Received: from postgresql.org.org (webmail.postgresql.org [216.126.85.28]) by postgresql.org (8.11.3/8.11.1) with SMTP id f38LLr885549 for ; Sun, 8 Apr 2001 17:21:53 -0400 (EDT) (envelope-from pgsql-interfaces-owner@postgresql.org) Received: from shienar.alanne.com (cx253468-a.chspk1.va.home.com [24.4.151.217]) by postgresql.org (8.11.3/8.11.1) with SMTP id f38H09849242 for ; Sun, 8 Apr 2001 13:00:10 -0400 (EDT) (envelope-from jeff@alanne.com) Message-Id: <200104081700.f38H09849242@postgresql.org> Received: (qmail 11453 invoked from network); 8 Apr 2001 21:33:06 -0000 Received: from unknown (HELO cairhien) (10.0.0.3) by 10.0.0.1 with SMTP; 8 Apr 2001 21:33:06 -0000 Content-Type: text/plain; charset="iso-8859-1" Content-Disposition: inline Content-Transfer-Encoding: 7bit MIME-Version: 1.0 From: "Jeff Duffy" To: pgsql-interfaces@postgresql.org Cc: "Lane, Tom" , "Momjian, Bruce" Subject: Status of DBD:Pg and Pg interfaces for Perl X-Mailer: Pronto v2.2.3 On linux/Pg Date: 08 Apr 2001 12:00:11 EDT Reply-To: "Jeff Duffy" X-Archive-Number: 200104/21 X-Sequence-Number: 1404 Greetings. The RC3 release has an ancient version of the perl Pg interface in /src/interfaces/Perl5, which predates the DBI architecture, the de-facto standard for interfacing with a database from Perl. The upshot of this is that most people will install the DBD::Pg module from CPAN rather than use the older Pg interface in the source tree, and neither Pg.pm or DBD::Pg include the segfault-on-text-insert fix Tom provided. This means that as of now there is no obvious way to get a Perl5 driver that will work correctly with toasted fields in 7.1. I have emailed the maintainer (Edmund Mergl) twice in two weeks in an attempt to get the patch applied, and he hasn't responded to either of the emails, or any of the relevant threads on the DBI mailing lists. What are the odds of getting DBD::Pg version 0.95 (the latest) patched with Tom's fix, into the 7.1 release (maybe as an addition to /src/interfaces/Perl5)? Can a new version of an interface library be considered a bugfix if it does not work properly, as this one does not? Or is it just too late in the cycle for his kind of addition (in which case another method of access needs to be available)? Jeff -- Jeff Duffy jeff@alanne.com