Karim BELABAS on Fri, 6 Dec 2002 13:56:42 +0100 (MET)

 Re: gp --texmacs

On Thu, 5 Dec 2002, Justin C. Walker wrote:
> On Thursday, December 5, 2002, at 07:10 PM, Michael Somos wrote:
> > pari-dev
> >
> >     This may be a silly question, but I just tried "gp --texmacs"
> > and now I don't know how to get anything useful out of it or even
> > how to quit. The usual control characters don't get me out. Shalom,
> > Michael
>
> Well, your old friend ^\ should work :-}.
>
> You can check here for a description of TeXmacs if you're unfamiliar
> with it: http://www.texmacs.org.  I don't know for sure what goes on,
> but it's a WYSIWYG editor that can talk with CAS, so I suppose it's
> waiting for something intelligible on its stdin.
>
> If you pass it some recognizable arguments, it will prompt with
> 'verbatim:'.
>
> That's what I know.  No doc that I can see.

The --texmacs option is not usable from within a shell. It is reserved to
the texmacs editor. gp --emacs will also have many display oddities if you
try it from the shell. TeXmacs is much worse, since the integration is tighter.

Karim.

P.S: I developped the TeXmacs interface to GP: you can start a GP shell from
within TeXmacs exactly as within emacs, with a nicer looking output. And in
fact many other CASs, as Justin said, about 15 different ones last time I
checked; the Web site only documents the ones which are free software.

TeXmacs input is "region based", i.e if your cursor is in a "GP environment"
(think of \begin{GP}...\end{GP}), what you type is sent to GP. The
interaction between the different environments is remarkably transparent, e.g
you can copy/paste a GP expression in TeX form, or to a different CAS [ with
a different syntax, with the right setup!!! ], or a TeX expression to any
CAS. It is quite funny to write a paper interactively, the editor (e.g
Macaulay or GP behind the scenes) doing all the algebraic manipulations in
the middle of (what I still think as) a TeX document.

The GP interface includes readline completion since 2.2.5, and texmacs has
fully customizable keybindings. So it is quite usable. It is mentioned in my
PARI pages ("Related stuff" section). See

http://www.math.u-psud.fr/~belabas/pari/texmacs-4.png

for a screenshot (there are others on http://www.texmacs.org, unrelated to
CAS capabilities).

TeXmacs is very impressive and produces nicer output than TeX, but I do not
use it myself. My main problems are
0) yet another lisp-based monster shell [ I don't use Emacs either ]. It is
powerful, open-source, and infinitely customizable, but I simply don't have
the time to tailor it to my needs.
1) it's not 100% TeX-compatible: it cannot flawlessly import my (La)TeX
documents. I found the 0.5% it does not understand painful to fix by hand.
2) it's slow for me on my main machines (OK on a 1.4 GHz PC, though).
3) most PARI output is not meant to be printed in a beautified form. This
is hard to fix with the 'lines' default; clipping must come from the TeXmacs
side, and is currently not available.
4) I just love vi(m:-)
--
Karim Belabas                    Tel: (+33) (0)1 69 15 57 48
Dép. de Mathématiques, Bât. 425  Fax: (+33) (0)1 69 15 60 19
Université Paris-Sud             Email: Karim.Belabas@math.u-psud.fr
F-91405 Orsay (France)           http://www.math.u-psud.fr/~belabas/
--