Re: [HACKERS] compile errors in new PL/Pler
On Fri, 2 Jul 2004, Joshua D. Drake wrote: So we have three choices as I see it: 1) revert the change 2) require some minimally recent version of perl 3) fix the issue in place Preferences? Joshua/Andrew -- do you want to take a shot at making this work on perl 5.00503? I personally don't have any desire to make this work on a version of perl that the perl community itself suggests that you should upgrade. Well, backing it out is always an option, since the old version apparently did work *shrug* Personally, rather have the new on in there though ... :) Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: [EMAIL PROTECTED] Yahoo!: yscrappy ICQ: 7615664 ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faqs/FAQ.html
Re: [HACKERS] compile errors in new PL/Pler
Joe Conway said:
Marc G. Fournier wrote:
On Fri, 2 Jul 2004, Christopher Kings-Lynne wrote:
I am going to bet dollars to donuts that it is your perl version.
Perl 5.00503 is ancient. Try upgrading to at least 5.6.
Not much I can do about that - it's builtin as part of FreeBSD 4.x
series.
And I bet its still the 'standard' for alot of ppl out there running
it ... its only within the past couple of months that we've even
upgraded mail.postgresql.org to it, and only because I wanted to see
if it would improve majordomo's performance (which it didn't) ...
So we have three choices as I see it:
1) revert the change
2) require some minimally recent version of perl
3) fix the issue in place
Preferences?
Joshua/Andrew -- do you want to take a shot at making this work on perl
5.00503?
Joe
1 is not acceptable, and 2 kinda sucks.
Please try adding this to SPI.xs,
#ifndef aTHX_
#define aTHX_
#endif
recompile and test by calling spi_exec_query with more than 2 arguments like
this:
create function should_croak() returns text language plperl as $$
my $rv = spi_exec_query(select 1,5,blurfl);
return $rv-{status};
$$;
select should_croak();
cheers
andrew
---(end of broadcast)---
TIP 8: explain analyze is your friend
Re: [HACKERS] compile errors in new PL/Pler
1 is not acceptable, and 2 kinda sucks. Please try adding this to SPI.xs, #ifndef aTHX_ #define aTHX_ #endif recompile and test by calling spi_exec_query with more than 2 arguments like this: Hmmm, compiled fine, but then I get this: -bash-2.05b$ createlang -d test plperl ERROR: could not load library /home/chriskl/local/lib/postgresql/plperl.so: dlopen '/home/chriskl/local/lib/postgresql/plperl.so' failed. (/home/chriskl/local/lib/postgresql/plperl.so: Undefined symbol eval_pv) createlang: language installation failed: ERROR: could not load library /home/chriskl/local/lib/postgresql/plperl.so: dlopen '/home/chriskl/local/lib/postgresql/plperl.so' failed. (/home/chriskl/local/lib/postgresql/plperl.so: Undefined symbol eval_pv) Chris ---(end of broadcast)--- TIP 8: explain analyze is your friend
Re: [HACKERS] compile errors in new PL/Pler
Christopher Kings-Lynne said: 1 is not acceptable, and 2 kinda sucks. Please try adding this to SPI.xs, #ifndef aTHX_ #define aTHX_ #endif recompile and test by calling spi_exec_query with more than 2 arguments like this: Hmmm, compiled fine, but then I get this: -bash-2.05b$ createlang -d test plperl ERROR: could not load library /home/chriskl/local/lib/postgresql/plperl.so: dlopen '/home/chriskl/local/lib/postgresql/plperl.so' failed. (/home/chriskl/local/lib/postgresql/plperl.so: Undefined symbol eval_pv) createlang: language installation failed: ERROR: could not load library /home/chriskl/local/lib/postgresql/plperl.so: dlopen '/home/chriskl/local/lib/postgresql/plperl.so' failed. (/home/chriskl/local/lib/postgresql/plperl.so: Undefined symbol eval_pv) progress is good We seem to have a mixture of old-style and new-style calls which needs to be cleaned up. I would relly appreciate some experienced eyes being cast over this, if there are any around. In the meantime, does this solve your problem?: #ifndef eval_pv #define eval_pv perl_eval_pv #endif cheers andrew ---(end of broadcast)--- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [HACKERS] compile errors in new PL/Pler
At 2004-07-02 08:55:56 -0400, [EMAIL PROTECTED] wrote: In the meantime, does this solve your problem?: #ifndef eval_pv #define eval_pv perl_eval_pv #endif The right way to do this is to #include ppport.h from Devel::PPPort. -- ams ---(end of broadcast)--- TIP 7: don't forget to increase your free space map settings
Re: [HACKERS] compile errors in new PL/Pler
Christopher Kings-Lynne [EMAIL PROTECTED] writes: I am going to bet dollars to donuts that it is your perl version. Perl 5.00503 is ancient. Try upgrading to at least 5.6. Not much I can do about that - it's builtin as part of FreeBSD 4.x series. Chris isn't the only one who's going to insist on fixing this ;-) $ perl -v This is perl, version 5.005_03 built for PA-RISC2.0 regards, tom lane ---(end of broadcast)--- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [HACKERS] compile errors in new PL/Pler
On Fri, 2 Jul 2004, Tom Lane wrote: Christopher Kings-Lynne [EMAIL PROTECTED] writes: I am going to bet dollars to donuts that it is your perl version. Perl 5.00503 is ancient. Try upgrading to at least 5.6. Not much I can do about that - it's builtin as part of FreeBSD 4.x series. Chris isn't the only one who's going to insist on fixing this ;-) $ perl -v This is perl, version 5.005_03 built for PA-RISC2.0 And on FreeBSD 4.x, you can force use of the perl 5.[68] ports by installing one of them, and then issuing: use.perl port from a root shell. (It won't help Tom, but will on FreeBSD. ) LER -- Larry Rosenman http://www.lerctr.org/~ler Phone: +1 972-414-9812 E-Mail: [EMAIL PROTECTED] US Mail: 1905 Steamboat Springs Drive, Garland, TX 75044-6749 ---(end of broadcast)--- TIP 7: don't forget to increase your free space map settings
Re: [HACKERS] compile errors in new PL/Pler
In the meantime, does this solve your problem?: #ifndef eval_pv #define eval_pv perl_eval_pv #endif Hmmm, I put it in there and rebuilt and reinstalled and tried again: -bash-2.05b$ createlang -d test plperl ERROR: could not load library /home/chriskl/local/lib/postgresql/plperl.so: dlopen '/home/chriskl/local/lib/postgresql/plperl.so' failed. (/home/chriskl/local/lib/postgresql/plperl.so: Undefined symbol eval_pv) createlang: language installation failed: ERROR: could not load library /home/chriskl/local/lib/postgresql/plperl.so: dlopen '/home/chriskl/local/lib/postgresql/plperl.so' failed. (/home/chriskl/local/lib/postgresql/plperl.so: Undefined symbol eval_pv) Same error? I wonder if I did something wrong... Chris ---(end of broadcast)--- TIP 1: subscribe and unsubscribe commands go to [EMAIL PROTECTED]
Re: [HACKERS] compile errors in new PL/Pler
Larry Rosenman said: On Fri, 2 Jul 2004, Tom Lane wrote: Christopher Kings-Lynne [EMAIL PROTECTED] writes: I am going to bet dollars to donuts that it is your perl version. Perl 5.00503 is ancient. Try upgrading to at least 5.6. Not much I can do about that - it's builtin as part of FreeBSD 4.x series. Chris isn't the only one who's going to insist on fixing this ;-) $ perl -v This is perl, version 5.005_03 built for PA-RISC2.0 And on FreeBSD 4.x, you can force use of the perl 5.[68] ports by installing one of them, and then issuing: use.perl port from a root shell. (It won't help Tom, but will on FreeBSD. ) We will fix this, and we won't require this upgrade, I believe. cheers andrew ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] compile errors in new PL/Pler
We seem to have a mixture of old-style and new-style calls which needs to be cleaned up. I would relly appreciate some experienced eyes being cast over this, if there are any around. In the meantime, does this solve your problem?: I went back to clean CVS and I noticed I had these warnings: gcc -O2 -fno-strict-aliasing -fpic -DPIC -I. -I/usr/libdata/perl/5.00503/mach/CORE -I../../../src/include -c -o plperl.o plperl.c -MMD plperl.c: In function `plperl_get_keys': plperl.c:417: warning: passing arg 3 of `Perl_av_store' makes pointer from integer without a cast plperl.c: In function `plperl_get_elem': plperl.c:452: warning: passing arg 2 of `Perl_hv_exists_ent' makes pointer from integer without a cast plperl.c: In function `plperl_func_handler': plperl.c:930: warning: passing arg 3 of `Perl_av_store' makes pointer from integer without a cast Chris ---(end of broadcast)--- TIP 7: don't forget to increase your free space map settings
Re: [HACKERS] compile errors in new PL/Pler
On Fri, 2 Jul 2004, Larry Rosenman wrote: On Fri, 2 Jul 2004, Tom Lane wrote: Christopher Kings-Lynne [EMAIL PROTECTED] writes: I am going to bet dollars to donuts that it is your perl version. Perl 5.00503 is ancient. Try upgrading to at least 5.6. Not much I can do about that - it's builtin as part of FreeBSD 4.x series. Chris isn't the only one who's going to insist on fixing this ;-) $ perl -v This is perl, version 5.005_03 built for PA-RISC2.0 And on FreeBSD 4.x, you can force use of the perl 5.[68] ports by installing one of them, and then issuing: use.perl port from a root shell. (It won't help Tom, but will on FreeBSD. ) On a *reasonably* established FreeBSD box, that makes heavy use of perl (ie. has *alot* of modules installed), I find that installing the port can be a bit of a hardship ... on new machines, its something that I do automatically, but it isn't a solution for older boxes, IMHO ... Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: [EMAIL PROTECTED] Yahoo!: yscrappy ICQ: 7615664 ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [HACKERS] compile errors in new PL/Pler
Christopher Kings-Lynne said: We seem to have a mixture of old-style and new-style calls which needs to be cleaned up. I would relly appreciate some experienced eyes being cast over this, if there are any around. In the meantime, does this solve your problem?: I went back to clean CVS and I noticed I had these warnings: gcc -O2 -fno-strict-aliasing -fpic -DPIC -I. -I/usr/libdata/perl/5.00503/mach/CORE -I../../../src/include -c -o plperl.o plperl.c -MMD plperl.c: In function `plperl_get_keys': plperl.c:417: warning: passing arg 3 of `Perl_av_store' makes pointer from integer without a cast plperl.c: In function `plperl_get_elem': plperl.c:452: warning: passing arg 2 of `Perl_hv_exists_ent' makes pointer from integer without a cast plperl.c: In function `plperl_func_handler': plperl.c:930: warning: passing arg 3 of `Perl_av_store' makes pointer from integer without a cast I suspect it's the same problem, i.e. it doesn't know about eval_pv Please see if the patch I just sent to -patches works for all these problems. thanks andrew ---(end of broadcast)--- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [HACKERS] compile errors in new PL/Pler
So we have three choices as I see it: 1) revert the change 2) require some minimally recent version of perl 3) fix the issue in place Preferences? Joshua/Andrew -- do you want to take a shot at making this work on perl 5.00503? I personally don't have any desire to make this work on a version of perl that the perl community itself suggests that you should upgrade. Perl 5.00503 is RedHat 6.2 genre... That is scary old. I believe even RedHat 7.3 came with Perl 5.6 and that is old as well. Sincerely, Joshua D. Drake Joe -- Command Prompt, Inc., home of Mammoth PostgreSQL - S/ODBC and S/JDBC Postgresql support, programming shared hosting and dedicated hosting. +1-503-667-4564 - [EMAIL PROTECTED] - http://www.commandprompt.com Mammoth PostgreSQL Replicator. Integrated Replication for PostgreSQL begin:vcard fn:Joshua D. Drake n:Drake;Joshua D. org:Command Prompt, Inc. adr:;;PO Box 215;Cascade Locks;Oregon;97014;USA email;internet:[EMAIL PROTECTED] title:Consultant tel;work:503-667-4564 tel;fax:503-210-0034 note:Command Prompt, Inc. is the largest and oldest US based commercial PostgreSQL support provider. We provide the only commercially viable integrated PostgreSQL replication solution, but also custom programming, and support. We authored the book Practical PostgreSQL, the procedural language plPHP, and adding trigger capability to plPerl. x-mozilla-html:FALSE url:http://www.commandprompt.com/ version:2.1 end:vcard ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faqs/FAQ.html
Re: [HACKERS] compile errors in new PL/Pler
Joshua D. Drake said: So we have three choices as I see it: 1) revert the change 2) require some minimally recent version of perl 3) fix the issue in place Preferences? Joshua/Andrew -- do you want to take a shot at making this work on perl 5.00503? I personally don't have any desire to make this work on a version of perl that the perl community itself suggests that you should upgrade. Perl 5.00503 is RedHat 6.2 genre... That is scary old. I believe even RedHat 7.3 came with Perl 5.6 and that is old as well. I have my fingers crossed that the patch I sent in with an upgraded ppport.h (which is supposed to handle these issues) will fix the matter. It was simply generated with this command: perl -MDevel::PPPort -e 'Devel::PPPort::WriteFile();' I would like to support as wide a user base as is reasonably possible. If this doesn't work we could restrict support to some version, but that would be a fallback. Certainly, we should not let it get to the stage of compile/link errors. cheers andrew ---(end of broadcast)--- TIP 8: explain analyze is your friend
[HACKERS] compile errors in new PL/Pler
gmake[3]: Entering directory `/space/1/home/chriskl/pgsql/src/pl/plperl' gcc -O2 -fno-strict-aliasing -g -fpic -DPIC -I. -I/usr/libdata/perl/5.00503/mach/CORE -I../../../src/include -c -o SPI.o SPI.c -MMD SPI.xs: In function `XS__spi_exec_query': SPI.xs:51: `aTHX_' undeclared (first use in this function) SPI.xs:51: (Each undeclared identifier is reported only once SPI.xs:51: for each function it appears in.) SPI.xs:51: syntax error before string constant gmake[3]: *** [SPI.o] Error 1 gmake[3]: Leaving directory `/space/1/home/chriskl/pgsql/src/pl/plperl' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/space/1/home/chriskl/pgsql/src/pl' gmake[1]: *** [all] Error 2 gmake[1]: Leaving directory `/space/1/home/chriskl/pgsql/src' gmake: *** [all] Error 2 Chris ---(end of broadcast)--- TIP 8: explain analyze is your friend
Re: [HACKERS] compile errors in new PL/Pler
Christopher Kings-Lynne wrote: gmake[3]: Entering directory `/space/1/home/chriskl/pgsql/src/pl/plperl' gcc -O2 -fno-strict-aliasing -g -fpic -DPIC -I. -I/usr/libdata/perl/5.00503/mach/CORE -I../../../src/include -c -o SPI.o SPI.c -MMD I am going to bet dollars to donuts that it is your perl version. Perl 5.00503 is ancient. Try upgrading to at least 5.6. Sincerely, Joshua D. Drake SPI.xs: In function `XS__spi_exec_query': SPI.xs:51: `aTHX_' undeclared (first use in this function) SPI.xs:51: (Each undeclared identifier is reported only once SPI.xs:51: for each function it appears in.) SPI.xs:51: syntax error before string constant gmake[3]: *** [SPI.o] Error 1 gmake[3]: Leaving directory `/space/1/home/chriskl/pgsql/src/pl/plperl' gmake[2]: *** [all] Error 2 gmake[2]: Leaving directory `/space/1/home/chriskl/pgsql/src/pl' gmake[1]: *** [all] Error 2 gmake[1]: Leaving directory `/space/1/home/chriskl/pgsql/src' gmake: *** [all] Error 2 Chris ---(end of broadcast)--- TIP 8: explain analyze is your friend -- Command Prompt, Inc., home of Mammoth PostgreSQL - S/ODBC and S/JDBC Postgresql support, programming shared hosting and dedicated hosting. +1-503-667-4564 - [EMAIL PROTECTED] - http://www.commandprompt.com PostgreSQL Replicator -- production quality replication for PostgreSQL ---(end of broadcast)--- TIP 7: don't forget to increase your free space map settings
Re: [HACKERS] compile errors in new PL/Pler
gmake[3]: Entering directory `/space/1/home/chriskl/pgsql/src/pl/plperl' gcc -O2 -fno-strict-aliasing -g -fpic -DPIC -I. -I/usr/libdata/perl/5.00503/mach/CORE -I../../../src/include -c -o SPI.o SPI.c -MMD I am going to bet dollars to donuts that it is your perl version. Perl 5.00503 is ancient. Try upgrading to at least 5.6. Not much I can do about that - it's builtin as part of FreeBSD 4.x series. Chris ---(end of broadcast)--- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
Re: [HACKERS] compile errors in new PL/Pler
On Fri, 2 Jul 2004, Christopher Kings-Lynne wrote: gmake[3]: Entering directory `/space/1/home/chriskl/pgsql/src/pl/plperl' gcc -O2 -fno-strict-aliasing -g -fpic -DPIC -I. -I/usr/libdata/perl/5.00503/mach/CORE -I../../../src/include -c -o SPI.o SPI.c -MMD I am going to bet dollars to donuts that it is your perl version. Perl 5.00503 is ancient. Try upgrading to at least 5.6. Not much I can do about that - it's builtin as part of FreeBSD 4.x series. And I bet its still the 'standard' for alot of ppl out there running it ... its only within the past couple of months that we've even upgraded mail.postgresql.org to it, and only because I wanted to see if it would improve majordomo's performance (which it didn't) ... Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: [EMAIL PROTECTED] Yahoo!: yscrappy ICQ: 7615664 ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
Re: [HACKERS] compile errors in new PL/Pler
Marc G. Fournier wrote: On Fri, 2 Jul 2004, Christopher Kings-Lynne wrote: I am going to bet dollars to donuts that it is your perl version. Perl 5.00503 is ancient. Try upgrading to at least 5.6. Not much I can do about that - it's builtin as part of FreeBSD 4.x series. And I bet its still the 'standard' for alot of ppl out there running it ... its only within the past couple of months that we've even upgraded mail.postgresql.org to it, and only because I wanted to see if it would improve majordomo's performance (which it didn't) ... So we have three choices as I see it: 1) revert the change 2) require some minimally recent version of perl 3) fix the issue in place Preferences? Joshua/Andrew -- do you want to take a shot at making this work on perl 5.00503? Joe ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
