in genius per se, but
it might be nice to write little test proggies, plugins, etc.

I am slowly slowly writing a sub-genius that will handle finite fields
etc. It will be GPL'd and will exist mostly in the hopes that George will
fix it and put it in genius. I find that people accept my code ideas alot
better if they see a working prototype. Even though they usually trash
every line of code I wrote :)

Tips on writing that whole parser dealy are welcome. I'll be putting that
off for a bit, while I get the proggie able to handle internally
constructed ETree's. Clear happy friendly tutorials are a plus. I learn
quick but I really just dont GET what lex and yacc are supposed to do.
This is probably very much related to the fact that I can't write a parser
in c either :)

Hope all you student types are through with finals and starting to relax.


ps add bitwise ops to integers (and gaussian integers). I'll love you
for it, and we'll already have addition in GF(2^n). I'll write it within a
few days, if you don't want to, but you'll still have to look over it.
Like I said I don't know radishes about parsers.
Received on Fri May 21 1999 - 17:11:07 CDT

This archive was generated by hypermail 2.2.0 : Sun Apr 17 2011 - 21:00:02 CDT