New draft (#12) of SRFI 224: Integer Mappings Arthur A. Gleckler (21 Jun 2021 17:06 UTC)
Re: New draft (#12) of SRFI 224: Integer Mappings Marc Nieper-Wißkirchen (21 Jun 2021 17:59 UTC)
Re: New draft (#12) of SRFI 224: Integer Mappings Wolfgang Corcoran-Mathe (21 Jun 2021 18:33 UTC)
Re: New draft (#12) of SRFI 224: Integer Mappings Marc Nieper-Wißkirchen (21 Jun 2021 19:04 UTC)
Re: New draft (#12) of SRFI 224: Integer Mappings Wolfgang Corcoran-Mathe (21 Jun 2021 20:23 UTC)
Re: New draft (#12) of SRFI 224: Integer Mappings Marc Nieper-Wißkirchen (21 Jun 2021 20:30 UTC)

Re: New draft (#12) of SRFI 224: Integer Mappings Wolfgang Corcoran-Mathe 21 Jun 2021 20:23 UTC

On 2021-06-21 21:04 +0200, Marc Nieper-Wißkirchen wrote:
> (fxmapping-filter-map (lambda (k v skip)
>                         (when (negative? v) (skip))
>                         (square v))
>                       (fxmapping 0 2 1 -4 2 8))

Thanks, I understand now.

> But as it does not mimic some legacy API, it won't be a good idea to
> establish a flawed version (the one that handles #f specifically), I think.

I'm thinking that I'll just remove it, because the Right Thing is
going to take more research.

filter-map is related to the (deleted) fxmapping-partition-map, which,
as you've mentioned, seemed to be a special case of some more general
form for mapping a fxmapping into a number of new fxmappings.  With
filter-map, we also have the option of discarding associations, which
is just another continuation:

    (fxmapping-filter/partition-map  ; silly sample name
     (lambda (k v insert1 insert2 ... skip)
       ...)
     fxmap)

It's worth figuring out how this could work, but I'd like to do that
outside of the context of SRFI 224.

--
Wolfgang Corcoran-Mathe  <xxxxxx@sigwinch.xyz>

"Heresy is a cradle; orthodoxy, a coffin.  Why should man be
afraid to think, and why should he fear to express his thoughts?"
--Robert G. Ingersoll