X-Original-To: pgsql-sql@postgresql.org Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id D4E64474E5C for ; Fri, 14 Feb 2003 23:54:05 -0500 (EST) Received: from web12707.mail.yahoo.com (web12707.mail.yahoo.com [216.136.173.244]) by postgresql.org (Postfix) with SMTP id 2579D474E42 for ; Fri, 14 Feb 2003 23:54:05 -0500 (EST) Message-ID: <20030215045410.15935.qmail@web12707.mail.yahoo.com> Received: from [203.107.130.9] by web12707.mail.yahoo.com via HTTP; Fri, 14 Feb 2003 20:54:10 PST Date: Fri, 14 Feb 2003 20:54:10 -0800 (PST) From: chester c young Subject: Re: rownum To: Tom Lane Cc: pgsql-sql@postgresql.org In-Reply-To: <7746.1045239664@sss.pgh.pa.us> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Virus-Scanned: by AMaViS new-20020517 X-Archive-Number: 200302/354 X-Sequence-Number: 11752 > Considering that it can trivially be done on the client side, I think > it would be a hard sell to put such a wart into the protocol ... having used Oracle for many years, I cannot think of a project in which I did not use rownum, including within procedures and triggers and subqueries. can usually be easily done - loop within loop - but more easily and bugfree as column within select. by "wart" do you mean aesthetically or code-wise? naively, it seems like pg is already keeping a row counter to implement limit. __________________________________________________ Do you Yahoo!? Yahoo! Shopping - Send Flowers for Valentine's Day http://shopping.yahoo.com