That’s not the job of the macro expander. And the Racket optimizer does
reduce (if '#t (#%app not '#f) '#f) — in fact to #t.
If you use Racket CS, try running:
PLT_LINKLET_SHOW_CP0=1 racket file.rkt
where file.rkt is
#lang racket
(displayln (if #t (not #f) #f))
Here’s the optimized program:
;; cp0 ---------------------
(lambda (instance-variable-reference .get-syntax-literal!1
.set-transformer!2 displayln3 print-values4)
(call-with-module-prompt
(lambda ()
(#%call-with-values
(lambda () (displayln3 #t))
print-values4)))
(#2%void))
On Mon, Mar 22, 2021 at 1:28 AM [email protected] <[email protected]>
wrote:
> The [WordCount] thread caused me to play around with the solutions to see
> whether I could see anything else that can be tweaked, I tried a few things
> but they didn't improve performance, but it was fun anyway.
>
> Then I though I could try the macro stepper and look at the expansion of
> [crowdsourced].
> I disabled macro hiding and stepped to the end.
> There I saw if statements with a condition that is `(quote #t)` with a
> little lock symbol before it.
>
> I would like to understand why those terms aren't reduced further.
> Are they placeholders and are replaced and resolved at a later time?
>
> terms copied (without the lock symbol) from the macro stepper
> (if (if '#t (#%app not '#f) '#f)
> (#%app for-loop (#%app unsafe-fx+ '1 pos))
> (#%app values))
>
> From a naive standpoint I would expect it to be simplified:
> (if '#t (#%app not '#f) '#f) -> (#%app not '#f)
>
> (if (#%app not '#f)
> (#%app for-loop (#%app unsafe-fx+ '1 pos))
> (#%app values))
>
> So expressed another way: why isn't the if with boolean literal as
> condition reduced to the corresponding branch? Or does this happen, just
> isn't shown in the macro stepper?
> What am I misunderstanding here?
>
> If there aren't simplifications like that, what is preventing them or
> making them not worthwhile?
>
>
> [WordCount]
> https://groups.google.com/g/racket-users/c/lGA60P6jboY/m/2ZLDAunBBgAJ
> [crowdsourced]
> https://gist.github.com/Bogdanp/b9256e1a91de9083830cb616b3659ff8
>
> --
> 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/fce863fe-85d7-44a6-a2ba-5778df40e6bfn%40googlegroups.com
> <https://groups.google.com/d/msgid/racket-users/fce863fe-85d7-44a6-a2ba-5778df40e6bfn%40googlegroups.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/CADcuegs1qGL0MnAzUmP9gFOOVL%2BRp-EB5wghpaY8sjiKObDbkA%40mail.gmail.com.
