Re: 7.1Beta 4&5 - clean build problem with '--with-perl' - Mailing list pgsql-bugs

From Tom Lane
Subject Re: 7.1Beta 4&5 - clean build problem with '--with-perl'
Date
Msg-id 21035.983829214@sss.pgh.pa.us
Whole thread Raw
In response to 7.1Beta 4&5 - clean build problem with '--with-perl'  ("Creager, Robert S" <CreagRS@LOUISVILLE.STORTEK.COM>)
List pgsql-bugs
"Creager, Robert S" <CreagRS@LOUISVILLE.STORTEK.COM> writes:
> Alrighty.  I should of just looked at the Makefile.PL in the first place...
> I have the environment variables POSTGRES_INCLUDE and POSTGRES_LIB defined,
> so Makefile.PL assumes those directories exist.  When it makes, no include
> files...  I've included the patch to have the Makefile.PL check for the
> directories existence.  I added a '-d' before the $ENV{POSTGRES_???} stuff
> to make sure the directory exists.

> *** Makefile.PL    Mon Mar  5 14:31:27 2001
> --- Makefile.PL.org    Mon Mar  5 14:25:12 2001
> ***************
> *** 12,18 ****
>
>   my %opts;
>
> ! if (! -d $ENV{POSTGRES_LIB} || ! -d $ENV{POSTGRES_INCLUDE}) {
>
>       # Check that we actually are inside the Postgres source tree
>       if (! -d "../libpq") {
> --- 12,18 ----
>
>   my %opts;
>
> ! if (! $ENV{POSTGRES_LIB} || ! $ENV{POSTGRES_INCLUDE}) {
>
>       # Check that we actually are inside the Postgres source tree
>       if (! -d "../libpq") {

Ah so.  That seems like a good solution.  Do you think we should also
modify the GNUmakefile to unset POSTGRES_INCLUDE and POSTGRES_LIB when
doing 'make all'?  Or would that get in the way of legitimate uses
of these variables?

            regards, tom lane

pgsql-bugs by date:

Previous
From: pgsql-bugs@postgresql.org
Date:
Subject: Possible Addition of a CREATE TABLE Warning
Next
From: Philip Warner
Date:
Subject: Re: pg_dump -a strangeness/bug? (7.1beta5)