I wrote a macro once to `flatten' this kind of idiom, because I was
writing code where I was doing a *lot* of this. It lets you mix
let-style bindings with tests that throw an error if they fail.
Warning: there is almost certainly a better way to write it (it is among
the first `real' macros I wrote in Racket, and I haven't looked at it in
a long time), but maybe you could adapt it to do what you're looking
for.
(define-syntax assertive-let*
; let* that simplifies intermixing of assertions and bindings
(syntax-rules (?)
; base case: sequentially evaluate body forms, as in a regular let
[(assertive-let* () body ...)
(begin body ...)]
; when head form only has two expressions, just introduce a binding
[(assertive-let* ([name expr] fs ...) body ...)
(let ((name expr))
(assertive-let* (fs ...) body ...))]
; when head form has three expressions beginning with "?", it's an
; assertion: expr evaluates or an error with message msg is raised
[(assertive-let* ([? expr msg] fs ...) body ...)
(if (not expr)
(error msg)
(assertive-let* (fs ...) body ...))]
; otherwise, a three-place head form simultaneously makes an
; assertion and introduces a binding for the result of the assertion
[(assertive-let* ([name expr msg] fs ...) body ...)
(let ([name expr])
(if (not name)
; TODO: there's probably something more sensible to do here
; than just throwing an error, e.g., return a default value?
(error msg)
(assertive-let* (fs ...) body ...)))]))
Then the code you linked to becomes something like (untested):
(define/private (get-x-spot char-width)
(assertive-let*
([? char-width "Non-false char-width is required"]
[dc (get-dc) "Failed to get dc"]
[style (or (send (get-style-list) find-named-style "Standard")
(send (get-style-list) find-named-style "Basic"))
"Failed to get a style"]
[fnt (send style get-font)])
(define-values (xw _1 _2 _3) (send dc get-text-extent "x" fnt))
(+ left-padding (* xw char-width))))
Though note that this code will now raise errors instead of returning #f
when one of the earlier bindings fails.
Hope that's helpful!
--
Best,
Richard
____________________
Racket Users list:
http://lists.racket-lang.org/users
