Re: doc/src/sgml/Makefile versus VPATH - Mailing list pgsql-hackers

From Alvaro Herrera
Subject Re: doc/src/sgml/Makefile versus VPATH
Date
Msg-id 20091006164121.GK5929@alvh.no-ip.org
Whole thread Raw
In response to doc/src/sgml/Makefile versus VPATH  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: doc/src/sgml/Makefile versus VPATH
List pgsql-hackers
Tom Lane wrote:
> So isn't this still pretty broken?  I notice that the clean and
> distclean targets still use addprefix on a lot of temporary/intermediate
> files that I would think get made in the build directory, not the source
> directory.

Yeah, I noticed that too.  I'm not too sure about it, because some of
those files we do want shipped in source tarballs; and they are
definitely cleaned in the builddir by maintainer-clean.  I didn't want
to get into the detail of what's the ultimate distclean charter; my
immediate problem was that "make -C doc maintainer-clean" was not
getting rid of the stamp files and thus I was getting bit by the problem
that openjade was running all the time, even after I maintainer-cleaned. 

By Peter's recent decree that tarballs are supposed to be built in
non-vpath-builds only, I am not really sure what should actually happen
here, both on build and on the various clean targets.

> The references to man files in srcdir in nonsql_manpage_files and
> adjacent macros seem a tad suspicious as well.

I agree, it probably merits more investigation.

-- 
Alvaro Herrera                                http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 support


pgsql-hackers by date:

Previous
From: KaiGai Kohei
Date:
Subject: Re: [PATCH] Largeobject access controls
Next
From: Alvaro Herrera
Date:
Subject: Re: [PATCH] Largeobject access controls