first bunch of patches with mkstemp()

[Russ Allbery]

Matus \"fantomas\" Uhlar <uhlar at fantomas.sk> writes:

> True, but is there a possibility to force using our function if
> <something>?

Yes, actually... that's a good point.  If there's something in particular
that should be tested, we can add a configure test for it and use our own
tmpfile and mkstemp if the platform fails.

-- 
Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>