conjoin and disjoin Rutger van Beusekom (30 Mar 2026 09:50 UTC)
Re: conjoin and disjoin Arthur A. Gleckler (30 Mar 2026 21:31 UTC)
Re: conjoin and disjoin Peter McGoron (30 Mar 2026 21:47 UTC)
Re: conjoin and disjoin Arthur A. Gleckler (30 Mar 2026 21:50 UTC)
Re: conjoin and disjoin John Cowan (02 Apr 2026 03:39 UTC)
Re: conjoin and disjoin Peter McGoron (02 Apr 2026 10:22 UTC)
Re: conjoin and disjoin Arthur A. Gleckler (02 Apr 2026 21:37 UTC)
Re: conjoin and disjoin Wolfgang Corcoran-Mathe (02 Apr 2026 21:57 UTC)
Re: conjoin and disjoin Arthur A. Gleckler (11 Apr 2026 22:01 UTC)

Re: conjoin and disjoin John Cowan 02 Apr 2026 03:39 UTC

On Mon, Mar 30, 2026 at 5:47 PM Peter McGoron <xxxxxx@mcgoron.com> wrote:

> Although `predicate` is usually a procedure of one argument

Definitely not.   R[57]RS says:  "By convention, ? is the final
character of the names of procedures that always return a boolean
value. Such procedures are called predicates." There is nothing said
about the number of arguments:  In particular, `eq?`, `eqv?`, and
`equal?` are predicates that take two arguments, and the various
ordering predicates take two or more of them.

> "predicate ... applied ... to the args" seems
> to imply that the "predicate"s are applied to each of the list of
> arguments in turn.

Just so.

> So I believe that a modification to the sample implementation that is
> equivalent to the code you posted would be sufficient, and it does not
> require a change to the SRFI.

I agree.