Return-Path: X-Original-To: apmail-subversion-dev-archive@minotaur.apache.org Delivered-To: apmail-subversion-dev-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id C28061131C for ; Fri, 23 May 2014 11:48:40 +0000 (UTC) Received: (qmail 70441 invoked by uid 500); 23 May 2014 11:48:40 -0000 Delivered-To: apmail-subversion-dev-archive@subversion.apache.org Received: (qmail 70390 invoked by uid 500); 23 May 2014 11:48:40 -0000 Mailing-List: contact dev-help@subversion.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Delivered-To: mailing list dev@subversion.apache.org Received: (qmail 70383 invoked by uid 99); 23 May 2014 11:48:40 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 23 May 2014 11:48:40 +0000 X-ASF-Spam-Status: No, hits=-0.7 required=5.0 tests=RCVD_IN_DNSWL_LOW,SPF_PASS X-Spam-Check-By: apache.org Received-SPF: pass (nike.apache.org: domain of philip.martin@wandisco.com designates 74.125.82.172 as permitted sender) Received: from [74.125.82.172] (HELO mail-we0-f172.google.com) (74.125.82.172) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 23 May 2014 11:48:36 +0000 Received: by mail-we0-f172.google.com with SMTP id k48so4816232wev.17 for ; Fri, 23 May 2014 04:48:12 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=wandisco.com; s=gapps; h=from:to:cc:subject:references:date:in-reply-to:message-id :user-agent:mime-version:content-type; bh=OkIcI0T/y36fPsoVIxJzTK2lvnTj+FMl8BSqNsGFKJ8=; b=iTPmnzfb6D1FbbAnPbzg4qwGZWrEEmPjdyAh29y0ClEDDZuqrx1Q7rvDTuMBfB2c3r U7UE/ZuTar8kIJP2+VUUXIvLJRy+MLUW894cvBdysiZxS9VexTLbI0YYq9sQtB+PoBlr qlVzfnncruZVPuUY1lTUxKL1rYR7cv15OSkco= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:from:to:cc:subject:references:date:in-reply-to :message-id:user-agent:mime-version:content-type; bh=OkIcI0T/y36fPsoVIxJzTK2lvnTj+FMl8BSqNsGFKJ8=; b=kvU+wD67VOmnsALaJzIeMTJkwDfKeceaOgCzatP7F84obYpT+8KwiTFKA2i/uE3rO6 tsq5tyF9cOYqRFg8hoSUakp6nydFg2zHt6wa/Q/32BscmVkblgt0mOP3x5ITtHGQN7ac x65WHZ0AAQWFT094mYQzqRiu11NoooaIBO4DYGmx7xi/vVh5qNcrI7Pho5M7p2/5Xavz U8u+s2oAQU0Tl3GaY3/UKRAMdTTynlSQ9kLCV2RC5fVjUgoCDLHQ9fs88Ze2aeXcdYpf UCAkpesYfrqCkqYN+wzKHfeRebUnE9QzXqR4uwZSnODlJNE7bXhtUcszQm7PmGWa1DX5 fPKw== X-Gm-Message-State: ALoCoQlm4IA12mrMDX5SKylZqOP6pfBc0qIU2w6B6tWX1mtIQ2fho/4w95bIl0lHbnhRFT3s+vMU X-Received: by 10.180.106.194 with SMTP id gw2mr2729317wib.47.1400845692596; Fri, 23 May 2014 04:48:12 -0700 (PDT) Received: from localhost (cpc20-farn7-2-0-cust13.6-2.cable.virginm.net. [86.15.228.14]) by mx.google.com with ESMTPSA id nb8sm2524846wic.18.2014.05.23.04.48.11 for (version=TLSv1.2 cipher=RC4-SHA bits=128/128); Fri, 23 May 2014 04:48:11 -0700 (PDT) From: Philip Martin To: James McCoy Cc: dev@subversion.apache.org Subject: Re: Ruby 2.0 support References: <20140401023849.GE2219@jamessan.com> <20140401065657.GA2198@tarsus.local2> <20140517141726.GH5105@jamessan.com> <871tvlc2xk.fsf@ntlworld.com> Date: Fri, 23 May 2014 12:48:11 +0100 In-Reply-To: <871tvlc2xk.fsf@ntlworld.com> (Philip Martin's message of "Thu, 22 May 2014 15:18:15 +0100") Message-ID: <87bnuo90n8.fsf@ntlworld.com> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.4 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Virus-Checked: Checked by ClamAV on apache.org Philip Martin writes: > James McCoy writes: > >> Updated patch attached. > > I tried Ruby 1.8.7 and 2.1.2 and the regression tests pass so I > committed this in r1596882. Thanks! I think our Ruby installation stuff is a bit broken. We have --with-ruby-sitedir=SITEDIR install Ruby bindings in SITEDIR (default is same as ruby's one) Ruby, at least since 1.8.7, has both sitedir and vendordir. vendordir is the system installation, typically under /usr, while sitedir is the local stuff, typically under /usr/local. I have Debian/wheezy 1.8.7 ruby and it has sitedir /usr/local/share/ruby/site_ruby sitelibdir /usr/local/lib/site_ruby/1.8 sitearchdir /usr/local/lib/site_ruby/1.8/x86_64-linux vendordir /usr/lib/ruby/vendor_ruby vendorlibdir /usr/lib/ruby/vendor_ruby/1.8 vendorarchdir /usr/lib/ruby/vendor_ruby/1.8/x86_64-linux I have Debian/wheezy 1.9.1/1.9.3 ruby and it has sitedir /usr/local/share/ruby/site_ruby sitelibdir /usr/local/lib/site_ruby/1.9.1 sitearchdir /usr/local/lib/site_ruby/1.9.1/x86_64-linux vendordir /usr/lib/ruby/vendor_ruby vendorlibdir /usr/lib/ruby/vendor_ruby/1.9.1 vendorarchdir /usr/lib/ruby/vendor_ruby/1.9.1/x86_64-linux Note that libdir is a prefix of sitedir for both vendor and site. That allows us to generate installation paths when --with-ruby-sitedir is used: AC_MSG_CHECKING([where to install Ruby extensions]) AC_CACHE_VAL([svn_cv_ruby_sitedir_archsuffix],[ svn_cv_ruby_sitedir_archsuffix="`echo "$rbconfig_sitearchdir" | \ $SED -e "s,^$rbconfig_sitedir,,"`" ]) SWIG_RB_SITE_ARCH_DIR="${svn_ruby_installdir}${svn_cv_ruby_sitedir_archsuffix}" AC_MSG_RESULT([$SWIG_RB_SITE_ARCH_DIR]) However that breaks down when libdir is not a prefix of sitedir. I have Debian/sid 2.1.2 ruby and it has sitedir /usr/local/lib/site_ruby sitelibdir /usr/local/lib/site_ruby/2.1.0 sitearchdir /usr/local/lib/x86_64-linux-gnu/site_ruby vendordir /usr/lib/ruby/vendor_ruby vendorlibdir /usr/lib/ruby/vendor_ruby/2.1.0 vendorarchdir /usr/lib/x86_64-linux-gnu/ruby/vendor_ruby/2.1.0 I have Fedora/20 2.0.0 and it has sitedir /usr/local/share/ruby/site_ruby sitelibdir /usr/local/share/ruby/site_ruby/ sitearchdir /usr/local/lib64/ruby/site_ruby vendordir /usr/share/ruby/vendor_ruby vendorlibdir /usr/share/ruby/vendor_ruby/ vendorarchdir /usr/lib64/ruby/vendor_ruby On these systems we get the wrong installation directories. If I use ./configure --with-ruby-sitedir=/home/pm/ruby the 1.8.7 system gives a reasonable looking: checking where to install Ruby extensions... /home/pm/ruby/1.8/x86_64-linux while the 2.1.2 system gives a very odd looking: checking where to install Ruby extensions... /home/pm/ruby/usr/local/lib/x86_64-linux-gnu/site_ruby I'm not sure whether the different layout is due to the more modern Ruby or the more modern OS. -- Philip Martin | Subversion Committer WANdisco // *Non-Stop Data*