|
From: Dan S. <dan...@gm...> - 2021-08-26 18:32:28
|
Hello All, I just put a version of a maple to macsyma translator program in the share folder under translators. The grammar was modified from the original ml grammar developed years ago to translate maple to mupad by François Thomasset, to be processed by antlr. The antlr parser generator can be used in a language independent way so that only a small amount of code actually has to be written to translate. I have learned that there is a mathematica antlr grammar, which could be used to translate mathematica, and I was thinking of translating rubi, which appears to handle, by using rules more integrals than any of the other CA systems out there. Now I know there is mockmma, which also included an earlier version of rubi available. So which does the list think is better: 1) Getting the current version of rubi to work with mockmma and integrating it with Maxima 2) Translating rubi rules into Macsyma rules, and getting them integrated into Maxima. There are about 6500 rubi rules, and it seems to me that they probably should not appear to the user the same way a user rule would. Also, there is a antlr python grammar as well, so it could be used in a translator to translate the symbolic python code available. Please let me know your opinions, I am always interested in them. Dan Stanger |