On Thu, 16 Dec 2010 12:56:12 -0500, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> "H.Merijn Brand" <h.m.brand@xs4all.nl> writes:
> > On Thu, 16 Dec 2010 12:31:21 -0500, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> >> So what I'm thinking is happening is that libpq expects size_t as
> >> the argument type, but it's getting linked against a libc that
> >> expects int as the argument type, and whatever HP is doing under
> >> the hood fails to cope with that case. You might try removing
> >> -D_XOPEN_SOURCE_EXTENDED from the template file --- I don't know if
> >> that will have any bad side-effects, but it's worth a try to build
> >> libpq that way.
>
> > I'll try that first after cleaning up my environment.
>
> I looked in the git history and found that I added
> -D_XOPEN_SOURCE_EXTENDED in a patch that made getaddrinfo work
> correctly:
> http://git.postgresql.org/gitweb?p=postgresql.git;a=commit;h=1da6eb7fdaa79a93972c290174bd7136672a894f
> So taking it out may indeed have unpleasant side-effects.
Which brings me to a rather awkward position.
A fresh built perl, using the same sockets that postgres uses (X/Open)
works perfectly fine with postgres, but now Oracle, DBD::Oracle also
freshly compiled with the same options however now doesn't work anymore
because of exactly the same reasons why DBD::Pg refused to work when I
started this quest.
As I use perl with DBD::Oracle and DBD::Pg at the same time to convert
between these two databases, there doesn't seem to be a way out.
My hopes in getting Oracle to change their policy is higher than IBM
changing user support, but still very close to zero.
How much are the postgres people willing to find a solution that would
force postgres to optional use BSD type sockets?
--
H.Merijn Brand http://tux.nl Perl Monger http://amsterdam.pm.org/
using 5.00307 through 5.12 and porting perl5.13.x on HP-UX 10.20, 11.00,
11.11, 11.23 and 11.31, OpenSuSE 10.1, 11.0 .. 11.3 and AIX 5.2 and 5.3.
http://mirrors.develooper.com/hpux/ http://www.test-smoke.org/
http://qa.perl.org http://www.goldmark.org/jeff/stupid-disclaimers/
