[comp.sources.wanted] lisp interpreting library

thoth@reef.cis.ufl.edu (Robert Forsman) (04/14/91)

  Many of you know about TCL, the embeddable interpreter.  It's nice,
but I want an equivalent for lisp.  I'm thinking of rewriting xconq
using lisp (which would make it much more flexible).

  The things I think I need from the lisp interpreter are:

  multiple interpreters.  I want to be able to have several separate
	lisp workspaces at once (I don't really need this, it just
	would be nice).
  Primitive C routine access to variables, setting and retrieving.
  ability to evaluate strings (how else would you use the interpreter?).
  Garbage collection (is there any lisp that is missing this?)
  Ability to define new intrinsic functions.

  Sorry if any of these specifications seem silly, but I'm not really
a lisp hacker.  I only write GEET code.

  And, no, I probably never will finish the rewrite of xconq, but if I
do you'll see it on comp.sources.??? ?
--
"Rob, you're stupid, and that makes you dangerous." - chess opponent
I deal with Reality as you _don't_ understand it. 
Hammor -= Supreme Commander of the Ice Pirates =-

thoth@snail.cis.ufl.edu (Robert Forsman) (04/16/91)

  Recently I posted asking for an embeddable lisp interpreter.  I've
recieved a few responses.


  They suggest I use elk.  Elk is a scheme interpreter (acceptable)
with lots of X stuff (nice, but R3) but it supports 3 architectures
and doesn't seem to work on the one I tried (sun3, twelve thumbs down,
man).  I probably did something wrong, but if I can't compile it, the
people on the net will have problems compiling it.

  They suggest I use xlisp.  The only versions (notice plural; I
checked a couple of places, but not all 13 kerjillion) of xlisp I
could find were labeled MS-DOS specific.  MS-DOS makes me ill.  None
of the docs were TeX or roffable either, so that's out.

  They suggest I use eli.  eli is part of the andrew system that
happens to come with x11r4 (it may come with other things).  Parts of
andrew are required to use eli, and andrew is huge, twisted and nifty.
Due to the overhead associated with eli I can count that out.



  Allow me to add a few more features I want from this embeddable lisp
interpreter:

  self-contained.  I want a relatively small tar that I can compile
and have the library.  I don't care if I have to pull the pieces out
of a large system, I just want a self-contained package.  (eli fails
here, elk is rather big)

  highly portable.  Writing code that requires machine-specific
assembly language is almost as stupid as writing free software that
uses Motif widgets.  Unless this embeddable lisp interpreter has
assembly for almost every architecture in the world (or at least the
ones that have X and Unix) I don't want to worry about porting it or
worry about making other people worry about porting it. (elk fails
miserably here, I don't know about andrew)

  I want to write a FREE GAME (Gnu General Public License), and I want
a lisp (or scheme) that will go anywhere and do anything (well
almost).  If all else fails I'll snarf the emacs kernel and see how to
embed that.

>   And, no, I probably never will finish the rewrite of xconq, but if I
> do you'll see it on comp.sources.??? ?
--
"Rob, you're stupid, and that makes you dangerous." - chess opponent
I deal with Reality as you _don't_ understand it. 
Hammor -= Supreme Commander of the Ice Pirates =-