On Fri, Apr 16, 2010 at 10:36 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
Scott Mead <scott.lists@enterprisedb.com> writes: > Huh? This I find hard to believe. Whenever I've had problems with readline > it was actually libtermcap that was giving me a headache. Are you sure that > there's nothing in there that's pointing you to your libtermcap being .... > 'wonky'
I think he's confusing what happens at compile and link time (where -I and -L do make a difference) with what happens at run time (where these switches do *not* make a difference, at least not on Linux).
Right, I've had issues during compile time with a scenario similar to this, where I ended up solving it by setting CFLAGS before compiling it. The configure script (at some point) was picking the new readline, but missing some other termcap dependency when trying to compile the test program. ---
