On 11 August 2016 at 20:31, 'Martin R' via FriCAS - computer algebra system <[email protected]> wrote: > Dear all! > > given the continued interest in some features of fricas from some sage > users, I had a look at the fricas interface in sage, and I'm thinking of > improving it. >
Great. I would be very happy to see a better Sage interface for FriCAS and would be willing to help with such an effort. > The main question is: how can sage receive the output from fricas? > > currently, sage essentially sends "unparse(result:InputForm)" to fricas, and > then parses the output: it strips away the prompt and the step number and it > looks for the type information, all using regular expressions. > If I recall correctly there is also a peculiar "synchronization" procedure that tries to overcome synchronization problems that were probably due to the old version of pexpect that was in use in Sage until a few releases ago. One side-effect of this is that it is not convenient to use the %%() history function in the Sage interface. To the best of my knowledge this procedure can now be eliminated. > a first improvement I was thinking of is to use the ioHook to separate the > algebra output from the remaining information. however, at least the step > number appearing just before the algebra output is currently not decorated. > I think using ioHook would definitely be an improvement. > is there a better way to get the results of computations - and also to send > instructions to fricas? > It seems to me that the best way to call FriCAS would be via the socket interface that is used by Hyperdoc. Unfortunately it is not documented anywhere. The only source of information that I know of is the source code for Hyperdoc. > I'm afraid, python cannot call lisp directly, can it? > In principle Python can call Lisp if Lisp is implemented as a library. This is supposed to be possible with ECL. But building FriCAS as a library might be a significant challenge. It is not clear if there would be a significant gain for the Sage-Fricas interface since most communication would still have to take place as exchange of command and results strings in any case - unless you want to tackle direct conversion of data types. Bill Page. -- You received this message because you are subscribed to the Google Groups "FriCAS - computer algebra system" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/fricas-devel. For more options, visit https://groups.google.com/d/optout.
