(User.programs) Option: 
Item: 113 by _sirmax at hpcvbbs.UUCP
Author: [Jeff Maxwell]
  Subj: Radical/Fraction program
  Keyw: radical fractions program
  Date: Mon Sep 24 1990 19:26 
----------
  Resp: 2 of 3 by _michaelowry at hpcvbbs.UUCP
Author: [Michael Andrew Lowry]
  Date: Fri Sep 28 1990 20:04 

Hi, I am a friend of Jeff, and I have recently risen to the challenge of
writing such a program. I TRIED to use s method similar to your's, but I
ran into so many problems that I started loosing sleep; I then quickly
wrote a rougher program that will do just about the same thing (it can't
find fractions inside the radical) I simply used nested FOR statements to
generate succesive fractions, took them out of the original # one by one,
until the result squared yeilded an integer. I have another program that
will simplify a radical expression (the expression, however has to be
stored interms of different variables representing distinct parts of the
expression- I don't want to bother with dissassembling an algebraic using
POS and SUB!!) Well, I am intrigued by your idea- I'll attempt to design
a working algorithm. I have a MATCH program that will expand a binomial
in the form (A+B)^N  It will be attached to another msg. Happy
programming. Oh, by the way, I'm just 18 - do you know many people my age
that are into this kind of stuff? 
                                    - Michael Lowry

----------
  Resp: 3 of 3 by _dsl at hpcvbbs.UUCP
Author: [David Lowenstein]
  Date: Wed Oct 17 1990 22:26 

[Note: Also see FRAD2, someone else's approach to this.  -jkh-]

   Here's a program that works using the method described above. It calls on
'PF', my own prime factor program, and 'FS', which is useless for anything
else. 'OQF' will do what you ask and 'QF' will do the same except with an
integer in the radical, but the expression will be equivalent. 'MN' only
returns a simple mixed number. I'm working on a smaller, faster version, so let
me know if this is on the right track.

qrad (1641 bytes)
[converts a decimal to A/B*SQR(C/D)]

[Note: David's PF is very slow for big numbers.  See the PF in the PRIME
directory on disk for a much faster factorizer.  -jkh-]