This looks quite a bit like you're trying to implement tracing, which is like logging, but instead of emitting messages associated with *points* in time, you emit messages for *ranges* of time. Rust has an excellent library <https://docs.rs/tracing/0.1.28/tracing/> for tracing. Perhaps that will provide some good inspiration?
On Friday, September 3, 2021 at 8:30:18 AM UTC-7 [email protected] wrote: > On Thu, Sep 2, 2021 at 5:32 PM Sorawee Porncharoenwase < > [email protected]> wrote: > >> Thoughts: >> >> - Perhaps the logger should be optional. The default value would be >> (current-logger). >> - The event name (like on-complete) could also be optional. The >> default would be the source location of the macro invocation site >> - Instead of “time: ~a”, I think it would be nice to support many >> “pairs”, which are formatted like raise-arguments-error. >> >> Example: >> >> (define (on-complete x) >> (log-test-debug "I'm in on-complete") >> x) >> >> (with-log () >> 1) >> >> (with-log (#:logger test-debug >> #:msg "on-complete" >> ["time" (current-seconds)]) >> (on-complete (person 'bob))) >> >> would output: >> >> test: entering test.rkt:5:1 >> test: exiting test.rkt:5:1 >> result: 1 >> 1 >> test: entering on-complete >> time: 123 >> test: I'm in on-complete >> test: exiting on-complete >> time: 124 >> result: (person 'bob) >> (person 'bob) >> >> >> > First of all, thank you. I like the idea of the event name being optional > and the logger defaulting but I'm not keen on the syntax. It's very > verbose for something that might occasionally be wrapped around a single > line of code. The raise-arguments-error formatting would be a nice default > but I prefer to give the option to use a format string if you want > something different. > > >> On Thu, Sep 2, 2021 at 2:06 PM Martin DeMello <[email protected]> >> wrote: >> >>> I do like the second form better, especially since the actual code being >>> run is not obscured by simply being the last argument to a long log >>> function. >>> >> > Cool. I'll move towards that. > > >>> martin >>> >>> On Thu, Sep 2, 2021 at 1:55 PM David Storrs <[email protected]> wrote: >>> >>>> I often find that for debugging I want to see a log message saying "I'm >>>> about to do X" followed by X followed by "I'm done with X" and I want it >>>> to >>>> return the result of X. >>>> >>>> I wrote this macro and posted it to the package server: >>>> https://pkgs.racket-lang.org/package/log-bracketed >>>> >>>> In retrospect, the syntax is bad and I should change it. Can anyone >>>> suggest something better? >>>> >>>> (define (on-complete x) (log-test-debug "entering on-complete") x) >>>> (struct person (name) #:transparent) >>>> >>>> (log-bracketed test-debug "on-complete" "time: ~a" (current-seconds) >>>> (on-complete (person 'bob))) >>>> (log-bracketed test-debug "on-complete" "" "no user-specified logging >>>> information") >>>> >>>> Spits out: >>>> >>>> >>>> test: about to on-complete. time: 1630611613 >>>> test: entering on-complete >>>> test: after on-complete. time: 1630611613. result: (person 'bob) >>>> (person 'bob) >>>> test: about to on-complete >>>> test: after on-complete. result: "no user-specified logging information" >>>> "no user-specified logging information" >>>> >>>> >>>> The problem is that this looks like it's a simple logging message when >>>> in fact it's real code that should not be ignored. I'm trying to think of >>>> a better way to do it...maybe something like this?: >>>> >>>> (with-bracketing-logs ([test-debug "on-complete" "time: ~a" >>>> (current-seconds)]) >>>> >>>> (on-complete (person 'bob)) >>>> >>>> >>>> >>>> -- >>>> 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]. >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/racket-users/CAE8gKocZha-NpiFAAKT1c8QTG3MDFRnvxCD4T0P269EncZW3KQ%40mail.gmail.com >>>> >>>> <https://groups.google.com/d/msgid/racket-users/CAE8gKocZha-NpiFAAKT1c8QTG3MDFRnvxCD4T0P269EncZW3KQ%40mail.gmail.com?utm_medium=email&utm_source=footer> >>>> . >>>> >>> -- >>> 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]. >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/racket-users/CAFrFfuEqt1NVjE2Ft1JVArvWnKUBvK7jPVoLqPhYCd-dB00A3Q%40mail.gmail.com >>> >>> <https://groups.google.com/d/msgid/racket-users/CAFrFfuEqt1NVjE2Ft1JVArvWnKUBvK7jPVoLqPhYCd-dB00A3Q%40mail.gmail.com?utm_medium=email&utm_source=footer> >>> . >>> >> -- 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]. To view this discussion on the web visit https://groups.google.com/d/msgid/racket-users/cd37bf92-11b4-4b93-8baa-77b40c50d9f0n%40googlegroups.com.
