On Tue, 10 Aug 2004, Ralf Stephan wrote:
So, assuming I have summarized the situation correctly, which way?
A while ago, I had followed the way of invoking the (powerful) univariate factoring in NTL but gave that up (due to non-technical reasons). I had the impression that the architecture of NTL was much more consistent than that of the other candidates. Also, I would be reluctant to add too many prerequesite libraries (libfac + ntl). As for the interface, I think the issue is much simpler than in the case of CLN: there're already normal(), fracfree() and friends for guidance. Maybe another routine for power users that populates a vector of factors/exponents would be handy. Regards -richy. -- Richard B. Kreckel <http://www.ginac.de/~kreckel/>