X-Original-To: pgsql-sql@postgresql.org Received: from localhost (postgresql.org [64.49.215.8]) by postgresql.org (Postfix) with ESMTP id 9AB6E475AE5 for ; Mon, 24 Feb 2003 04:32:48 -0500 (EST) Received: from postgresql.org ([64.49.215.8]) by localhost (postgresql.org [64.49.215.8:10024]) (amavisd-new) with SMTP id 48348-02 for ; Mon, 24 Feb 2003 04:32:37 -0500 (EST) Received: from fzkmail2.fzk.de (fzkmail2.fzk.de [141.52.27.52]) by postgresql.org (Postfix) with SMTP id 2F3C8475956 for ; Mon, 24 Feb 2003 04:30:18 -0500 (EST) Received: FROM resy5.fzk.de BY fzkmail2.fzk.de ; Mon Feb 24 10:30:17 2003 +0100 Received: œby rodos.fzk.de with ESMTP (8.8.6 (PHNE_17135)/8.7.1) id KAA07246; Mon, 24 Feb 2003 10:30:16 +0100 (MET) Message-ID: <3E59E628.ED83D72@rodos.fzk.de> Date: Mon, 24 Feb 2003 10:30:16 +0100 From: Christoph Haller X-Mailer: Mozilla 4.77 [en] (X11; U; HP-UX B.10.20 9000/780) X-Accept-Language: en MIME-Version: 1.0 To: pgsql-sql@postgresql.org Cc: wdarmst@bacchus.com.au Subject: Re: Porting from db2 problem Content-Type: text/plain; charset=iso-8859-2 Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new X-Archive-Number: 200302/527 X-Sequence-Number: 11925 > > I am trying to port an app that currently runs over db2 and oracle (on windows > os/2 linux hpux aix etc) to also run over postgres. > I am currently porting the windows client (using odbc). > I am having major problems because of the lack of with hold cursors (cursors > that remain open over a commit) in postgres. There are a lot of places where > the app will - open a cursor with hold - fetch the first 20 or so rows, allow > the user to scroll thru the result set, updating and commiting changes to > particular rows as they go. When the user gets to the end of the current set of > fetched rows - the app will fetch the next 20 or so during a scroll down > operation etc. The user may hold the read-only cursor over the result-set all > day, and it is important that rows that they do update, be immediately > available for other users. > > What do folk do to support this kind of scenario in the abscence of with hold > cursors ? Have a look at SELECT ... FOR UPDATE ... LIMIT ... OFFSET ... ; > Is implementation of with hold cursors likely in the near future? > I have no idea. Sorry. Regards, Christoph