From owner-freebsd-apache@freebsd.org Wed Sep 16 17:45:53 2015 Return-Path: Delivered-To: freebsd-apache@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 794D89CECD8 for ; Wed, 16 Sep 2015 17:45:53 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from mailman.ysv.freebsd.org (mailman.ysv.freebsd.org [IPv6:2001:1900:2254:206a::50:5]) by mx1.freebsd.org (Postfix) with ESMTP id 6341F1418 for ; Wed, 16 Sep 2015 17:45:53 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: by mailman.ysv.freebsd.org (Postfix) id 5F69D9CECD7; Wed, 16 Sep 2015 17:45:53 +0000 (UTC) Delivered-To: apache@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 5F0019CECD6 for ; Wed, 16 Sep 2015 17:45:53 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 36A8F1410 for ; Wed, 16 Sep 2015 17:45:53 +0000 (UTC) (envelope-from bugzilla-noreply@freebsd.org) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.15.2/8.15.2) with ESMTP id t8GHjrX2003224 for ; Wed, 16 Sep 2015 17:45:53 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: apache@FreeBSD.org Subject: maintainer-feedback requested: [Bug 203157] www/apache24: mistakenly tries to build expat if old magic files exist while citing a wrong filename Date: Wed, 16 Sep 2015 17:45:51 +0000 X-Bugzilla-Type: request Message-ID: In-Reply-To: References: X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" X-BeenThere: freebsd-apache@freebsd.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: Support of apache-related ports List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Sep 2015 17:45:53 -0000 freebsdpr-20121114@billmail.scconsult.com has reassigned Bugzilla Automation 's request for maintainer-feedback to apache@FreeBSD.org: Bug 203157: www/apache24: mistakenly tries to build expat if old magic files exist while citing a wrong filename https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=203157 --- Description --- I tried building ("make build" in www/apache24) apache24 from the port (to get LDAP support not in the binary package) and it chased the textproc/expat2 dependency and tried to install it from the port even though there was a perfectly usable binary package of expat already installed. The output around the detour was: ===> apache24-2.4.16_1 depends on executable: libtoolize - found ===> apache24-2.4.16_1 depends on shared library: libexpat.soEither /usr/share/misc/magic.mime or /usr/share/misc/mime.magic.mgc exist and must be removed. These are legacy files from an older release and may safely be deleted. Please see UPDATING 20150213 for more details. - not found ===> expat-2.1.0_3 depends on file: /usr/local/sbin/pkg - found Note the lack of space after libexpat.so and the 2 file names. I removed /usr/share/misc/magic.mime but /usr/share/misc/mime.magic.mgc did not exist. However /usr/share/misc/magic.mime.mgc DID exist, with the same mtime as /usr/share/misc/magic.mime (suspicious...) "make clean ; make build" resulted in the same detour into building expat with the same error message, even though no such files existed. I removed the suspect file, tried again, and the build went perfectly. This seems like it is maybe 2 bugs, but I'm not sure: 1. /usr/ports/Mk/Scripts/find-lib.sh looks for files of the correct names, emits an error message with one of them mangled. 2. The way /usr/ports/Mk/Scripts/find-lib.sh fails, the message it emits, and the resulting upstream reaction of chasing a dependency unnecessarily is not a recipe for least harm. If those obsolete files are so dangerous that they must break port builds, perhaps the check for them should live somewhere higher up where it can be made to break every port build from the start unconditionally.