From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-lj1-f196.google.com (mail-lj1-f196.google.com [209.85.208.196]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by dev.tarantool.org (Postfix) with ESMTPS id 0B312469719 for ; Tue, 18 Feb 2020 22:11:26 +0300 (MSK) Received: by mail-lj1-f196.google.com with SMTP id q8so24217688ljj.11 for ; Tue, 18 Feb 2020 11:11:26 -0800 (PST) Date: Tue, 18 Feb 2020 22:11:25 +0300 From: Konstantin Osipov Message-ID: <20200218191125.GD20569@atlas> References: <20200218175511.55344-1-k.sosnin@tarantool.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20200218175511.55344-1-k.sosnin@tarantool.org> Subject: Re: [Tarantool-patches] [PATCH] sql: fix update of a self-referenced table List-Id: Tarantool development patches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Chris Sosnin Cc: tarantool-patches@dev.tarantool.org * Chris Sosnin [20/02/18 20:56]: > An update of a self-referenced table is not handled correctly: > if a new row doesn't reference itself, it is considered to violate > the constraint. This patch makes vdbe to inspect the table on updates > too with a little change: the row we find should be different from the > one we are updating. How is tarantool different from upstream that it needs this fix? Why is the bug not present in the upstream (I assume it isn't, it's quite obvious)? Did you file a bug in the upstream if it's present in it? -- Konstantin Osipov, Moscow, Russia