[Mailman-Users] Compilation problems
Barry A. Warsaw
bwarsaw at cnri.reston.va.us
Thu Dec 3 23:30:45 CET 1998
>>>>> "VG" == Victoriano Giralt <vic at vgg.sci.uma.es> writes:
VG> I've just downloaded mailman-1.0b6 and tryed to compile it in
VG> a DEC AXP with Digital Unix, and get the following errors
VG> while doing make:
VG> With gcc: gcc -DSCRIPT="\"admin\"" -I. -DCGI_GID=65534 -g -O2
VG> -DHAVE_STRERROR=1 -DSTDC_HEADERS =1 -DHAVE_SYSLOG_H=1
VG> -DGETGROUPS_T=gid_t -DHAVE_VPRINTF=1 common.o -o admin
VG> common.c common.c:26: `PREFIX' undeclared here (not in a
VG> function) common.c:26: parse error before string constant
VG> common.c:27: `PREFIX' undeclared here (not in a function)
VG> common.c:28: `PYTHON' undeclared here (not in a function)
Something is truly strange here. It looks like your system is trying
to build the admin binary by compiling common.c. It shouldn't be
doing that! It should be using cgi-wrapper.c to build all the CGI
binaries. The common.c -> common.o compilation step should happen
first and should have macros that define things like PREFIX and
PYTHON. e.g. on my system:
gcc -c -I. -DPREFIX="\"/export/home/mailman\"" -DPYTHON="\"/depot/sundry/plat/bin/python\"" -g -O2 -DHAVE_STRERROR=1 -DSTDC_HEADERS=1 -DHAVE_SYSLOG_H=1 -DGETGROUPS_T=gid_t -DHAVE_VPRINTF=1 common.c
gcc -DSCRIPT="\"admin\"" -I. -DCGI_GID=60001 -g -O2 -DHAVE_STRERROR=1 -DSTDC_HEADERS=1 -DHAVE_SYSLOG_H=1 -DGETGROUPS_T=gid_t -DHAVE_VPRINTF=1 common.o -o admin cgi-wrapper.c
What version of make are you running? It's possible I've got something
incompatible in src/Makefile with your default make. I only have GNU
make and Solaris make to test with, and the build process should
complete successfully with both of those.
Try making sure you're building with GNU make instead of whatever make
you're using by default.
If this really is the problem, we'll have to add a note to the
installation instructions. Or if you can figure out the fix to
Makefile.in and send us a patch, that would be best!
-Barry
More information about the Mailman-Users
mailing list