Billy G. Allie wrote:
> Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> > ... it does not build on HPUX, nor on any other platform that requires
> > specialized switches to build shared libraries. I recommend the author
> > consult the makefile for libpq, and transpose the necessary fooling
> > around into plpgsql. (On HPUX it fails for lack of -fPIC in the compile
> > commands for the pl/plpgsql/src subdirectory; there might also be
> > problems in the link command, but the compile commands are certainly
> > wrong.)
> I have made change to configure/configure.in and added a Makefile.in to help
> with the problem (I had problems compiling for UnixWare). I will be testing
> [the UnixWare part of] the patches over the new few days. I will submit the
> after I complete prlimiary testing. I will also mail the patch to anyone else
> who would like to test for a specific port.
> > We have enough different shared libraries now that it'd be a good idea
> > to try to eliminate most of that cruft from the individual module
> > makefiles, replacing it with definitions made in the port makefiles.
> > But it's a bit too late in the 6.4 beta cycle to try to do that for 6.4,
> > I think --- the testing problem looms too large. Something for next time.
> I aggree that the shared library handling needs to be standardized and the
> port specific information moved to the port specific Makefile.
> > BTW there is also a configuration check that needs to be applied: the
> > pl/tcl subdirectory should be built only if a *shared library* version
> > of Tcl is available. I happen to have only a static version of libtcl.a
> > installed, which is fine for pgtclsh, but pl/tcl tries to link with it
> > and falls over...
> Hmmmm.... I don't now autoconf well enough (yet) to handle this change.
I'd really appreciated to have PL/pgSQL installed in
template1 with the 6.4 release. But I see that the problems
we have with the make process of shared libraries is too much
trouble that close to release.
Since PL/Tcl also causes trouble let's put both into 6.5 and
don't delay 6.4 longer than required for the other fixes.
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me. #
#======================================== jwieck(at)debis(dot)com (Jan Wieck) #
In response to
pgsql-hackers by date
|Next:||From: Jan Wieck||Date: 1998-10-05 08:35:27|
|Subject: Re: [HACKERS] Dumping of views -- done!|
|Previous:||From: Jan Wieck||Date: 1998-10-05 07:17:57|
|Subject: Re: [HACKERS] Problem dumping PL/pgsql functions.|