I've been working on an old ticket, #18138 
(https://trac.sagemath.org/ticket/18138), regarding the Mathematica 
interface. Here's a quick summary of the ticket:

   - Starting with Mathematica 9, the command line front end shipped with 
   Mathematica no longer works with pexpect.
   - The short-term fix is a wrapper 
   script,  $SAGE_ROOT/src/bin/math-readline,  for the proprietary Mathematica 
   front end, but something more robust is desired. (See `git log -p -- 
   $SAGE_Root/src/bin/math-readline`.)
   - The long-term fix is to author a replacement command line Mathematica 
   front end which uses MathLink/WSTP to communicate to Mathematica, as it is 
   assumed MathLink/WSTP will be relatively stable. 

I've written a suitable front end, MathLine, which lives 
here: https://github.com/rljacobson/MathLine. My recent commit introduces 
MathLine as an experimental package.  The associated tarball is produced 
from the 1.0rc tag of the MathLine repository with:

$ git archive --prefix=mathline-1.0/ -o mathline-1.0.tar.gz 1.0rc

MathLine can be tested as a math-readline replacement by building MathLine 
as an experimental package as usual and changing the command in the 
Mathematica interface from command = "math-readline" to command = 
"mathline". I have tested MathLine on macOS Sierra 10.12 and Ubuntu 16.04.2 
LTS with Mathematica 11.01. 

It is not clear how best to proceed. Because the proposal is to replace 
math-readline with the alternative front end (presumably with 
math-readline as a fallback), the front end code needs to be thoroughly 
peer reviewed. Provided MathLine is acceptable, Sage's Mathematica 
interface will need to be modified to use MathLine and, ideally, to take 
advantage of MathLine's options to avoid unnecessary pre-parsing of 
Mathematica output. The question is: Does the peer review of MathLine's 
code occur at the time MathLine is proposed as an experimental package? Or 
does it occur at the time the accompanying modified Mathematica interface 
is proposed? Or some combination?

Best,

Robert

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.

Reply via email to