On Apr 20, 2:50 am, "Dr. David Kirkby" <david.kir...@onetel.net> wrote:
> This is interesting. > > http://code.google.com/p/symja/ > > Apparently it can parse a large percentage of the Mathematica language and has > nearly 295 functions. I note it has "D[]" and "Integrate[]" as two of them. Well, to the extent that it's interesting to be able to call symbolic math functions in Java, I think one's time would be better spent trying to build an interface to an existing system rather than creating a new one. As you know, symbolic math is hard, and while it is easy to get started, it is a tremendously long haul. Existing symbolic math systems are mostly written in languages other than Java, and often weren't created with the idea that programmatic interaction is any kind of goal, so bolting another program onto them is often tedious and messy. But even so that's a much, much easier problem to solve than replicating all of the algorithms. As a proof of concept, I've compiled Maxima with ABCL which is a Lisp implementation in Java and called Maxima functions from a Java program. It is clumsy, but, if I really wanted to do it, it would be a lot easier to improve the interface than to reimplement the algorithms of interest. I suppose a similar exercise could be carried out with Jython (Python implementation in Java) and Sympy or any pure-Python code in Sage. There is also CL-Python which is an implementation of Python is Lisp. One could, I guess, link Lisp functions into the image and call them from Python; then Python could be your user language. FWIW Robert Dodier -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
