Re: PERSISTANT PREPARE (another point of view)
От | Steve Midgley |
---|---|
Тема | Re: PERSISTANT PREPARE (another point of view) |
Дата | |
Msg-id | 20080711235417.57459650695@postgresql.org обсуждение исходный текст |
Ответ на | PERSISTANT PREPARE (another point of view) (Milan Oparnica <milan.opa@hotmail.com>) |
Список | pgsql-sql |
At 04:31 PM 7/11/2008, pgsql-sql-owner@postgresql.org wrote: >Date: Fri, 11 Jul 2008 23:31:03 +0000 >From: Milan Oparnica <milan.opa@hotmail.com> >To: <pgsql-sql@postgresql.org> >Subject: PERSISTANT PREPARE (another point of view) >Message-ID: <BLU109-W5078B5ABDE57D183AA6695F8900@phx.gbl> >[snip] >What could we gain by introducing a kind of global prepared statement >area, is SIMPLICITY of DB DEVELOPMENT AND MAINTENANCE. > >Here is our point of view: >[snip] >Now, instead of preparing statements on each connection request (and >we use around 900 prepared statements), why couldn't we simply prepare >these statements ONCE and keep them in some global storage for future >everyday usage. Hi, What's wrong with using complex views, stored procedures, functions and maybe even custom data types to accomplish what you want here? It seems like you could build a lot of "prepared statements" using these tools, providing your application layer developers with a consistent set of interfaces to obtain data that are not tied to the data tables themselves. And allowing them to insert/update/manage tables via structured interfaces as well. Am I missing something? Best, Steve
В списке pgsql-sql по дате отправления: