[racket] Style or and/define
On Tue, Jun 11, 2013 at 10:39 AM, Клочков Роман <kalimehtar at mail.ru> wrote:
> You may use srfi/2 http://docs.racket-lang.org/srfi-std/srfi-2.html
>
> (require srfi/2)
> (define (get-x-spot char-width)
> (and-let*
> ([char-width]
> [dc (get-dc)]
>
> [style (or (send (get-style-list) find-named-style "Standard")
> (send (get-style-list) find-named-style "Basic"))])
> (let*-values ([(fnt) (send style get-font)]
> [(xw _1 _2 _3) (send dc get-text-extent "x" fnt)])
> (+ left-padding (* xw char-width))))))
This is cool. It seems to support some idioms very well. Guarding
arguments, fetching some unstable external thing. As long as #f is an
acceptable answer for everything but the correct solution.
In my CL code I would tend to do this through guard expressions which
return early or throw an exception. I find that some of the things
that cause guards to fail are things the user needs to know about,
like misconfigurations, so exceptions are warranted. I also find that
it is clearer to separate the cruft of managing guards from the main
task of the function.
In Racket I still like to separate guards into a distinct conditional,
like if, when, unless. The nesting works well for me. I like a little
bit of it. I would tend to do something like this:
(define (get-x-spot char-width)
;; Get my environment into nice short variables.
(let ([dc (get-dc)]
[sl (get-style-list)])
;; Guard
(when (and char-width dc sl)
(let ([style (or (send sl find-named-style "Standard")
(send sl find-named-style "Basic"))])
;; Another guard
(when style
(let*-values ([(fnt) (send style get-font)]
[(xw _1 _2 _3) (send dc get-text-extent "x" fnt)])
(+ left-padding (* xw char-width))))))))
Erik.
>
>
> Вторник, 11 июня 2013, 18:49 +02:00 от Laurent <laurent.orseau at gmail.com>:
>
> When I see what Robby is forced to write when following the Style:
> https://github.com/plt/racket/commit/09d636c54573522449a6591c805b38f72b6f7da8#L4R963
>
> I cannot help but think that something is wrong somewhere (it may not be the
> Style, and in case it wasn't clear I'm certainly not criticizing Robby's
> code).
> Using `let' and `and' instead, although being a bit better since it avoids
> all the [else #f], is not that big an improvement:
>
> (define (get-x-spot char-width)
> (and
> char-width
> (let ([dc (get-dc)])
> (and
> dc
> (let ([style (or (send (get-style-list) find-named-style "Standard")
> (send (get-style-list) find-named-style "Basic"))])
> (and
> style
> (let*-values ([(fnt) (send style get-font)]
> [(xw _1 _2 _3) (send dc get-text-extent "x" fnt)])
> (+ left-padding (* xw char-width)))))))))
>
>
> Actually I think here the right thing to do might be to allow for internal
> definitions inside `and':
>
> (define (get-x-spot char-width)
> (and char-width
> (define dc (get-dc))
> dc
> (define style (or (send (get-style-list) find-named-style "Standard")
> (send (get-style-list) find-named-style "Basic")))
> style
> (define fnt (send style get-font))
> (define-values (xw _1 _2 _3) (send dc get-text-extent "x" fnt))
> (+ left-padding (* xw char-width))))
>
>
> Isn't it *much* more readable? (shorter, avoid rightward drift, less parens,
> vertical alignment)
>
> Since it's not the first time I find the need for such internal definitions
> in `and', maybe this is something to consider for future addition to Racket?
> Or have some people already identified some problems with this idea?
>
> I've played a bit with it if you want to try by your own:
> https://gist.github.com/Metaxal/5758394
>
> (not sure I got it all good with syntax-parse though)
>
> Laurent
>
> ____________________
> Racket Users list:
> http://lists.racket-lang.org/users
>
>
>
> --
> Роман Клочков
>
> ____________________
> Racket Users list:
> http://lists.racket-lang.org/users
>
--
Erik Pearson
Adaptations
;; web form and function