I’m really going to put scheme and common lisp in the same category?


  • Racket
  • Chez
  • Guile
  • Gambit
  • Chicken

  • SBCL

Gambit vs Chez vs Guile scheme. Me dunno Gerbil is a system on top of Gambit approximately similar to racket

(print "hello world")
(print "~a foo" 'bar)
#lang racket
(print "hello world")



ribbit scheme

scheme primer

Pre-Scheme: A Scheme Dialect for Systems Programming

https://gitlab.com/spritely/guile-hoot hoot is scheme to wasm

https://github.com/Calysto/calysto_scheme jupyter enabled scheme



https://github.com/michaelballantyne/faster-minikanren raco pkg install minikanren

#lang racket
(require minikanren)
(run 1 (q) (fresh (x y z) (== x q) (== 3 y)))

http://io.livecode.ch/ more minikanren examples implementing microkanren

http://tca.github.io/veneer/examples/editor.html minikanren examples.


lambda kanren assume all A New Higher-order Unification Algorithm for 𝜆Kanren



raco pkg install

Creating Languages in Racket (2011)


beautiful racket

https://github.com/wilbowma/cur a curious dependently typed proof assistant


racket contest good links

Mythical Macros tutorial https://soegaard.github.io/mythical-macros/

Macros and Languages in Racket book draft http://rmculpepper.github.io/malr/ 1

Racket Guide: Creating Languages https://docs.racket-lang.org/guide/languages.html

Racket Docs search for all languages: https://docs.racket-lang.org/search/index.html?q=H%3A

Extensible Pattern Matching in an Extensible Language

Bowman how to hashlang Macro-embedding Compiler Intermediate Languages in Racket


implementing type systems as a macros Type Systems as macros

Rhombus: A New Spin on Macros without All the Parentheses



Rosette - should be a confluence of my interests. I feel like I want some kind of macro engine for smtlib. Perhaps it makes intermiedtae queries, perhaps not. stage0 would be “just” a macro expander for smtlib.

#lang rosette
(define-symbolic b boolean?)
(assert #t)
(assume b)
(verify b)

;(define-symbolic b boolean?)

https://rdivyanshu.github.io/ma122.html Solving a mathemattic problem using Rosette

Common Lisp

common lisp resources

Lisp-2 - seperate namespace for functions and vasriables -Common lisp Condition system lisp class system CLOS

https://letoverlambda.com/index.cl/guest/chap2.html let over lambda. Let in common lisp makes a ref cell?



alive vscode extension

getting started sbcl manual quick lisp library manager This makes it easy to load up packages. I think using this is quite strandard. Follow instructions to curl and load and add to init file

lisp books

Practical Common Lisp Common Lisp the Language 2nd ed


echo $'
(write-line "Hello, World!")
(ql:quickload "vecto")
(format t "Hello, world!")
; comment one
Multiline comment are nestable

(defun fib (n)
  "return fib nuymber"
  (if (< n 2)
      (+ (fib (- n 1)) (fib (- n 2)))
(fib 3)
(funcall #\'fib 3)  ; funcall takes multiarity args. Functions quoting
(apply #\'fib (list 3)) ; apply takes list of args

; multiple value returns
(defun many (n)
  (values n (* n 2) (* n 3)))

(let ((str "Hello, world!")) ; let and let*
  (string-upcase str))

; defparameter and defvar

(defparameter *string* "I\'m global")

(defun print-variable ()
  (print *string*))


(defun foobar (a b c) (+ a (* b c)))
(compiled-function-p #\'foobar)
(disassemble #\'foobar)

(format t "~a" t)

(vector 1 2 3)
#(1 2 3)
' | sbcl 


(declaim (optimize (debug 3))) https://malisper.me/category/debugging-common-lisp/

save-lisp-and-die to save an image of current process.


art of metaobject protocol


Condition System

Like exceptions except you can restrta from them book


https://www.khoury.northeastern.edu/home/pete/research.html Manolios teaches courses https://www.cs.utexas.edu/users/hunt/research/index.html Hunt

The main theorem prover associated with common lisp. It is a granddaddy of a lot of other systems https://github.com/acl2/acl2

sudo apt-get install acl2 works.

proofpad? http://proofpad.org/ online version too http://new.proofpad.org/ There’s a little tutorial button

“sedan”. Looks out of date, eclipse doesn’t warm my heart. I dunno.

applicative common lisp, it’s a variant of common lisp using only pure (applicative) functions.

echo "
(+ 1 1)
(append (list 1 2) (list 3 4))
(cons 1 nil)
(car '(1 2 3)) ; cdr first rest
(defun triple (x) (* 3 x))
; Since TRIPLE is non-recursive, its admission is trivial.  We observe
; that the type of TRIPLE is described by the theorem 
; (ACL2-NUMBERP (TRIPLE X)).  We used primitive type reasoning.
(endp nil) ; t
(evenp 3) ; nil
(equal t t)

(defun hailstone (x)
  (if (evenp x) ; condition
      (/ x 2) ; what to do if it passes
      (+ (* 3 x) 1))) ; what to do if it fails

; uses guards. acl2 is proving termination
(defun factorial (n)
  (if (zp n)
      (* n (factorial (- n 1)))))

(thm (> (factorial n) 0)) ; QED

:q ; exit acl2 rep back to lisp
(lp) ; back to acl2

:args assoc-eq ; guards and conditions associated wth


; :mini-proveall

; (include-book )
" | acl2

Needs to reason about terminatin to define stuff.

https://hilbert.math.wisc.edu/logic/kunen/ramsey.pdf - ramsey theorem in ACL2

ACL2 hypperreals

Gamboa course on discrete math has some vids https://www.youtube.com/watch?v=R0vgJbskXbM&ab_channel=RubenGamboa https://arxiv.org/pdf/2205.11695.pdf teaching acl2 gamboa https://github.com/rubengamboa/acl2_bridge https://github.com/rubengamboa/acl2_jupyter https://github.com/rubengamboa/acl2-math-book https://github.com/rubengamboa/acl2-docker-images https://github.com/rubengamboa/acl2-docker-images/tree/main/context/acl2-notebooks/programming-tutorial jupter notebooks

rex page

https://www.cs.utexas.edu/users/moore/publications/km97a.pdf A Precise Description of the ACL2 Logic https://cstheory.stackexchange.com/questions/32749/proof-software-for-primitive-recursive-arithmetic https://proofassistants.stackexchange.com/questions/1867/is-acl2-first-order-logic-and-is-this-a-limitation-for-proving-math :doc ordinals

Defines ordinals hmm https://www.khoury.northeastern.edu/home/pete/pub/acl2-ordinal-arithmetic-acl2.pdf odinal airthmetic in acl2. cantor normal form. epsilon 0

from z3 import *
U = Datatype('Univ')
U.declare('cons', ('car', U), ('cdr', U))
U.declare('bool', ('bool_val', BoolSort()))
U.declare('real', ('real_val', RealSort()))
U.declare('string', ('string_val', StringSort()))
U.declare('int', ('int_val', IntSort()))
U = U.create()
print(U.cons(U.int(1), U.nil))

# Can I add operations?
#U.__add__ = lambda self, other: U.int(self, other)
car = U.car
cdr = U.cdr
cons = U.cons
nil = U.nil

def atom(x):
    return Not(U.is_cons(x))
def consp(x):
    return U.is_cons(x)
def integerp(x):
    return U.is_int(x)

Function("order", U, U, BoolSort())
= RecursiveFunc()
  And(e0_ordinalp(car(x)), car(x) != U.int(0), 
      Not(e0-ord-<(car(x), cadr(x)))),
  And(integerp(x), U.int_val(x) >= 0)))

https://www.cs.utexas.edu/users/moore/publications/index.html moore publications https://www.cs.utexas.edu/users/moore/publications/gentle-intro-to-acl2-programming.html gentl introduction CLI veirifed stack - linker loader? https://www.cs.utexas.edu/users/moore/best-ideas/piton/index.html piton verified os stack https://www.cs.utexas.edu/ftp/boyer/cli-reports/022.pdf


Computer-Aided Reasoning: An Approach Computer-Aided Reasoning: ACL2 Case Studies A computational logic handbook - book on ACL2

“the little prover” is about a simplified versio nof acl2

https://www.cs.utexas.edu/users/moore/acl2/manuals/current/manual/?topic=ACL2____The_02Proof_02of_02the_02Associativity_02of_02App the proof of the associativty of app

https://www.cs.utexas.edu/users/moore/acl2/v8-5/combined-manual/?topic=ACL2____GUARD guards

https://www.cs.utexas.edu/users/moore/acl2/v8-5/combined-manual/?topic=ACL2____STATE von nuemmanesque state. Reads to me like th state monad


(declare (xargs :guard (g x y) :measure (m x y))) xargs let’s you state guard and measure

Hmm. Totalize in the logic, but have partiality guard rails at metalevel? A global notion of ordinal and well founded?

; actually make a datatype? But then not extendable. Maybe that's ok.
(declare-sort Atom)
(declare-datatype Value
  (cons (car Value) (cdr Value))
  (number Int)
  (real Real)
  (string String)
  (atom Atom)
(declare-sort Value)
(declare-fun cons (Value Value) Value)
(define-const nil Value)
(declare-fun car (Value) Value)
(assert (forall ((x Value)) (= (car x) (ite (is-cons x) ()  nil)))

Other Lisps



Lisp in easy pieces


Gambit compiling to javascript

old gist discussion

You need to compile the javascript runtime special

make _gambit.js # This command finally build the needed Gambit Javascript runtime
LD_LIBRARY_PATH=..: ../gsc/gsc -:~~bin=../bin,~~lib=../lib,~~include=../include -f -target js  -prelude "(define-cond-expand-feature|enable-type-checking|)(define-cond-expand-feature|disable-auto-forcing|)(define-cond-expand-feature|enable-sharp-dot|)(define-cond-expand-feature|enable-bignum|)(define-cond-expand-feature|enable-ratnum|)(define-cond-expand-feature|enable-cpxnum|)(define-cond-expand-feature|disable-smp|)(##include\"../lib/header.scm\")" -o _gambit.js ../lib/_univlib.scm
cp _gambit.js /c/tools/gambit-c/lib/_gambit.js # This command publish the Javascript runtime to the installed Gambit-C.

Hmm. make_gambit does this stuff? https://gitter.im/gambit/gambit?at=5bc7fb95435c2a518ec448d1

This final concatenation step is bizarre. Hmm wasn’t necessayr on minimal example in gitter above

% cat test.scm
(println "hello!")
% gsc/gsc -:=. -target js -exe test.scm
% ./test

contrib/try repl gambit in the browwser experiment using emscripten. Different from try.scheme.org using universal backend


tinylisp 99 lines of C. pretty cool https://news.ycombinator.com/item?id=32100035 - interrsting comment about generic multiple dispatch. Is there a subsumption indexing issue here?

build your own lisp

https://github.com/nojb/ocs-ng ocaml scheme

Papers on writing virtual machines for scheme? 3 implementation models

scheme bibliography

State of Scheme to Javascript The Ribbit pathway seems interesting. Or maybe rather the Gambit pathway scheme interp in browser

“Let’s Build a Hygienic Macro Expander” by Matthew Flatt Macro scope. So de bruijn indices are a way of dealing with some of the troubles of binding forms. But here I think he’s saying that de bruijn indices are the wrong abstraction.

Scope sets Syntax objects are paired with scope sets. type syntax = Sexp.t * Int.Set.t

syntax_of_datum : Sexp.t -> syntax =

HUh that was an interesting trick. Use ref () and physical equality for gensym. Binding as Sets of Scopes Notes on a new model of macro expansion for Racket

https://github.com/jcubic/lips lips - a scheme in javascript

https://ianthehenry.com/posts/janet-game/the-problem-with-macros/ interesting blog post cross stage persistence. How does one serialize a function? http://www.nhplace.com/kent/Papers/Technical-Issues.html why a lisp 2 i guess

https://turtleware.eu/ common lisp essay

https://www.cs.utah.edu/plt/publications/macromod.pdf composable and compileable macros - you want in whe?

Oleg of course. http://okmij.org/ftp/Scheme/

  • Has an implementation of shift reset delimitted contianutations

Macro examples:

  • Making a pattern matcher might be kind of interesting exercise.
  • making your own short circuiting “or” or other non call by value constructs. In some sense this is defining a new interpreter back into scheme?
  • making your own binding forms or introducing variables into the environment

https://www.cs.utexas.edu/ftp/garbage/cs345/schintro-v14/schintro_toc.html - an intorudcition to scheme and it’s impllementation

https://ds26gte.github.io/tyscheme/index-Z-H-1.html teach yourself scheme in fixnum days

In particular the extneded examples section is kind of interesting https://www.scheme.com/tspl4/examples.html#./examples:h0

The scheme objects technique. Let bindings are mutable. Isn’t that nuts. Let’s you do all kinds of shenanigans that you could do in ocaml with refs, but would be unlikely to.

let make_counter () = let state = mk_ref 0 in fun () -> let state := !state + 1 in !state

amb is genuinely surprising. It isn’t even that it’s a macro. Its callcc that makes it possible call/cc for break statements.

continuations https://www.ps.uni-saarland.de/~duchier/python/continuations.html call/cc let’s you pretend you were writing in CPS the whole time.

  • call/cc for break statements
  • call/cc for coroutines
  • search
  • amb

https://felleisen.org/matthias/7480-s21/lectures.html history of PL. interesting discussion of hygienic macros at the least https://www.sweetjs.org/ - hygienic macros


What is the deal.

https://github.com/namin/inc incremental compilter construction

Hmm a scheme in the broser biwacheme chicken -> C -> wasm gambit has a backend racket has ab ackend https://www.reddit.com/r/scheme/comments/fhvuwl/the_best_uptodate_and_mature_schemejavascript/

Like livecode.io right clojurescript. But that’s clojure

https://github.com/webyrd/faster-miniKanren https://github.com/jasonhemann/microKanren/blob/master/microKanren.scm

<script src="https://www.biwascheme.org/release/biwascheme-0.7.1-min.js">
   (console-log "Hello, world!")

<div id="bs-console">

<script type="text/biwascheme">
  (print "Hello, world!")
  (print (current-date))
  (console-log "ok.")
  (load "/assets/microkanren.scm")
  (print (call/goal (call/fresh (lambda (q)  (== q 1) ))))
  (print (call/run (lambda (q)
  (call/fresh (lambda (r)
  (conj (== q `(,r ,r))  (== r 1) ))))))
  (define-macro (fizz x)
    `(buzz ,x))
  (print (macroexpand '(fizz 3)))
  (print (macroexpand '(Zzz f)))
  (print (macroexpand '(conj+ x y z)))
  (print (macroexpand '(fresh (x y) g1 g2)))
  (print (macroexpand '( conde  ((x y) (p q))  )))
  (print (run 1 (q) (== q 1)) )
  (print (run 2 (q)
           ((== q 1))
           ((== q 2)))))
  (print (run 2 (x)
        (fresh (p q)
         (== x `(,p ,q))
           ((== q 1) (== p q))
           ((== q 2))))))
    (print (run 2 (q p)
           ((== q 1) (== p q))
           ((== q 2)))))

    (print (macroexpand '(run 2 (q p)
    ( print (macroexpand-1 '(conde
           ((== q 1) (== p q))
           ((== q 2)))))