From patchwork Fri Oct 7 08:07:24 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Andy Patterson X-Patchwork-Id: 16327 Received: (qmail 79309 invoked by uid 89); 7 Oct 2016 08:07:59 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Checked: by ClamAV 0.99.2 on sourceware.org X-Virus-Found: No X-Spam-SWARE-Status: No, score=-4.6 required=5.0 tests=BAYES_00, RP_MATCHES_RCVD, SPF_PASS autolearn=ham version=3.3.2 spammy=nearer, Hx-languages-length:3999, snippet, 2016-10-07 X-Spam-Status: No, score=-4.6 required=5.0 tests=BAYES_00, RP_MATCHES_RCVD, SPF_PASS autolearn=ham version=3.3.2 X-Spam-Checker-Version: SpamAssassin 3.3.2 (2011-06-06) on sourceware.org X-Spam-Level: X-HELO: lists.gnu.org Received: from lists.gnu.org (HELO lists.gnu.org) (208.118.235.17) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with ESMTP; Fri, 07 Oct 2016 08:07:49 +0000 Received: from localhost ([::1]:32985 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bsQC2-00036V-Gb for patchwork@sourceware.org; Fri, 07 Oct 2016 04:07:46 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:42157) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bsQBw-00035L-9H for guix-devel@gnu.org; Fri, 07 Oct 2016 04:07:41 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bsQBq-0003qD-4c for guix-devel@gnu.org; Fri, 07 Oct 2016 04:07:39 -0400 Received: from mailservices.uwaterloo.ca ([129.97.128.141]:39691 helo=mailchk-m04.uwaterloo.ca) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bsQBp-0003mV-St for guix-devel@gnu.org; Fri, 07 Oct 2016 04:07:34 -0400 Received: from localhost (bas1-jockvale05-1176150086.dsl.bell.ca [70.26.160.70]) (authenticated bits=0) by mailchk-m04.uwaterloo.ca (8.14.4/8.14.4) with ESMTP id u9787OiL021117 (version=TLSv1/SSLv3 cipher=AES256-GCM-SHA384 bits=256 verify=NO); Fri, 7 Oct 2016 04:07:28 -0400 Date: Fri, 7 Oct 2016 04:07:24 -0400 From: Andy Patterson To: iyzsong@member.fsf.org (=?UTF-8?B?5a6L5paH5q2m?=) Subject: Re: [PATCH v2 01/13] build-system: Add asdf-build-system. Message-ID: <20161007040724.6b28b18d@uwaterloo.ca> In-Reply-To: <20161005165907.3e5ed2c2@uwaterloo.ca> References: <20160927041532.27097-1-ajpatter@uwaterloo.ca> <20161003024139.19975-1-ajpatter@uwaterloo.ca> <20161003024139.19975-2-ajpatter@uwaterloo.ca> <87ponf9yy0.fsf@member.fsf.org> <20161005165907.3e5ed2c2@uwaterloo.ca> X-Mailer: Claws Mail 3.14.0 (GTK+ 2.24.30; x86_64-unknown-linux-gnu) MIME-Version: 1.0 X-UUID: 510043a3-8659-47f2-8f79-db4301966384 X-Miltered: at mailchk-m04 with ID 57F757BD.001 by Joe's j-chkmail (http://j-chkmail.ensmp.fr)! X-Virus-Scanned: clamav-milter 0.99.1 at mailchk-m04 X-Virus-Status: Clean X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.5.9 (mailchk-m04.uwaterloo.ca [129.97.128.141]); Fri, 07 Oct 2016 04:07:28 -0400 (EDT) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 3.x X-Received-From: 129.97.128.141 X-BeenThere: guix-devel@gnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: guix-devel@gnu.org Errors-To: guix-devel-bounces+patchwork=sourceware.org@gnu.org Sender: "Guix-devel" Hi again, Things took a little longer than expected as I ran into a bug with ECL (fix attached). I'll try to upstream it later. It can be re-ordered somewhere nearer the beginning of the series. On Wed, 5 Oct 2016 16:59:07 -0400 Andy Patterson wrote: > > What’s the ‘compile-dependencies’ used for? I think when compile a > > system, ASDF will load all its depends (from the :depends-on) first. > > Sometimes we need load extra systems manually? > > > > It's for packages like slynk, where many systems are defined within > the same file. When telling asdf to find a system, it will only > search for a file with the same name unless it already knows about > that system. Therefore, in these cases the system which contains the > definitions is loaded first. This mechanism is a bit clunky so > suggestions are welcome. For now I'll document the reason it's done > that way. > I found a better way to solve this. I hope I've managed to cover everything you've suggested. The slynk packages look quite a bit nicer now. I'll attach the updates as replies to the corresponding v2 patches. Let me know what you think. --- Andy From 7646691c77c9a01e620e5f0c4695db8abb276863 Mon Sep 17 00:00:00 2001 From: Andy Patterson Date: Fri, 7 Oct 2016 03:19:26 -0400 Subject: [PATCH v3 12/12] gnu: ecl: Fix `truename' on long target files. When the target of a symbolic link exceeded 128 characters, `truename' would unexpectedly fail. * gnu/packages/patches/ecl-fix-si-readlink.patch: New file. * gnu/local.mk (dist_patch_DATA): Add it. --- gnu/local.mk | 1 + gnu/packages/lisp.scm | 4 +++- gnu/packages/patches/ecl-fix-si-readlink.patch | 20 ++++++++++++++++++++ 3 files changed, 24 insertions(+), 1 deletion(-) create mode 100644 gnu/packages/patches/ecl-fix-si-readlink.patch diff --git a/gnu/local.mk b/gnu/local.mk index 206b794..d057d20 100644 --- a/gnu/local.mk +++ b/gnu/local.mk @@ -497,6 +497,7 @@ dist_patch_DATA = \ %D%/packages/patches/doxygen-test.patch \ %D%/packages/patches/duplicity-piped-password.patch \ %D%/packages/patches/duplicity-test_selection-tmp.patch \ + %D%/packages/patches/ecl-fix-si-readlink.patch \ %D%/packages/patches/elfutils-tests-ptrace.patch \ %D%/packages/patches/elixir-disable-failing-tests.patch \ %D%/packages/patches/einstein-build.patch \ diff --git a/gnu/packages/lisp.scm b/gnu/packages/lisp.scm index 458862c..7068800 100644 --- a/gnu/packages/lisp.scm +++ b/gnu/packages/lisp.scm @@ -130,7 +130,9 @@ interface to the Tk widget system.") (snippet ;; Add ecl-bundle-systems to 'default-system-source-registry'. `(substitute* "contrib/asdf/asdf.lisp" - ,@(asdf-substitutions name))))) + ,@(asdf-substitutions name))) + (patches + (search-patches "ecl-fix-si-readlink.patch")))) (build-system gnu-build-system) ;; src/configure uses 'which' to confirm the existence of 'gzip'. (native-inputs `(("which" ,which))) diff --git a/gnu/packages/patches/ecl-fix-si-readlink.patch b/gnu/packages/patches/ecl-fix-si-readlink.patch new file mode 100644 index 0000000..5cd0efc --- /dev/null +++ b/gnu/packages/patches/ecl-fix-si-readlink.patch @@ -0,0 +1,20 @@ +--- a/src/c/unixfsys.d 2016-02-25 02:06:19.000000000 -0500 ++++ b/src/c/unixfsys.d 2016-10-07 00:50:20.172282246 -0400 +@@ -236,15 +236,15 @@ static cl_object + si_readlink(cl_object filename) { + /* Given a filename which is a symlink, this routine returns + * the value of this link in the form of a pathname. */ +- cl_index size = 128, written; ++ cl_index size = 0, written; + cl_object output, kind; + do { ++ size += 128; + output = ecl_alloc_adjustable_base_string(size); + ecl_disable_interrupts(); + written = readlink((char*)filename->base_string.self, + (char*)output->base_string.self, size); + ecl_enable_interrupts(); +- size += 256; + } while (written == size); + output->base_string.self[written] = '\0'; + kind = file_kind((char*)output->base_string.self, FALSE); -- 2.10.0