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 RAA81024 for ; Sun, 23 Apr 2000 17:32:07 -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 RAA14701; Sun, 23 Apr 2000 17:32:01 -0400 (EDT) To: mig@utdt.edu cc: pgsql-sql@postgresql.org Subject: Re: Re: Returning sets from functions? In-reply-to: <200004231502.MAA05163@ant.utdt> References: <200004231502.MAA05163@ant.utdt> Comments: In-reply-to mig@utdt.edu message dated "Sun, 23 Apr 2000 12:02:01 -0300" Date: Sun, 23 Apr 2000 17:32:01 -0400 Message-ID: <14698.956525521@sss.pgh.pa.us> From: Tom Lane X-Archive-Number: 200004/265 mig@utdt.edu writes: > Consider > create function followers(text) returns setof text as > 'select id from ids where id > $1 ' > language 'sql'; > The function seems to work OK: > select followers('h'); > does return all indexes after the index 'h'. > On the other hand, > select * from ids where id in (select followers('h')); > produces the message > ERROR: An operand to the '=' operator returns a set of text, > but '=' takes single values, not sets. Seems to work in 7.0, which actually surprises me quite a bit ;-). Functions returning sets are only very shakily implemented right now. We've cured a few bugs since 6.5, but it's going to be difficult to make much further progress with them until the fabled querytree redesign happens (which will also bring outer joins and some other good stuff). Current plans are to tackle that for 7.2, which might be out by the end of the year or so. > Sorry, I forgot part of my question in the previous message: > How do you return a "setof" value from a PG/Tcl procedure? Right now I think you can't ... regards, tom lane