Re: A handful of LISP questions

> Yes. It's a great exercise.


It sure is! I started off basically copy and pasting Paul Graham's
LISP and changing it into C code. And from there, I was able to
implement first- class functions (dynamic scope though... still need
to figure out lexical from C). I finally got to a critical point a
few weeks ago where I was just able to add feature after feature and
see the results of my project. But I don't like working on messy code,
and so the last week and a half, I have avoided doing anything
further.


> You might enjoy looking at xlisp. http://almy.us/xlisp.html .
> You can learn something about compiling a small lisp with


http://hedgehog.oliotalo.fi/




> [1] http://t3x.org/bits/s9fes/
[2] Jones, Lins
                "Garbage Collection"
                Wiley & Sons, 1996




> http://www.faqs.org/faqs/scheme-faq/part1/section-4.html
> http://halogen.note.amherst.edu/~jdtang/scheme_in_48/tutorial/overview.html
> http://halogen.note.amherst.edu/~jdtang/scheme_in_48/tutorial/parser.html
> http://www-swiss.ai.mit.edu/~jaffer/SCM.html
> http://search.cpan.org/~sfink/parrot-0.0.11.2/languages/scheme/Scheme/Parser.pm






> A complication is the mechanism of read-time macros. For instance, the
QUOTE facility.


I have the ' expanding properly to quote (thank god). However, the
last thing I added was support for quasiquotes and unquotes. I don't
have the ` and , operators working, and that is the big motivation for
me to rewrite my parser. That and because of dotted pairs (which I
don't yet support).


Additionally, my goal is to make a more usable LISP. I come from a
Python background, and I would like to make a language with very
clean, elegant syntax. I like what ' ` and , do for LISP, and I want
to add more of it where it is appropriate. And because of that, I need
a very clean way to make changes to the language's grammar.




>> In either case, what is the best way to ensure I still have access to
>> the current environment?


>What do you mean by this?


Let me rephrase. When garbage collecting with Mark and Sweep, you need
a root set of objects. This is the set of objects in immediate view of
the program. For example, the objects on the call stack or in the
current environment. I guess from what people have told me on this
list is that you call the collector at the start of the eval function
when memory exceeds a threshold. Then, in that case, do I just flag
everything in the current environment as being part of the root set?




> Do you want to implement a Lisp-1 (like Scheme) or Lisp-2 (like Common
Lisp) ?


Oh god, lisp-1 please. I am not friends with Common Lisp. It reminds
me of perl with parentheses =-) Again, I'm looking for elegance, so
I'm going to have a strong biased towards the Scheme ends of things.


> However, if you seriously want to write a lisp interpreter then you
should have a look into "Lisp in small pieces" by Queinnec & Callaway


I actually purchased this book a week and a half ago. It is very
interesting. It is a bit tricky at some parts though, and requires
you to have a fair understand the concepts in question. The part about
continuations completely blew me away (as I had never used call/cc
before). But I'll keep at it, even if it takes me a while.


Thank you everyone who replied. You have all been very helpful. I hope
to make good progress with this over the next few months... or
whenever I have time =-)

Return to the comp.compilers page.
Search the comp.compilers archives again.