Обсуждение: Support for DBD::PgSPI with Pg 7.4.8?
Hello, I am wondering if there is support for DBD::PgSPI in PostgreSQL 7.4.8 (that is, should I be able to build DBD::PgSPI with this Pg?). I tried to make it to no avail (make failed with several errors). This is on a Fedora Core 3 box with RPM installed postgres. Any suggestions would be appreciated! Thanks, Scott ------------ Here is the output of make: [root@localhost DBD-PgSPI-0.02]# export POSTGRES_HOME=/var/lib/pgsql [root@localhost DBD-PgSPI-0.02]# perl Makefile.PL Configuring PgSPI Remember to actually read the README file ! OS: linux Multiple copies of Driver.xst found in: /usr/lib/perl5/site_perl/5.8.5/i386-linux-thread-multi/auto/DBI/ /usr/lib/perl5/vendor_perl/5.8.3/i386-linux-thread-multi/auto/DBI/at Makefile.PL line 25 Using DBI 1.48 (for perl 5.008005 on i386-linux-thread-multi) installed in /usr/lib/perl5/site_perl/5.8.5/i386-linux-thread-multi/auto/DBI/ Checking if your kit is complete... Looks good Writing Makefile for DBD::PgSPI [root@localhost DBD-PgSPI-0.02]# make cp PgSPI.pm blib/lib/DBD/PgSPI.pm /usr/bin/perl /usr/lib/perl5/5.8.5/ExtUtils/xsubpp -typemap /usr/lib/perl5/5.8.5/ExtUtils/typemap PgSPI.xs > PgSPI.xsc&& mv PgSPI.xsc PgSPI.c gcc -c -I/var/lib/pgsql/include -I/var/lib/pgsql/backend -I/var/lib/pgsql/interfaces/libpq -I/usr/lib/perl5/site_perl/5.8.5/i386-linux-thread-multi/auto/DBI/-D_REENTRANT -D_GNU_SOURCE -DTHREADS_HAVE_PIDS -DDEBUGGING-fno-strict-aliasing -pipe -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -I/usr/include/gdbm-g -DVERSION=\"0.02\" -DXS_VERSION=\"0.02\" -fPIC "-I/usr/lib/perl5/5.8.5/i386-linux-thread-multi/CORE" PgSPI.c In file included from PgSPI.xs:14: PgSPI.h:16:22: postgres.h: No such file or directory PgSPI.h:17:21: funcapi.h: No such file or directory PgSPI.h:18:26: executor/spi.h: No such file or directory PgSPI.h:26:20: ppport.h: No such file or directory In file included from PgSPI.h:32, from PgSPI.xs:14: dbdimp.h:36: error: syntax error before "TupleDesc" dbdimp.h:36: warning: no semicolon at end of struct or union dbdimp.h:37: warning: data definition has no type or storage class dbdimp.h:47: error: syntax error before '*' token dbdimp.h:47: warning: data definition has no type or storage class dbdimp.h:52: error: syntax error before '}' token PgSPI.xs: In function `XS_DBD__PgSPI__st__prepare': PgSPI.xs:175: error: dereferencing pointer to incomplete type PgSPI.xs: In function `XS_DBD__PgSPI__st_execute': PgSPI.xs:233: error: dereferencing pointer to incomplete type PgSPI.xs:234: error: dereferencing pointer to incomplete type PgSPI.xs:235: error: dereferencing pointer to incomplete type PgSPI.xs: In function `XS_DBD__PgSPI__st_finish': PgSPI.xs:294: error: dereferencing pointer to incomplete type PgSPI.xs:301: error: dereferencing pointer to incomplete type PgSPI.xs: In function `XS_DBD__PgSPI__st_DESTROY': PgSPI.xs:345: error: dereferencing pointer to incomplete type PgSPI.xs:346: error: dereferencing pointer to incomplete type PgSPI.xs:351: error: dereferencing pointer to incomplete type PgSPI.xs:352: error: dereferencing pointer to incomplete type PgSPI.xs:352: error: dereferencing pointer to incomplete type PgSPI.xs:352: error: dereferencing pointer to incomplete type PgSPI.xs:354: error: dereferencing pointer to incomplete type PgSPI.xs: In function `boot_DBD__PgSPI': PgSPI.xs:36: error: invalid application of `sizeof' to incomplete type `/usr/lib/perl5/site_perl/5.8.5/i386-linux-thread-multi/auto/DBI/DBIXS.h' make: *** [PgSPI.o] Error 1 -- ------------------------------------------------------------------------ Scott Cain, Ph. D. cain@cshl.edu GMOD Coordinator (http://www.gmod.org/) 216-392-3087 Cold Spring Harbor Laboratory
Scott Cain <cain@cshl.edu> writes:
> gcc -c -I/var/lib/pgsql/include -I/var/lib/pgsql/backend -I/var/lib/pgsql/interfaces/libpq
-I/usr/lib/perl5/site_perl/5.8.5/i386-linux-thread-multi/auto/DBI/-D_REENTRANT -D_GNU_SOURCE -DTHREADS_HAVE_PIDS
-DDEBUGGING-fno-strict-aliasing -pipe -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64
-I/usr/include/gdbm-g -DVERSION=\"0.02\" -DXS_VERSION=\"0.02\" -fPIC
"-I/usr/lib/perl5/5.8.5/i386-linux-thread-multi/CORE" PgSPI.c
> In file included from PgSPI.xs:14:
> PgSPI.h:16:22: postgres.h: No such file or directory
> PgSPI.h:17:21: funcapi.h: No such file or directory
> PgSPI.h:18:26: executor/spi.h: No such file or directory
> PgSPI.h:26:20: ppport.h: No such file or directory
Either you don't have those header files available, or the PgSPI
makefile is confused about where to look for them --- the
-I/var/lib/pgsql/include -I/var/lib/pgsql/backend switches look
like rather generic guesses as to where they are...
All the other errors are probably just cascading from that.
regards, tom lane
Tom, Thanks! Of course I should have figured that out on my own. Anyway, to get it to work, I had to modify the Makefile.PL file for the distribution. It requires a $POSTGRES_HOME environment variable, which doesn't help in Fedora, because files are spread all over the place. I just added a few items to the include list so that the appropriate header files would be found. Thanks, Scott On Wed, 2005-07-13 at 13:36 -0400, Tom Lane wrote: > Scott Cain <cain@cshl.edu> writes: > > gcc -c -I/var/lib/pgsql/include -I/var/lib/pgsql/backend -I/var/lib/pgsql/interfaces/libpq -I/usr/lib/perl5/site_perl/5.8.5/i386-linux-thread-multi/auto/DBI/-D_REENTRANT -D_GNU_SOURCE -DTHREADS_HAVE_PIDS -DDEBUGGING-fno-strict-aliasing -pipe -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -I/usr/include/gdbm-g -DVERSION=\"0.02\" -DXS_VERSION=\"0.02\" -fPIC "-I/usr/lib/perl5/5.8.5/i386-linux-thread-multi/CORE" PgSPI.c > > In file included from PgSPI.xs:14: > > PgSPI.h:16:22: postgres.h: No such file or directory > > PgSPI.h:17:21: funcapi.h: No such file or directory > > PgSPI.h:18:26: executor/spi.h: No such file or directory > > PgSPI.h:26:20: ppport.h: No such file or directory > > Either you don't have those header files available, or the PgSPI > makefile is confused about where to look for them --- the > -I/var/lib/pgsql/include -I/var/lib/pgsql/backend switches look > like rather generic guesses as to where they are... > > All the other errors are probably just cascading from that. > > regards, tom lane -- ------------------------------------------------------------------------ Scott Cain, Ph. D. cain@cshl.edu GMOD Coordinator (http://www.gmod.org/) 216-392-3087 Cold Spring Harbor Laboratory