From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Alvaro Herrera <alvherre(at)commandprompt(dot)com> |
Cc: | zhong ming wu <mr(dot)z(dot)m(dot)wu(at)gmail(dot)com>, pgsqlrpms-hackers(at)pgfoundry(dot)org, pgsql-general <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: weird bug in rebuilding RPMs |
Date: | 2010-02-08 15:33:06 |
Message-ID: | 10542.1265643186@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs pgsql-general |
Alvaro Herrera <alvherre(at)commandprompt(dot)com> writes:
> zhong ming wu escribi:
>> I can now rebuild rpms so that all files go under a specific directory
>> specified by _prefix directive in ~/.rpmmacros Forgetting lots of
>> other bugs that I managed to fix, this weird bug is making me use
>> _prefix that does not include the word "pgsql" in the path name any
>> where.
> Yeah, the Makefiles contain a trick to put append "/pgsql" to the
> include dir (and others) if the prefix does not already contain "pgsql"
> or "postgresql" or similar. This is probably what is causing the build
> to fail. See src/Makefile.global.in:
Yeah, I suspect that the RPMs depend in various ways on the assumption
that those pathname substitutions will happen. I haven't got round to
testing, but it might be a good idea to change rpm-pgsql.patch to make
the substitutions unconditional, so that the file layout below the
_prefix is fixed regardless of what the _prefix is.
But having said that, I don't get the point of trying to build a
nonstandard installation from the RPM. That seems more or less
antithetical to most of the design concepts of RPM-based distros;
and it certainly seems pretty silly if your intent is to transition
to the standard RPM later.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Devrim GÜNDÜZ | 2010-02-08 16:37:00 | Re: [Pgsqlrpms-hackers] weird bug in rebuilding RPMs |
Previous Message | Alvaro Herrera | 2010-02-08 15:23:05 | Re: weird bug in rebuilding RPMs |
From | Date | Subject | |
---|---|---|---|
Next Message | erobles | 2010-02-08 16:06:58 | which the best way to start postgres. |
Previous Message | Alvaro Herrera | 2010-02-08 15:23:05 | Re: weird bug in rebuilding RPMs |