Received: from sss2.sss.pgh.pa.us (sss.pgh.pa.us [209.114.166.2]) by hub.org (8.9.3/8.9.3) with ESMTP id NAA95058 for ; Fri, 19 May 2000 13:48:35 -0400 (EDT) (envelope-from tgl@sss.pgh.pa.us) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss2.sss.pgh.pa.us (8.9.3/8.9.3) with ESMTP id NAA08872; Fri, 19 May 2000 13:48:15 -0400 (EDT) To: Peter Eisentraut cc: Bruce Momjian , gav , Mitch Vincent , pgsql-sql@postgresql.org Subject: Re: Re[2]: lower() for varchar data by creating an index In-reply-to: References: Comments: In-reply-to Peter Eisentraut message dated "Fri, 19 May 2000 19:39:38 +0200" Date: Fri, 19 May 2000 13:48:15 -0400 Message-ID: <8869.958758495@sss.pgh.pa.us> From: Tom Lane X-Archive-Number: 200005/86 Peter Eisentraut writes: > Tom Lane writes: >> What's the point of a notice? "You just deleted OID equals. Better >> luck with your next database." Either we think this is too dangerous to >> be allowed even to the dbadmin, or we don't. > Anyway, shouldn't you be able to do CREATE FUNCTION xxx (...) LANGUAGE > 'internal'; to recreate it? (And that wouldn't actually require things > like oideq to be in pg_proc, would it?) I was thinking that the CREATE needs to insert index entries, which depends on having the datatype-specific procs that will be called by the index access method. (Not sure if oideq is actually one of the ones used by any of the indexes on pg_proc, but you get my drift.) If I had some spare time I'd try it in a junk database... regards, tom lane