Received: from news.tht.net (news.hub.org [216.126.91.242]) by postgresql.org (8.11.3/8.11.4) with ESMTP id f8B737712693 for ; Tue, 11 Sep 2001 03:03:16 -0400 (EDT) (envelope-from news@news.tht.net) Received: (from news@localhost) by news.tht.net (8.11.4/8.11.4) id f8B706h72005 for pgsql-sql@postgresql.org; Tue, 11 Sep 2001 03:00:06 -0400 (EDT) (envelope-from news) From: "Gledatelj" X-Newsgroups: comp.database.ms-access, comp.database.oracle, comp.databases.postgresql.sql, comp.databases.sybase, comp.databases Subject: Re: Combine query views into one SQL string Date: Tue, 11 Sep 2001 08:33:30 +0200 Organization: CARNet, CROATIA Lines: 53 Message-ID: <9nkbg3$9db$1@bagan.srce.hr> References: <63e57aa8.0109060835.33719290@posting.google.com> <3B97DC1F.94433F65@mbit.nl> <9na15r$ku4$1@narses.hrz.tu-chemnitz.de> <63e57aa8.0109091520.4f975abd@posting.google.com> X-Complaints-To: abuse@carnet.hr X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 To: pgsql-sql@postgresql.org X-Archive-Number: 200109/82 X-Sequence-Number: 4738 Hy Maik, you can select it directly from the tables using following sql statement (but it's same as using views): select t1.id, t1.f1, t2.f2 from (select id, sum(amount) f1 from table1 group by id) t1, (select id, sum(amount) f2 from table2 group by id) t2 where t1.id = t2.id but, you must also know that in this way you may not get all the records (sum of amount of all id-s) if this is your intention, but just sum of amount of identical id-s (contained in both tables). Maybe this will help you. "Eimas" wrote in message news:63e57aa8.0109091520.4f975abd@posting.google.com... > "Maik" wrote in message news:<9na15r$ku4$1@narses.hrz.tu-chemnitz.de>... > > Its clear, union concat the two results. > > > > But he can also use this join version, if its the intention. > > > > select t1.id, sum(t1.amount), t2.id, sum(t2.amount) from table1 as t1, > > table2 as t2 where t1.id=t2.id; > > > > Ciao Maik > > This is not right, i wouldnt even bather you if this was the simple > answer. > > t1.id=t2.id would mean in doubled or tripled aggregation, > since t1.id and t2.id are not unique, thats why I had to aggregate > them in first Q1, Q2 querires, and link them by ID in the last one. > > I don't understand what UNION got to do here. then you probably have > to make three union links of inner, left and right joined tables. > > This is not that easy as it seems. > Is there a universal SQL string "expanding" rules, like in math or so? > > Thanks