I realize it's not ideal Racket, but I've gotten a lot of mileage out of
the following:
;;----------------------------------------------------------------------
(define prefix-for-say (make-parameter ""))
(define-syntax (say stx)
(syntax-case stx ()
[(say a ...)
#'(displayln (~a (prefix-for-say) a ...))]
))
(define-syntax (__WHERE:__ stx) ; Note the ':'
(with-syntax ((fpath (syntax-source stx))
(line (syntax-line stx)))
(syntax-case stx ()
[_ #'(~a "file:" fpath " (line:" line "): ")])))
;;----------------------------------------------------------------------
I then sprinkle my code with:
(say __WHERE:__ "the thing was: " thing ", and it had " num-owners " owners)
which gets me output like:
file:/Users/dstorrs/some/path/to/file.rkt (line:131): the thing was:
<thing>, and it had 7 owners
For threading, I have this:
(define/contract (threaded thnk)
(-> (-> any) any)
(define current-prefix (prefix-for-say))
(define thread-label
(cond [(empty-string? current-prefix) (~a (rand-val "thread") ": ")]
[else (~a (rand-val "thread") " / " current-prefix)]))
(parameterize ([prefix-for-say thread-label])
(thread (thunk (begin0 (thnk) (sleep 0)))))) ; sleep 0 suggests that
the thread run now
(define (foo x) (threaded (thunk (bar x))))
(define (bar x) (threaded (thunk (say x))))
(foo "hello, world")
And I get output like this:
thread-261855 / thread-861893: hello, world
There are improvements that I could make to this -- for example, make
'threaded' a macro so you don't need to pass a thunk, then special-case it
so that thunks were executed so that existing code continues to work. Also,
'say' could have an option that lets you use println instead of displayln,
or automatically map ~v over arguments. And obviously this system doesn't
take advantage of Racket's hierarchical logging infrastructure. That said,
it's simple and it works really well for me.
On Tue, Aug 7, 2018 at 11:22 AM, Sean Kanaley <[email protected]> wrote:
> Thanks, Neil. Unfortunately #2 didn't work. The historically written
> define-logger calls like log-something-debug overwrites the "foo" topic
> with "something". If only they were appended together. #3 can't intercept
> logs of define-logger loggers (if a woodchuck could...) even with
> topic/level hardcoded because the #:logger keyword is also needed. So it
> seems like the usage would be a series of nested with-intercepted-loggings,
> one for each define-logger. Lastly, #1 is what I'm trying to avoid :)
>
> To be fair, either #1 or #3 is fully *practical*, but the Racketeer in me
> would prefer to fully automate this. The "proper" solution seems to require
> a Racket change.
>
> Is #3 a bug? Or I might be getting confused at the timing with the
> define-logger = current-logger being used to write the log and the
> interceptor reverting to the previous logger, but the code that seems like
> it should work here:
>
> #lang racket/base
> (require racket/logging)
>
> (define-logger x)
>
> ;works, no define-logger topic though
> (let ([thread-id "hurray"])
> (with-intercepted-logging
> (λ (v)
> (printf "~a: ~a~n" thread-id (vector-ref v 1)))
> (λ ()
> (log-debug "we interrupt this program to bring you this important
> message"))
> 'debug))
>
> ;doesn't work
> (let ([thread-id "hurray"])
> (with-intercepted-logging
> (λ (v)
> (printf "~a: ~a~n" thread-id (vector-ref v 1)))
> (λ ()
> (log-x-debug "we interrupt this program to bring you this important
> message"))
> 'debug))
>
> ;works but messy for many define-loggers
> (let ([thread-id "hurray"])
> (with-intercepted-logging
> #:logger x-logger
> (λ (v)
> (printf "~a: ~a~n" thread-id (vector-ref v 1)))
> (λ ()
> (log-x-debug "we interrupt this program to bring you this important
> message"))
> 'debug))
>
> --
> You received this message because you are subscribed to the Google Groups
> "Racket Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
>
--
You received this message because you are subscribed to the Google Groups
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
