[Toybox] [PATCH] mktemp: add -t and fix behavior.

[enh]

On Wed, Dec 5, 2018 at 10:12 AM enh <enh at google.com> wrote:
>
> On Wed, Dec 5, 2018 at 9:57 AM Rob Landley <rob at landley.net> wrote:
> >
> > On 12/5/18 9:56 AM, enh via Toybox wrote:
> > > Yes, that's what we've been talking about for days. Search for mktemp and glibc.
> > >
> > > Bionic will warm you at compile time instead, and musl should be silent.
> >
> > He's not gonna be the last one to notice, though.
> >
> > Sigh, I may need to add a mktemp in portability.c...
>
> i don't think that's a good idea. we really _shouldn't_ be using
> mktemp(3) *except* to implement mktemp(1)'s -u --- just inlining
> "replace each X with a byte from getrandom" into mktemp(1) makes more
> sense.

(in an lstat(2) loop, that is. which does make me wonder what the
expected behavior of `mktemp -u` is if the directory also isn't
readable...)

>
> > Rob