Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA1:256) (Exim 4.89) (envelope-from ) id 1hK6J3-0006Pg-3c for pgsql-docs@arkaria.postgresql.org; Fri, 26 Apr 2019 19:14:45 +0000 Received: from localhost ([127.0.0.1] helo=malur.postgresql.org) by malur.postgresql.org with esmtp (Exim 4.89) (envelope-from ) id 1hK6J0-00061Q-CQ for pgsql-docs@arkaria.postgresql.org; Fri, 26 Apr 2019 19:14:42 +0000 Received: from magus.postgresql.org ([2a02:c0:301:0:ffff::29]) by malur.postgresql.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA1:256) (Exim 4.89) (envelope-from ) id 1hK6J0-0005zM-58 for pgsql-docs@lists.postgresql.org; Fri, 26 Apr 2019 19:14:42 +0000 Received: from sss.pgh.pa.us ([66.207.139.130]) by magus.postgresql.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA1:256) (Exim 4.89) (envelope-from ) id 1hK6It-0007ws-UH for pgsql-docs@lists.postgresql.org; Fri, 26 Apr 2019 19:14:41 +0000 Received: from sss1.sss.pgh.pa.us (localhost [127.0.0.1]) by sss.pgh.pa.us (8.14.4/8.14.4) with ESMTP id x3QJEWks010561; Fri, 26 Apr 2019 15:14:32 -0400 From: Tom Lane To: Alvaro Herrera cc: pgsql-docs@lists.postgresql.org Subject: Re: Documentation for partitioned indexes? In-reply-to: <20190426185333.GA12464@alvherre.pgsql> References: <20190426185333.GA12464@alvherre.pgsql> Comments: In-reply-to Alvaro Herrera message dated "Fri, 26 Apr 2019 14:53:33 -0400" MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-ID: <10559.1556306072.1@sss.pgh.pa.us> Date: Fri, 26 Apr 2019 15:14:32 -0400 Message-ID: <10560.1556306072@sss.pgh.pa.us> List-Id: List-Help: List-Subscribe: List-Post: List-Owner: List-Archive: Precedence: bulk Alvaro Herrera writes: > I suppose I better add something in Chapter 5 (DDL), possibly inside the > 5.10 section (Table Partitioning) -- a new 5.10.6 "Indexes for > Partitioned Tables" perhaps? I was expecting to find it in "5.11.2.2. Partition Maintenance". The structure of 5.11 seems a bit weird already, in that important (IMO) topics like preferred maintenance procedures are buried at a lower nesting level than essentially-historical trivia like "5.11.3. Implementation Using Inheritance". But maybe right now is not the time to redesign that. BTW, is there anything equivalent for unique/pkey constraints? I tried "add constraint unique ... not valid" and just got a raspberry. regards, tom lane