bothner@sevenlayer.cs.wisc.edu (Per Bothner) (02/07/91)
Archive-name: symbolic-math/big-numbers/bignum/1991-02-06 Archive: sevenlayer.cs.wisc.edu:pub/bignum.tar.Z [128.105.8.11] Original-posting-by: bothner@sevenlayer.cs.wisc.edu (Per Bothner) Original-subject: Re: Bignums, Rationals, Incremental gaussian elimination & simplex Reposted-by: emv@ox.com (Edward Vielmetti) In article <1722@n-kulcs.cs.kuleuven.ac.be>, bimandre@saturnus.cs.kuleuven.ac.be (Andre Marien) writes: >- Infinite Precision Arithmetic (i.e. bignums, rationals and others): > * Packages (any information on existing routines, their > availability (cost, where, etc.), experience with them would be > welcome). I've been using the Bignum package from DEC's Paris Research Lab. It has efficient assembler primitives for a number of machines, and generic C routines for other routines, so you get the best (more or less) of efficiency and portability. Its primitives implemenent unsigned arithmetic; on top of that there is a library that implements signed-magnitude integers (I don't use the latter). To get it, either send mail to: librarian@decprl.dec.com or ftp it from sevenlayer.cs.wisc.edu:pub/bignum.tar.Z. The latter version includes an improved Makefile I wrote that automatically selects the correct version to compile. I have a set of C++ classes and routines that implement generic arithmetic, including two's complement integers (which gives compatibility with Common Lisp), fractions, double floats, and complex numbers. It is by no means "complete", but most of the non-trivial rational and integer routines are implemented. The code is in sevenlayer.cs.wis.edu:pub/gennum.[Ch]. Please let me know if you use this code, or improve it. -- --Per Bothner bothner@cs.wisc.edu Computer Sciences Dept, U. of Wisconsin-Madison