Lisp/Schemeの準クォートの展開アルゴリズムは Quasiquotation in Lisp のAppendix A に書かれている。で生成される展開コードは効果的じゃない、最適化されたものが “Common Lisp the Language, 2nd Edition” の Appendix C. Backquote に載っている、とも書かれている。


;;; Common Lisp backquote implementation, written in Common Lisp.
;;; Author: Guy L. Steele Jr. Date: 27 December 1985
;;; Tested under Symbolics Common Lisp and Lucid Common Lisp.
;;; This software is in the public domain.

;;; $ is pseudo-backquote and % is pseudo-comma. This makes it
;;; possible to test this code without interfering with normal
;;; Common Lisp syntax.

;;; The following are unique tokens used during processing.
;;; They need not be symbols; they need not even be atoms.

(define *bq-clobberable* (gensym))
(define *bq-quote-nil* (list 'quote ()))

;;; Reader macro characters:
;;; $foo is read in as (BACKQUOTE foo)
;;; %foo is read in as (#:COMMA foo)
;;; %@foo is read in as (#:COMMA-ATSIGN foo)
;;; is read in as (#:COMMA-DOT foo)
;;; where #:COMMA is the value of the variable *COMMA*, etc.

;;; BACKQUOTE is an ordinary macro (not a read-macro) that
;;; processes the expression foo, looking for occurrences of
;;; #:COMMA, #:COMMA-ATSIGN, and #:COMMA-DOT. It constructs code
;;; in strict accordance with the rules on pages 349-350 of
;;; the first edition (pages 528-529 of this second edition).
;;; It then optionally applies a code simplifier.

;(set-macro-character #\`
; (lambda (stream char)
; ;(declare (ignore char))
; (list 'quasiquote (read stream))))
;(set-macro-character #\,
; (lambda (stream char)
; ;(declare (ignore char))
; (let1 c (read-char stream)
; (case c
; (#\@
; (list 'unquote-splicing (read stream)))
; (#\.
; (list 'unquote-dot (read stream)))
; (t (unread-char c stream)
; (list 'unquote (read stream)))))))

;;; If the value of *BQ-SIMPLIFY* is non-NIL, then BACKQUOTE
;;; processing applies the code simplifier. If the value is NIL,
;;; then the code resulting from BACKQUOTE is exactly that
;;; specified by the official rules.

(define-macro (quasiquote x)
(bq-completely-process x))

;;; Backquote processing proceeds in three stages:
;;; (1) BQ-PROCESS applies the rules to remove occurrences of
;;; #:COMMA, #:COMMA-ATSIGN, and #:COMMA-DOT corresponding to
;;; this level of BACKQUOTE. (It also causes embedded calls to
;;; BACKQUOTE to be expanded so that nesting is properly handled.)
;;; Code is produced that is expressed in terms of functions
;;; #:BQ-LIST, #:BQ-APPEND, and #:BQ-CLOBBERABLE. This is done
;;; so that the simplifier will simplify only list construction
;;; functions actually generated by BACKQUOTE and will not involve
;;; any user code in the simplification. #:BQ-LIST means LIST,
;;; but indicates places where "%." was used and where NCONC may
;;; therefore be introduced by the simplifier for efficiency.
;;; (2) BQ-SIMPLIFY, if used, rewrites the code produced by
;;; BQ-PROCESS to produce equivalent but faster code. The
;;; additional functions #:BQ-LIST* and #:BQ-NCONC may be
;;; introduced into the code.
;;; (3) BQ-REMOVE-TOKENS goes through the code and replaces
;;; #:BQ-LIST with LIST, #:BQ-APPEND with APPEND, and so on.
;;; #:BQ-CLOBBERABLE is simply eliminated (a call to it being
;;; replaced by its argument). #:BQ-LIST* is replaced by either
;;; LIST* or CONS (the latter is used in the two-argument case,
;;; purely to make the resulting code a tad more readable).

(define (bq-completely-process x)
(bq-simplify (bq-process x)))

(define (bq-process x)
(cond ((not (pair? x))
(list 'quote x))
((eq? (car x) 'quasiquote)
(bq-process (bq-completely-process (cadr x))))
((eq? (car x) 'unquote) (cadr x))
((eq? (car x) 'unquote-splicing)
(error ",@~S after `" (cadr x)))
((eq? (car x) 'unquote-dot)
(error ",.~S after `" (cadr x)))
(else (let loop ((p x)
(q '()))
(if (not (pair? p))
(cons 'append
(nreconc q (list (list 'quote p))))
(if (eq? (car p) 'unquote)
(begin (unless (null? (cddr p)) (error "Malformed ,~S" p))
(cons 'append
(nreconc q (list (cadr p)))))
(begin (when (eq? (car p) 'unquote-splicing)
(error "Dotted ,@~S" p))
(when (eq? (car p) 'unquote-dot)
(error "Dotted ,.~S" p))
(loop (cdr p)
(cons (bracket (car p)) q)))))))))

;;; This implements the bracket operator of the formal rules.

(define (bracket x)
(cond ((not (pair? x))
(list 'list (bq-process x)))
((eq? (car x) 'unquote)
(list 'list (cadr x)))
((eq? (car x) 'unquote-splicing)
(cadr x))
((eq? (car x) 'unquote-dot)
(list *bq-clobberable* (cadr x)))
(else (list 'list (bq-process x)))))

;;; This auxiliary function is like MAPCAR but has two extra
;;; purposes: (1) it handles dotted lists; (2) it tries to make
;;; the result share with the argument x as much as possible.

(define (maptree fn x)
(if (not (pair? x))
(fn x)
(let ((a (fn (car x)))
(d (maptree fn (cdr x))))
(if (and (equal? a (car x)) (equal? d (cdr x)))
(cons a d)))))

;;; This predicate is true of a form that when read looked
;;; like %@foo or

(define (bq-splicing-frob x)
(and (pair? x)
(or (eq? (car x) 'unquote-splicing)
(eq? (car x) 'unquote-dot))))

;;; This predicate is true of a form that when read
;;; loocked like %@foo or or just place %foo.

(define (bq-frob x)
(and (pair? x)
(or (eq? (car x) 'unquote)
(eq? (car x) 'unquote-splicing)
(eq? (car x) 'unquote-dot))))

;;; The simplifier essentially looks for calls to #:BQ-APPEND and
;;; tries to simplify them. The arguments to #:BQ-APPEND are
;;; processed from right to left, building up a replacement form.
;;; At each step a number of special cases are handled that,
;;; loosely speaking, look like this:
;;; (APPEND (LIST a b c) foo) U> (LIST* a b c foo)
;;; provided a, b, c are not splicing frobs
;;; (APPEND (LIST* a b c) foo) U> (LIST* a b (APPEND c foo))
;;; provided a, b, c are not splicing frobs
;;; (APPEND (QUOTE (x)) foo) U> (LIST* (QUOTE x) foo)
;;; (APPEND (CLOBBERABLE x) foo) U> (NCONC x foo)

(define (bq-simplify x)
(if (pair? x)
(let ((x (if (eq? (car x) 'quote)
(maptree bq-simplify x))))
(if (not (eq? (car x) 'append))
(bq-simplify-args x)))

(define (bq-simplify-args x)
(let loop ((args (reverse (cdr x)))
(result '()))
(if (not (null? args))
(loop (cdr args)
(cond ((not (pair? (car args)))
(bq-attach-append 'append (car args) result))
((and (eq? (caar args) 'list)
(not (any bq-splicing-frob (cdar args))))
(bq-attach-conses (cdar args) result))
((and (eq? (caar args) 'list*)
(not (any bq-splicing-frob (cdar args))))
(reverse (cdr (reverse (cdar args))))
(bq-attach-append 'append
(car (last (car args)))
((and (eq? (caar args) 'quote)
(pair? (cadar args))
(not (bq-frob (cadar args)))
(not (cddar args)))
(bq-attach-conses (list (list 'quote
(caadar args)))
((eq? (caar args) *bq-clobberable*)
(bq-attach-append 'append! (cadar args) result))
(else (bq-attach-append 'append
(car args)

(define (null-or-quoted x)
(or (null? x) (and (pair? x) (eq? (car x) 'quote))))

;;; When BQ-ATTACH-APPEND is called, the OP should be #:BQ-APPEND
;;; or #:BQ-NCONC. This produces a form (op item result) but
;;; some simplifications are done on the fly:
;;; (op '(a b c) '(d e f g)) U> '(a b c d e f g)
;;; (op item 'nil) U> item, provided item is not a splicable frob
;;; (op item ’nil) U>(op item), if item is a splicable frob
;;; (op item (op a b c)) U> (op item a b c)

(define (bq-attach-append op item result)
(cond ((and (null-or-quoted item) (null-or-quoted result))
(list 'quote (append (safe-cadr item) (safe-cadr result))))
((or (null? result) (equal? result *bq-quote-nil*))
(if (bq-splicing-frob item) (list op item) item))
((and (pair? result) (eq? (car result) op))
(list* (car result) item (cdr result)))
(else (list op item result))))

;;; The effect of BQ-ATTACH-CONSES is to produce a form as if by
;;; `(LIST* ,@items ,result) but some simplifications are done
;;; on the fly.
;;; (LIST* 'a 'b 'c 'd) U> '(a b c . d)
;;; (LIST* a b c 'nil) U> (LIST a b c)
;;; (LIST* a b c (list* d e f g)) U> (LIST* a b c d e f g)
;;; (LIST* a b c (list d e f g)) U> (LIST a b c d e f g)

(define (bq-attach-conses items result)
(cond ((and (every null-or-quoted items)
(null-or-quoted result))
(list 'quote
(append (map cadr items) (cadr result))))
((or (null? result) (equal? result *bq-quote-nil*))
(cons 'list items))
((and (pair? result)
(or (eq? (car result) 'list)
(eq? (car result) 'list*)))
(cons (car result) (append items (cdr result))))
(else (cons 'list* (append items (list result))))))
  • リーダーマクロの辺りはコメントアウトした
  • 元のコードでは backquote, comma, comma-atsign, comma-dot という名前だが、Scheme風に quasiquote, unquote, unquote-splicing, unquote-dot と変更した
  • 元のコードではいったん識別用のシンボルを埋め込んで後で変換しているが、最初から埋め込むようにした
  • do ループを使用していたところをnamed-letによるループに変更
  • Common Lispでは当然ながら nil をリストの終端と false の両方の意味で使うけど、Schemeでは使い分けないといけない


;; Reverse list and concatenate tail destructively.
(define (nreconc ls tail)
(let1 top (reverse! ls)
(set-cdr! ls tail)

(define (safe-car x)
(if (null? x)
(car x)))
(define (safe-cdr x)
(if (null? x)
(cdr x)))
(define (safe-cadr x) (safe-car (safe-cdr x)))
  • nreconc はリストをひっくり返して、その末尾に追加する関数
    • こんな関数があることが驚きだ…
  • Common Lispでは carcdrnil を渡すと nil が返るけど、Schemeではエラーが発生するので、nil チェックをするものを用意してやる


(print (bq-completely-process '(x ,y ,@z)))
;; => (list* 'x y z)



'(x . ,y)

(list* 'x y)