> thank you for creating SRFI-245! In Guile I enjoy it very much to be > able to intersperse definitions and expressions.
+1 > The specification in SRFI-245 looks a bit too strict for me, though: > > > It is an error for the evaluation of any expression or of the > > right-hand side of any definition in the sequence of definition or > > expressions to require knowledge of the value of a variable whose > > definition is to the right of it. > > > Do I understand it correctly that this would mean that this would not > work, because > > (define (using-later-procedure) > (define x (y)) > (define (y) #t) > x) > > Or that it would stop to work if I interspersed logging messages? > > (define (using-later-procedure) > (define x (y)) > (display x)(newline) > (define (y) #t) > x) > > Both of these work in Guile, because Guile treats expressions before the > final definition as implicit definitions: > > (define (using-later-procedure) > (define x (y)) > (define _1 (display x)) > (define _2 (newline)) > (define (y) #t) > x) > > For details, see > https://www.gnu.org/software/guile/manual/html_node/Internal-Definitions.html > > > (let () > > (define a 1) > > (foo) > > (define b 2) > > (+ a b)) > > > > is equivalent to > > > > (let () > > (letrec* ((a 1) (_ (begin (foo) #f)) (b 2)) > > (+ a b))) > > > It may be useful to enforce an ordering between expressions and blocks > of definitions, though, to ensure that logging code runs before the > later definitions. OT: It is dubious to log the return of procedure: if the procedure is buggy, the logs from initial deployment are all useless. In fact, outside the human readable message, no value should more complex than a literal number, boolean, symbol, or even a string. The only exception, I can think of, is time zone aware. My rationales for the SRFI specification are: 1. Re use-before-definition: I can see, or think for, no rationales that mention usability, ease of programming, debugging, maintenance. 2. The description of the interpretation, or compilation of use-before-definition is complicated; 3. Having the body of lambda behave the way the body of a program behave is easy > Do I understand the letrec* implementation in the SRFI correctly that it > would enforce the ordering of expressions before the directly following > definition? Thanks, Arne, for chiming in, I hope, again, you take part, once more, in SRFI process. Best, and due regards, > > Best wishes, > Arne > -- > Unpolitisch sein > heißt politisch sein, > ohne es zu merken. > draketo.de
