Re: openpty module broken on Solaris

[Reuben Thomas]

On 8 February 2013 16:09, Paul Eggert <address@hidden> wrote:

Thanks for the diagnosis.  I should have proofread the Jan 31
openpty patch better.  Reuben, if the openpty module is used,
application code should not inspect HAVE_OPENPTY to decide whether
to use openpty -- instead, the idea is that application code can
simply invoke openpty as if it were a GNU platform.

Thanks. In my case I've decided this is OK, but there was a reason I wanted to know: on mingw the gnulib module merely adds a stub that always returns an error, and the program I'm working on used to output a list of optional features as part of its help text. Without a way of testing whether I've *really* got openpty, the only way to find out is to try it, which is a little unfortunate (e.g. I then have to be sure I'm getting a "facility does not exist" error as opposed to some other failure), as opposed to just being able to test a macro at build time.

--

http://rrt.sc3d.org