C++ Logo

std-proposals

Advanced search

Re: Making parameter names clearer than func(true, false);

From: Ville Voutilainen <ville.voutilainen_at_[hidden]>
Date: Mon, 8 Jul 2019 22:31:30 +0300
On Mon, 8 Jul 2019 at 22:29, Jonny Grant via Std-Proposals
<std-proposals_at_[hidden]> wrote:
>
>
>
> On 08/07/2019 20:23, Ville Voutilainen via Std-Proposals wrote:
> > On Mon, 8 Jul 2019 at 22:19, Hyman Rosen via Std-Proposals
> > <std-proposals_at_[hidden]> wrote:
> >>
> >> On Mon, Jul 8, 2019 at 2:41 PM Thiago Macieira <thiago_at_[hidden]> wrote:
> >>>
> >>> So when coming up with examples of where named parameters would be useful,
> >>> please find good APIs that would become even better, or at least find bad ones
> >>> that have other reasons for being bad.
> >>
> >>
> >> Posix system calls, an API that isn't about to change and doesn't care whether you
> >> think it's good or bad:
> >> ssize_t int write(int fd, void *buf, size_t count);
> >
> > If the API isn't about to change, you're unlikely to be able to call
> > it with named arguments.
>
> What might be difficult is that different libc in this case have
> different parameter names. Unfortunately what I desire (named

I'm saying that it's unlikely that the committee will buy an attempt
to allow named-calls without
a declaration opting into that. Once you have a named-declaration, you
can decide what the names are.

Another problem is the standard library, implementations __uglify
parameter names. Once Modules are
available, they no longer need to do that, and named-arg functions in
stdlib become a possibility.

Received on 2019-07-08 14:33:35