Errata from the final draft of R7RS:
- In Section 3.1 (Variables, syntactic keywords, and regions), the claim in paragraph 2 that all variable binding constructs can be explained in terms of lambda does not apply to top-level bindings.
- Section 7.1.5 (Transformers) doesn't make it clear that a top-level syntax-rules pattern must be a list pattern, not a vector pattern or an identifier pattern.
- In Section 6.3 (Booleans), the procedure boolean=? is defined to return #t if the arguments are all booleans and are either all #t or all #f. The words "are all booleans and" incorrectly suggest that the value is #f if at least one argument is not a boolean. In fact it is an error to apply boolean=? to non-booleans.
- In Section 4.1.2 (Literal expressions), the examples '# and # should be '#\a and #\a respectively.
- In section 7.1.1 (Lexical structure), the escape sequence \| is not shown as permitted in strings. The list in Section 6.7 shows that it is equivalent to |. This makes the same escape sequences valid in both strings and symbols.
- In Section 7.3 (Derived expression types), the syntax-rules definition of case is incorrect; the fourth syntax rule must be moved after the fifth to avoid an improper match against the fourth rule when => is present. Here is the correct version:
(define-syntax case
(syntax-rules (else =>)
((case (key ...)
clauses ...)
(let ((atom-key (key ...)))
(case atom-key clauses ...)))
((case key
(else => result))
(result key))
((case key
(else result1 result2 ...))
(begin result1 result2 ...))
((case key
((atoms ...) => result))
(if (memv key '(atoms ...))
(result key)))
((case key
((atoms ...) => result)
clause clauses ...)
(if (memv key '(atoms ...))
(result key)
(case key clause clauses ...)))
((case key
((atoms ...) result1 result2 ...))
(if (memv key '(atoms ...))
(begin result1 result2 ...)))
((case key
((atoms ...) result1 result2 ...)
clause clauses ...)
(if (memv key '(atoms ...))
(begin result1 result2 ...)
(case key clause clauses ...)))))
- In Section 7.1.1, the lexical rule <special initial> incorrectly omits @.
- Bibliographic reference ![13] should link to SRFI 4.
- In section 4.2.2, add "interleaving evaluations with assignments" to the definition of letrec*. Replace the meaningless example with this:
;; Returns the arithmetic, geometric, and
;; harmonic means of a nested list of numbers
(define (means ton)
(letrec*
((mean
(lambda (f g)
(f (/ (sum g ton) n))))
(sum
(lambda (g ton)
(if (null? ton)
(+)
(if (number? ton)
(g ton)
(+ (sum g (car ton))
(sum g (cdr ton)))))))
(n (sum (lambda (x) 1) ton)))
(values (mean values values)
(mean exp log)
(mean / /))))
Note that evaluating (means '(3 (1 4))) returns three values: 8/3, 2.28942848510666 (approximately), and 36/19.
- In section 7.1.5, add <bytevector> to the alternatives for the <pattern datum> rule.
- Section 1.3.4 refers to "the initial environment" containing *, which is not true for programs. It should refer instead to "an environment containing the base library".
- An example in section 5.3.3 refers to integer-sqrt instead of exact-integer-sqrt.
- In section 4.3.1, the body of a let-syntax expression is said to contain "one or more definitions"; it should be "zero or more definitions".
- In section 6.6 for char-upcase/downcase/foldcase and 6.7 for string-upcase/downcase/foldcase the reader is referred to UAX #29, but it should be TR #44.
- In section 7.3, the definition of case-lambda should use letrec-syntax, not let-syntax.
- In section 4.3.2 (but not in 7.1.3), change <literal> to <pattern literal> to avoid confusion.
- In section 4.2.7 in the definition of guard, change the first reference to raise-continuable to raise. This was an editorial error that introduced an unwanted incompatibility with the R6RS definition. See ticket #431, which should have fixed this but didn't (mea culpa).