Re: Bugfix and new feature for PGXS - Mailing list pgsql-hackers

From Peter Eisentraut
Subject Re: Bugfix and new feature for PGXS
Date
Msg-id 1381193255.25702.4.camel@vanquo.pezone.net
Whole thread Raw
In response to Re: Bugfix and new feature for PGXS  (Andrew Dunstan <andrew@dunslane.net>)
Responses Re: Bugfix and new feature for PGXS
Re: Bugfix and new feature for PGXS
List pgsql-hackers
On Sun, 2013-09-29 at 19:09 -0400, Andrew Dunstan wrote:
> On 09/03/2013 04:04 AM, Cédric Villemain wrote:
> >> Simple one, attached.
> >> I didn't document USE_VPATH, not sure how to explain that clearly.
> > Just a remember that the doc is written and is waiting to be commited.
> >
> > There is also an issue spoted by Christoph with the installdirs prerequisite,
> > the attached patch fix that.
> >
>
> I applied this one line version of the patch, which seemed more elegant
> than the later one supplied.
>
> I backpatched that and the rest of the VPATH changes for extensiuons to
> 9.1 where we first provided for extensions, so people can have a
> reasonably uniform build system for their extensions.

I suspect this line

submake-libpq: $(libdir)/libpq.so ;

will cause problems on platforms with a different extension (e.g. OS X).

Given that this supposedly small and noninvasive set of changes has
caused a number of problems already, I suggest we revert this entire
thing until we have had time to actually test it somewhere other than in
the the stable branches.  As it stands, it is a new feature being
developed in the stable branches, which is clearly not acceptable.





pgsql-hackers by date:

Previous
From: Kevin Grittner
Date:
Subject: Re: [COMMITTERS] pgsql: Fix bugs in SSI tuple locking.
Next
From: Heikki Linnakangas
Date:
Subject: Re: SSI freezing bug