variadic arity Alex Shinn (17 Jun 2020 00:50 UTC)
Re: variadic arity Alex Shinn (17 Jun 2020 01:00 UTC)
Re: variadic arity Alex Shinn (17 Jun 2020 22:18 UTC)
Re: variadic arity Arthur A. Gleckler (17 Jun 2020 22:23 UTC)
Re: variadic arity Marc Nieper-Wißkirchen (18 Jun 2020 08:04 UTC)
Re: variadic arity Alex Shinn (18 Jun 2020 09:34 UTC)
Re: variadic arity Marc Nieper-Wißkirchen (18 Jun 2020 09:56 UTC)
Re: variadic arity Marc Nieper-Wißkirchen (18 Jun 2020 11:47 UTC)
Re: variadic arity Alex Shinn (18 Jun 2020 13:22 UTC)
Re: variadic arity Marc Nieper-Wißkirchen (18 Jun 2020 16:39 UTC)

Re: variadic arity Marc Nieper-Wißkirchen 18 Jun 2020 16:39 UTC

Am Do., 18. Juni 2020 um 15:22 Uhr schrieb Alex Shinn <xxxxxx@gmail.com>:

> The point is, relying on fast case-lambda is unreliable.  It will in fact
> make code slower in many implementations. On the other hand,

That's a problem that can only be cured by these implementations.
Besides lambda, case-lambda is the only form defined in the R7RS that
creates procedures and for procedures taking one optional argument, a
simple R7RS way to implement them is to use case-lambda. So
case-lambda *is* being used (as it is also shown by many SRFI
implementations.)

The alternative, rest argument parsing, is also usually less
performant for Schemes that do not have a slow case-lambda. Moreover,
rest argument parsing works even less with the ideas of SRFI 191.

Regardless of what we are discussing here, a faster case-lambda
implementation for Chibi would certainly be helpful (regardless of
whether the author of Chibi is fond of case-lambda :)).