From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from localhost (localhost [127.0.0.1]) by olra.theworths.org (Postfix) with ESMTP id 1045D431FBD for ; Mon, 27 Jan 2014 09:40:27 -0800 (PST) X-Virus-Scanned: Debian amavisd-new at olra.theworths.org X-Spam-Flag: NO X-Spam-Score: 0 X-Spam-Level: X-Spam-Status: No, score=0 tagged_above=-999 required=5 tests=[none] autolearn=disabled Received: from olra.theworths.org ([127.0.0.1]) by localhost (olra.theworths.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1TEPkg0bqOJC for ; Mon, 27 Jan 2014 09:40:19 -0800 (PST) Received: from yantan.tethera.net (yantan.tethera.net [199.188.72.155]) (using TLSv1 with cipher DHE-RSA-AES128-SHA (128/128 bits)) (No client certificate requested) by olra.theworths.org (Postfix) with ESMTPS id 57851431FBC for ; Mon, 27 Jan 2014 09:40:19 -0800 (PST) Received: from remotemail by yantan.tethera.net with local (Exim 4.80) (envelope-from ) id 1W7qAU-0006qi-21; Mon, 27 Jan 2014 13:40:18 -0400 Received: (nullmailer pid 6171 invoked by uid 1000); Mon, 27 Jan 2014 17:40:14 -0000 From: David Bremner To: Gregor Zattler , notmuch Subject: Re: Bug: notmuch new chokes on dangling symlinks and quits In-Reply-To: <20140127011427.GD10844@boo.workgroup> References: <20140126131846.GA10844@boo.workgroup> <878uu2n17j.fsf@zancas.localnet> <20140127011427.GD10844@boo.workgroup> User-Agent: Notmuch/0.17+53~g3e1d7f6 (http://notmuchmail.org) Emacs/24.3.1 (x86_64-pc-linux-gnu) Date: Mon, 27 Jan 2014 13:40:14 -0400 Message-ID: <878uu1l3wh.fsf@zancas.localnet> MIME-Version: 1.0 Content-Type: text/plain X-BeenThere: notmuch@notmuchmail.org X-Mailman-Version: 2.1.13 Precedence: list List-Id: "Use and development of the notmuch mail system." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 27 Jan 2014 17:40:27 -0000 Gregor Zattler writes: > Hi David, > * David Bremner [26. Jan. 2014]: >> Gregor Zattler writes: >>> I consider this to be a bug. Instead notmuch should simply >>> ignore the symlink. >>> >> >> Since there is a test for specifically this behaviour, I'd have to say >> it's a design decision you don't agree with, not a bug ;). > > May I ask why this is so? > Purely from memory (I wasn't involved, and didn't dig up the the discussion): - a common use case is linking different trees into one notmuch-index tree. - if a subtree disappears (e.g. by a network failure), then the choices are stop the index or ignore the missing files. - in the latter case, all tags from "deleted" messages are lost So, a simple solution which avoids data loss is to abort the index process. A more complicated solution would be possible of course, but nobody proposed it (or more importantly, did it) yet. d