Return-Path: X-Original-To: notmuch@notmuchmail.org Delivered-To: notmuch@notmuchmail.org Received: from localhost (localhost [127.0.0.1]) by arlo.cworth.org (Postfix) with ESMTP id 0E0846DE0130 for ; Fri, 1 Jul 2016 09:31:30 -0700 (PDT) X-Virus-Scanned: Debian amavisd-new at cworth.org X-Spam-Flag: NO X-Spam-Score: -0.011 X-Spam-Level: X-Spam-Status: No, score=-0.011 tagged_above=-999 required=5 tests=[AWL=0.000, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=disabled Received: from arlo.cworth.org ([127.0.0.1]) by localhost (arlo.cworth.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Is6Bw1l2xRTT for ; Fri, 1 Jul 2016 09:31:22 -0700 (PDT) Received: from fethera.tethera.net (fethera.tethera.net [198.245.60.197]) by arlo.cworth.org (Postfix) with ESMTPS id DFE5F6DE00E8 for ; Fri, 1 Jul 2016 09:31:21 -0700 (PDT) Received: from remotemail by fethera.tethera.net with local (Exim 4.84) (envelope-from ) id 1bJ1LI-0005dB-2U; Fri, 01 Jul 2016 12:31:00 -0400 Received: (nullmailer pid 13093 invoked by uid 1000); Fri, 01 Jul 2016 16:31:15 -0000 From: David Bremner To: Daniel Lee Harple , notmuch@notmuchmail.org Subject: Re: [PATCH] NEWS, python: add --libnotmuch-dir option to "setup.py install" In-Reply-To: References: User-Agent: Notmuch/0.22.1~rc0 (http://notmuchmail.org) Emacs/24.5.1 (x86_64-pc-linux-gnu) Date: Fri, 01 Jul 2016 18:31:15 +0200 Message-ID: <87furtgw18.fsf@maritornes.cs.unb.ca> MIME-Version: 1.0 Content-Type: text/plain X-BeenThere: notmuch@notmuchmail.org X-Mailman-Version: 2.1.20 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: Fri, 01 Jul 2016 16:31:30 -0000 Daniel Lee Harple writes: > This change fixes the python bindings on OS X 10.11, which introduced a > feature called System Integrity Protection. The /usr/bin/python > executable has the "restricted" flag set, which means dlopen() requires > an absolute path to the library that is to be loaded. Thanks for the patch. I'll let Justus (in copy) decide what the best way to deal with this issue is. I just wanted to point both of you (as an example) to commit 233b5528cd4 Which avoids re-computing the name of the shared library in the ruby bindings build system since it is already computed by the notmuch configure script. If possibly, I'd like to centralize these computations, so that we don't have to litter the entire source with various "if Darwin then blah" choices. And obviously it's better not to have update many places every time MacOS changes it's mind. I guess some people might like to build the bindings outside the notmuch source tree (via pypi or whatever). That's not a case I personally want to support, but I defer to Justus. d