Re: plpython thinks it's hooked into "make distprep", but not so much

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: pgsql-hackers(at)postgresql(dot)org
Cc: Peter Eisentraut <peter_e(at)gmx(dot)net>
Subject: Re: plpython thinks it's hooked into "make distprep", but not so much
Date: 2011-07-02 20:47:52
Message-ID: 19034.1309639672@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

I wrote:
> I noticed while wrapping 9.1beta2 that plpython tries to build a file
> spiexceptions.h to be included in the tarballs, but no such file is
> actually appearing therein. The reason is that src/pl/Makefile doesn't
> recurse into the plpython subdirectory unless with_python is set.
> Which it isn't, because the tarball build script doesn't configure
> --with-python (or with any other options for that matter). If it did,
> it would fail, since there's no python installation on
> developer.postgresql.org.

> I left this go for the moment, since the only implication is that you
> have to have perl available to build plpython from the tarball. But we
> oughta fix it before final. Any opinions on what's the least ugly way
> to get control to recurse to src/pl/plpython for "make distprep"
> regardless of with_python?

> (More generally, I would like to see cleanup targets such as make
> distclean and make maintainer-clean recurse to all subdirectories
> regardless of configure options, but that might be too much to ask for
> right now.)

Attached is a proposed patch to fix this. I found that the general
case was not too hard to solve, so this patch causes the make system
to recurse everywhere for distprep, clean, distclean, maintainer-clean
cases. Anybody have a cleaner way to do it, or want to bikeshed on
the macro names?

regards, tom lane

Attachment Content-Type Size
recurse-prep-clean.patch text/x-patch 5.5 KB

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Eisentraut 2011-07-02 21:02:05 merge pg_proc and pg_operator?
Previous Message Tom Lane 2011-07-02 19:46:04 Re: [v9.2] Fix leaky-view problem, part 1