This page is part of the web mail archives of SRFI 110 from before July 7th, 2015. The new archives for SRFI 110 contain all messages, not just those from before July 7th, 2015.
On 3/7/13, Jos Koot <jos.koot@xxxxxxxxx> wrote: > Nice idea, but how to deal with my frequent copy/paste actions in the > definitionss window of DrRacket? > > For sexprs, copy/paste is easy in the definitions window of DrRacket. Just > copy/paste and reindent. How easy would this be with sweet expressions? Can > we copy/paste at a choosen level of indentation? > > Say I have a definition > > (define (f a b c) > def/expr ... > last-expr) > > When I am not sure that function f produces correct results, I modify it > temporarily as: > > (define (f a b c) (let ((r (let() > def/expr ... > last-expr))) (printf "result of proc f: ~s~n~n" r) r)) > > Notice that here the indenting is deliberately wrong. Using the temporal > "let ((r" form I don't care about the indenting. After seeing what is going > on I use ctl Z in order to remove the extra let-form and the print-form. > How > would this be with sweet expressions? Normally for myself I keep a probe() definition handy: define probe(x) display "probe: " \\ write x \\ (newline) Then whatever I need to probe, I just convert: define (f a b c) def/expr ... last-expr to: define (f a b c) $ probe $ let () def/expr ... last-expr -- SUBLIST rocks. probe can also be a macro: define-syntax probe ! syntax-rules () ! \\ ! ! probe x ! ! \\ ! ! let ! ! $ v $ x ! ! write 'x ; show the expression ! ! display ": " ! ! write v ! ! v > > Wish you good luck. A sweet expressions language is certainly possible in > Racket. If i can be useful for (parts of) the implementation,mail me > privately. Yes, although I worry that it'll require a #lang declaration rather than our preferred #!sweet form... Is the latter possible in Racket? AFAIK all Racket languages require a #lang form, I'm not at all certain it's possible to use #!sweet. (Note: I haven't gone digging through Racket docs, so correct me if I'm wrong here) Still, not a big problem, and it seems that Racket's current #lang will do quite well. Sincerely, AmkG