_whitelogger_ has quit [Remote host closed the connection]
_whitelogger_ has joined #crux
rever_4192 has quit [Ping timeout: 248 seconds]
blackfrog78 has joined #crux
<blackfrog78>
revdep gives me for postgresql-17.1 this: /usr/lib/postgresql/plperl.so: error - recompiling postgresql does not fix it. /usr/lib/perl5/5.36/linux-thread-multi/CORE/libperl.so exists but is not found according to ldd /usr/lib/postgresql/plperl.so.
rever_4192 has joined #crux
zorz has quit [Ping timeout: 252 seconds]
zorz has joined #crux
ivandi has quit [Quit: WeeChat 4.4.3]
ivandi has joined #crux
joacim has quit [Ping timeout: 260 seconds]
<farkuhar>
blackfrog78: postgresql-perl-rpath.patch got deleted with the 17.0 -> 17.1 version bump, maybe that might explain the revdep and ldd error. Unfortunately the patch doesn't apply cleanly to the 17.1 Makefile, so it will have to be modified if you want to test this theory.
<cruxbridge>
<tim> i suspect it will need the patch as well, I'll try to rebase it
<blackfrog78>
This fixes it. No devdep error anymore
<cruxbridge>
<tim> nice, i'll push that in a bit
<cruxbridge>
<tim> thanks
<blackfrog78>
Thanks to you too
<farkuhar>
I'm always curious why one-line insertions aren't more often handled with something like sed -e '/SHLIB_LINK.*perl_embed_ldflags/aSHLIB_LINK += "$(rpath)"' -i src/pl/plperl/GNUmakefile, instead of a patch.
<farkuhar>
The patch has to be rebased when the line numbers change too much, but a sed command in build() should continue to work.
<blackfrog78>
I have to admit the "sed" solution is more elegant and probably more robust. But personally I am not that familiar with the "sed" syntax.
<cruxbridge>
<tim> farkuhar: can't say, but if you like I can take your sed which looks fine too
<farkuhar>
tim: no strong opinion here. You're free to use whichever solution you want.
<cruxbot>
[contrib/3.7]: postgresql: fix for perls rpath