Re: FYI: libunbound building broken

From: Murray S. Kucherawy <msk_at_blackops.org>
Date: Fri, 14 Aug 2009 23:42:12 -0700 (PDT)

On Fri, 14 Aug 2009, Mike Markley wrote:
> The bits that actually decide whether to set USE_UNBOUND happen just
> before that test, so it would probably require some minor refactoring.

Actually it should be possible to move the test to right after the
AC_ARG_WITH for unbound. At that point you know what the user tried on
the command line, whether or not the value actually pointed at an
installation of unbound. It's enough to error out there if both
resolvers were selected.

The alternative only lets us possibly continue if "--with-unbound" was
given but didn't point to anything, but that should produce a different
error anyway.

Testing now, will commit when done.
Received on Sat Aug 15 2009 - 06:42:27 PST