Re: [PATCH 1/1] Support mkdtemp via mkdtemp! and scm_mkdtemp

[Mike Gran]

On Mon, Jan 04, 2021 at 06:42:07PM -0600, Rob Browning wrote:
> Rob Browning <rlb@defaultvalue.org> writes:
> 
> > Mike Gran <spk121@yahoo.com> writes:
> >
> >> Since mkdtemp already returns a string of the new directory name,
> >> it might be more scheme-like to not modify the input string, and instead
> >> just return the new directory name.
> >
> > Perhaps, though I was just matching the existing semantics of guile
> > mkstemp, i.e. figured maybe they should behave the same.
> 
> ...mkstemp! I meant, and I think I might still somewhat favor keeping
> the correspondence between a mkstemp! and mkdtemp! that have the same
> semantics, and more or less directly match their POSIX functions.

OK, fair enough. I'll push a patch on mkdtemp! in a day or two if
you're good with the idea.  (I asked on IRC.) But, one thing I'd like
to change, if you don't mind, is the split into scm_i_mkdtemp and
scm_mkdtemp because there is no internal client of scm_i_mkdtemp.

Regards,
Mike Gran