Re: [Slackbuilds-users] pdfstudio and pdfstudioviewer up for grabs
On Fri, 25 Jun 2021, Alexander Verbovetsky wrote: > pdfstudio/pdfstudioviewer upstream stopped providing deb packages The download page still looks like there might be a debian package, »To install PDF Studio on multiple computers through command-line, you may use our 64 bit Debian package (right-click and save link as)«, but the download link is broken. I e-mailed their support about this, and as expected the answer was utterly useless, simply stating "there is no Debian package", completely ignoring the info and the broken link from the download page. I guess that's it then. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org https://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
[Slackbuilds-users] separating source archives and SlackBuilds
Hello there!
There's a certain feature in FreeBSD ports that I think I'd like to have
in SlackBuilds, too, and I want to know what you think about it:
FreeBSD ports keep all source archives in the /usr/ports/distfiles folder
while SlackBuilds expect the source archive in the same folder as the
SlackBuld itself.
With a simple "DISTFILES=${DISTFILES:-$CWD}" and "tar xvf $DISTFILES/..."
rather than "tar xvf $CWD/..." in the template we could have both ways.
The reason for keeping it the way it is: simplicity.
The reason for having a separate distfiles folder: you can keep as many
source archive versions around as you want without affecting the way you
work with the SlackBuilds directory tree, e.g. always using a pristine and
possibly read-only copy, or going all-in on git without ever accidentally
adding and comitting the LibreOffice tar ball.
Would this be desirable and feasible?
Cheers!
--
___
SlackBuilds-users mailing list
SlackBuilds-users@slackbuilds.org
https://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users
Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/
FAQ - https://slackbuilds.org/faq/
[Slackbuilds-users] Exim 4.94.2 security upgrade
Hello there, Today Exim 4.94.2 was released, fixing a number of local and remote vulnerabilities. An upgrade is highly recommened. See the announcement for full details and additional information for users of versions older than 4.94: Announcement: https://marc.info/?l=exim-users=162013704804315=2 Because submissions are closed there won't be an official update of my SlackBuild for some time, but you can either use the existing SlackBuild setting VERSION accordingly or get an updated SlackBuild from here: https://www.beingboiled.info/file/exim-4.94.2-SlackBuild.tar.gz Cheers! -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org https://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
[Slackbuilds-users] Submission confirmation mails have no Message-ID
Hello there, I just noticed that submission confirmation mails have no Message-ID. In fact the Message-ID is optional. I didn't know that. However, the RFC also states that the field SHOULD be present. In my experience a missing Message-ID can cause some interesting problems and is usually punished with a bad score in common spam filters. If the submission form doesn't generate a Message-ID, the first MTA should. This could solve some of the occational deliverability problems. Cheers! -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org https://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] IPv6 issue with git.slackbuilds.org
On Sun, 25 Oct 2020, Robby Workman wrote: >> since the SlackBuilds.org servers have been moved I have to get >> updates with "git pull -4". Can you open port 9418 for IPv6? > Oops. I wish you'd told us sooner. Fixed :-) While you're at it, can you open port 873 (rsync) for IPv6 on the slackbuilds.org main site, too? Cheers! -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org https://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] 3 Slackbuilds looking for new maintainer
On Mon, 8 Oct 2018, Willy Sudiarto Raharjo wrote: > I will take slack Any progress on that? The info file still lists Harald as the maintainer and there has been no update since. The old 3.1.1 version works well enough but propably has a number of security issues by now with its "bundled everything" approach. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org https://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
[Slackbuilds-users] mupdf and dynamic vs. static linking
Hello there, Something I got reminded of by the recent update... By default the mupdf software package builds and links against a static library, resulting in a 123MB txz package containing several 33MB executables. The LFS project has a patch that applies cleanly and switches to a dynamic build, resulting in a much more sensible 20MB txz package. http://www.linuxfromscratch.org/blfs/view/svn/pst/mupdf.html http://www.linuxfromscratch.org/patches/blfs/svn/mupdf-1.13.0-shared_libs-1.patch I've been using the LFS patch for quite a while over several versions of mupdf without any problems. Could the LFS patch be added to the SlackBuild? Cheers! -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org https://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] Request: please add a build option to skip running tests for perl-http-server-simple SBo buildscript
On Mon, 18 Dec 2017, Willy Sudiarto Raharjo wrote: >> In the past I have also asked the maintainer to add such a build >> option, but without success. > Done in my branch Maybe that was a bit too quick... I think the test is fine and the error is in OP's network config where localhost and 127.0.0.1 don't match. I've tried this in a VM and the test worked fine. I got the same error message, though, when I deliberately set localhost to eth0's address instead of 127.0.0.1 in /etc/hosts. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org https://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
[Slackbuilds-users] Exim remote code execution
Hello there, A remote code execution vulnerability has been reported in Exim. If you're running Exim, disable the ESMTP CHUNKING extension by putting... chunking_advertise_hosts = ...into the main section of your exim.conf. This should do as a workaround until a proper fix is available. https://marc.info/?l=exim-announce=151158351116562=2 -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org https://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
[Slackbuilds-users] libupnp security issue
Hello there, The SlackBuild for libupnp hasn't been updated for more than five years, omitting several security fixes and leaving the resulting package vulnerable to CVE-2016-6255: »Portable UPnP SDK (aka libupnp) before 1.6.21 allows remote attackers to write to arbitrary files in the webroot via a POST request without a registered handler.« This build should propably get an upgrade (and a new maintainer). -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] Update - 20170606.1
On Wed, 7 Jun 2017, Willy Sudiarto Raharjo wrote: > I have applied your suggestion > please review the changes here > https://git.slackbuilds.org/slackbuilds/commit/?id=fc810ed28 Builds and works fine, thanks! -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] Update - 20170606.1
On Tue, 6 Jun 2017, Willy Sudiarto Raharjo wrote: > network/slack: Fix path on x86_64. This broke the package because the /usr/bin/slack symlink still points to the then non-existing /usr/lib/slack/slack. Also, the slack binaries don't support anything else than x86_64 currently, so there's no need to test $ARCH. A simple rename+relink should do. # Fix path mv $PKG/usr/lib $PKG/usr/lib64 ( cd $PKG/usr/bin ; rm -f slack ; ln -s ../lib64/slack/slack ) -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
[Slackbuilds-users] Exim & CVE-2016-9963
Happy holidays, everyone! Two new versions of the Exim Internet Mailer were released today to fix a security issue where DKIM keys can be leaked to log files and remote hosts if several conditions are met. For details see: https://exim.org/static/doc/CVE-2016-9963.txt I've prepared builds for the fixed versions. These builds are somewhat improvised, though, as I currently don't have access to my usual build and test environment. If you use DKIM signing, have a look at https://beingboiled.info/file/exim-4.87.1.tar.gz or https://beingboiled.info/file/exim-4.88.tar.gz If you don't use DKIM the existing 4.87 should be safe. Nevertheless, I'll submit a regular update within the next few days. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] qcomicbook link does not exist
On Thu, 25 Aug 2016, Matteo Bernardini wrote: > that's strange: I just have this warning starting qcomicbook from the > terminal here > > libpng warning: iCCP: known incorrect sRGB profile That's interesting. I don't get a libpng warning message on Slackware 14.2. I get a different libpng warning message on Slackware 14.1 (never seen this before). I don't get a warning message on either regarding the desktop file. I've definitely seen this on 14.1 but can't reproduce it any longer. However, checking with "desktop-file-validate" confirms: /usr/share/applications/qcomicbook.desktop: error: (will be fatal in the future): value "qcomicbook.png" for key "Icon" in group "Desktop Entry" is an icon name with an extension, but there should be no extension as described in the Icon Theme Specification if the value is not an absolute path -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] qcomicbook link does not exist
On Wed, 24 Aug 2016, Matteo Bernardini wrote: > I have already updated it on my git branch and will go in the next public > update > https://slackbuilds.org/cgit/slackbuilds/commit/?id=47251c6e488807d50ae53d5f92098150aabd0096 While you're at it, do you want to add... # Icon info in the desktop file should not have a file extension. sed -i 's/\.png$//' data/qcomicbook.desktop ...to the SlackBuild? It's just a cosmetical fix to get rid of the warning message when starting qcomicbook from the terminal. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] Am trying to build a perl module, perl-gd, in order to bring shutter up-to-date
On Sun, 21 Aug 2016, B Watson wrote: > Doesn't need a patch, just different options to Build. You're right. To my excuse: the use of "perl Build.PL --prefix ..." is documented and GD certainly breaks this. Thereafter I didn't investigate any further and simply patched. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] Am trying to build a perl module, perl-gd, in order to bring shutter up-to-date
On Sun, 21 Aug 2016, unix_li...@f-m.fm wrote: > Is someone willing to build a perl module for me, perl-GD, and submit > it to slackbuilds so that I can update shutter? I do NOT want to become the maintainer of perl-GD, but I do know its pitfalls, so I prepared a SlackBuild for anyone willing to pick it up, see attachment. Just insert your name and e-mail address :-) Some notes... The tests don't work. The Makefile.PL doesn't work at all. The Build.PL will work but needs a patch because the use of Getopt renders the usual Build.PL options useless. My patch was made for Slackware 14.1 and also trims the required version down to 0.40 to work with what was a core module then, but since Module-Build is an external dependency now you can leave this out. I didn't have a need for the scripts and was too lazy to clean up $PKG/usr/bin, so I simply deleted it. As a careful maintainer you should not do that :-) Have fun! -- perl-GD.tar.gz Description: Binary data ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] Fwd: monit change request
On Mon, 8 Aug 2016, Bryan Harris wrote: > Interesting re: the restart command. I think it's fixed by using > "monit quit" for the implementation of /etc/rc.d/rc.monit stop It can't be, because "monit quit" just sends a SIGTERM to the monit daemon and thus is functionally the same as a "kill $(cat monit.pid)" or similar. Your approach is still better, though. The previous rc-script would kill any monit, including user instances. Then again... "reload" works just fine and a "restart" is practically never needed, so I'd consider a few seconds of sleep between stop and start as "good enough". -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] Fwd: monit change request
On Sat, 6 Aug 2016, Sebastian Arcus wrote: > I can't help but think that this is a little bit poor design from the part of > the Monit development team. What is the point in having a fully configurable > config file name and location, if half of your software is hard-coded to use a > particular config file name and location? There's nothing wrong or even unusual with monit's design in that respect. If you want to use the command line switch to specify a config file location other than the default you have to specify it each time. Otherwise "monit $command" simply can't know how to talk to the daemon started with "monit -c /i/want/this.conf". The -c option is propably more useful for testing and development, i.e. without affecting your running instance you can test a monit config file for syntax errors and then deploy it somewhere else. > The other problem I have with monit at the moment is that if you do > /etc/rc.d/rc.monit restart, most of the time this will kill the daemon > completely. What appears to be happening is that monit is taking a long time > to shut down - which means the start command is attempted while it is still > shutting down - so it never starts again. I can confirm this. It seems to be connected to the event notification, i.e. shutdown takes as long as is takes to send an e-mail + overhead, but I'm not entirely sure. A "sleep 4" between stop and start solved this for me so I didn't investigate any further. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - https://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - https://slackbuilds.org/faq/
Re: [Slackbuilds-users] upgrade to 14.2
On Mon, 1 Aug 2016, Gilles Chanteperdrix wrote: > - the exim slackbuild tries to download the 4.86.2 tarball, but this > version is no longer the latest stable so the URL in the slackbuild > is no longer valid, it should be: > ftp://ftp.exim.org/pub/exim/exim4/old/exim-4.86.2.tar.bz2 No more need for that because I submitted an update to Exim 4.87 just yesterday :-) -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Community Challenge! Help fix the last few SlackBuilds for 14.2
On Mon, 30 May 2016, David Spencer wrote: > 1. academic/Xyce > > http://pastebin.com/nsMwxhgR > > It looks like an underlinking error, but I can't see anything wrong > with the link. There is a new version 6.4 upstream, I have tried it, > but the same linking problem happens. Maybe its dep > libraries/trilinos is involved This one is a bit tricky... the problem indeed is trilinos which can't be built with neither "atlas" (as stated in the info file) nor "blas" and "lapack" as currently present on SBo. atlas is a very fragile library and highly dependent on certain compiler versions and CPU timings! The build is very old and the README is unclear on how to merge the atlas and lapack libraries to get the lapack functions that atlas doesn't cover. On the other hand, blas and lapack 3.6.0 are too new for trilinos. 3.6.0 replaced some deprecated functions, but the wrapper in trilinos still uses the old function names, resulting in a libepetra.so with unresolved symbols that will make Xyce fail. So, this gets us the following options: 1) Ditch atlas and build trilinos with blas and lapack downgraded to version 3.5.0. Easy, unless this breaks other packages. 2) Ditch atlas and patch trilinos to work with blas and lapack in version 3.6.0. There's no readymade patch available and I don't know how to do this as this involves FORTRAN code. 3) Bring atlas up to date and merge it with lapack for a fully functional high-performance library. Ideally done by a maintainer who understands how atlas works, aka "not me". Any other options? -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Community Challenge! Help fix the last few SlackBuilds for 14.2
On Mon, 30 May 2016, David Spencer wrote: > 4. graphics/photivo > > http://pastebin.com/v5Yt3SYV > > This looks like an underlinking error, but the build fails for me only > on x86_64 ... on i586 it builds ok! I made this one build on x86_64 with the help of an extra compiler flag: env CFLAGS="$SLKCFLAGS -fno-finite-math-only" \ CXXFLAGS="$SLKCFLAGS -fno-finite-math-only" make ...instead of the simple "make" in line 76. The resulting package runs without crashing. I don't know if it actually *works* as the tool is strange to me. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] Exim CVE-2016-1531
Dear Exim-users and SBo-admins,
A new version of Exim has been released, fixing a vulnerability that
allows a local privilege escalation. While the Exim from SBo doesn't
include the vulnerable feature and thus *should* not be affected, the
download link in the .info file needs editing and an upgrade to the
current Exim 4.86.2 seems desirable.
If you're still okay with Exim 4.86 you can find the archive at:
ftp://ftp.exim.org/pub/exim/exim4/old/exim-4.86.tar.bz2
If you prefer to upgrade to Exim 4.86.2, simply replace exim.SlackBuild
and exim.info with the attached files and build. Please note that the new
version contains changes that might break your existing config. See...
http://article.gmane.org/gmane.mail.exim.announce/168
...for more information.
-- #!/bin/sh
# Slackware build script for Exim
# Copyright 2012-2016 Thomas Morper, Augsburg, Germany
# All rights reserved.
#
# Redistribution and use of this script, with or without modification, is
# permitted provided that the following conditions are met:
#
# 1. Redistributions of this script must retain the above copyright
#notice, this list of conditions and the following disclaimer.
#
# THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY EXPRESS OR IMPLIED
# WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
# MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
# EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
# OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
# WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
# OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
# ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
PRGNAM=exim
VERSION=${VERSION:-4.86.2}
BUILD=${BUILD:-1}
TAG=${TAG:-_SBo}
if [ -z "$ARCH" ]; then
case "$( uname -m )" in
i?86) ARCH=i486 ;;
arm*) ARCH=arm ;;
*) ARCH=$( uname -m ) ;;
esac
fi
CWD=$(pwd)
TMP=${TMP:-/tmp/SBo}
PKG=$TMP/package-$PRGNAM
OUTPUT=${OUTPUT:-/tmp}
if [ "$ARCH" = "i486" ]; then
SLKCFLAGS="-O2 -march=i486 -mtune=i686"
LIBDIRSUFFIX=""
elif [ "$ARCH" = "i686" ]; then
SLKCFLAGS="-O2 -march=i686 -mtune=i686"
LIBDIRSUFFIX=""
elif [ "$ARCH" = "x86_64" ]; then
SLKCFLAGS="-O2 -fPIC"
LIBDIRSUFFIX="64"
else
SLKCFLAGS="-O2"
LIBDIRSUFFIX=""
fi
set -e
# Check if the exim user and group exist. If not, then bail.
if [ "$(id -g exim 2> /dev/null)" != "222" -o "$(id -u exim 2> /dev/null)" !=
"222" ]; then
echo " You must have an 'exim' user and group to run this script."
echo "# groupadd -g 222 exim"
echo "# useradd -d /var/spool/exim -g exim -s /bin/false -u 222 exim"
exit 1
fi
rm -rf $PKG
mkdir -p $TMP $PKG $OUTPUT
cd $TMP
rm -rf $PRGNAM-$VERSION
tar xvf $CWD/$PRGNAM-$VERSION.tar.bz2
cd $PRGNAM-$VERSION
chown -R root:root .
find -L . \
\( -perm 777 -o -perm 775 -o -perm 750 -o -perm 711 -o -perm 555 \
-o -perm 511 \) -exec chmod 755 {} \; -o \
\( -perm 666 -o -perm 664 -o -perm 640 -o -perm 600 -o -perm 444 \
-o -perm 440 -o -perm 400 \) -exec chmod 644 {} \;
# This is our custom config for an all-purpose Exim daemon.
# The SQL- and LDAP-lookups will not be built by default,
# but you can enable them easily by passing the directives
# on the command line when calling this script, e.g.
#
# LOOKUP_LDAP=yes LOOKUP_SQLITE=yes \
# LOOKUP_MYSQL=yes LOOKUP_PGSQL=yes \
# sh exim.SlackBuild
#
cat $CWD/exim.Makefile > Local/Makefile
# Use a stock config for the Exim Monitor (not built by default).
cat exim_monitor/EDITME > Local/eximon.conf
# Use the Exim build system to set the architecture-specific CFLAGS.
# This requires "make" to run twice, in case you wonder. If you want
# to use your own CFLAGS in exim.Makefile, you should put a '#' in
# front of the next 3 lines.
echo "CFLAGS=$SLKCFLAGS -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE" >
Local/Makefile-Linux
FULLECHO="" LIBDIRSUFFIX=$LIBDIRSUFFIX DESTDIR=$PKG make -e || true
FULLECHO="" LIBDIRSUFFIX=$LIBDIRSUFFIX DESTDIR=$PKG make -e makefile
# build & install
FULLECHO="" LIBDIRSUFFIX=$LIBDIRSUFFIX DESTDIR=$PKG make -e
FULLECHO="" LIBDIRSUFFIX=$LIBDIRSUFFIX DESTDIR=$PKG make -e install
find $PKG -print0 | xargs -0 file | grep -e "executable" -e "shared object" |
grep ELF \
| cut -f 1 -d : | xargs strip --strip-unneeded 2> /dev/null || true
# Rename the version specific binary to simply 'exim'.
mv $PKG/usr/sbin/exim-$VERSION-* $PKG/usr/sbin/exim
# Additional sy
[Slackbuilds-users] dead data in tor-browser
Hello there, The tor-browser SlackBuild needlessly extracts the source archive and moves its content to $PKG/opt, resulting in a huge Slackware package containing about 150MB of dead data that is never used and not even accessible by a non-privileged user. There shouldn't be anything else in $PKG/opt besides a "tor-browser" directory containing the source archive. Also, the whole "if $ARCH..." tree setting unused compiler flags etc. should propably be simplified. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] dead data in tor-browser
On Thu, 14 Jan 2016, Willy Sudiarto Raharjo wrote: > when you run tor-browser from your local account, the source will be > copied to your home directory as well and your user will use it > instead of the one in /opt. Have a look at the "tor-browser" script... the private installation in the user's home directory is made by unpacking the archive stored in the /opt/tor-browser directory, e.g. tor-browser-linux64-5.0.6_en-US.tar.xz. The SlackBuild script correctly stores the source archive, but first it unpacks the source archive and moves everything to /opt. This is wrong. The resulting /opt/Browser (and /opt/start-tor-browser.desktop) is 150MB of dead data. These files are not used for anything. In fact they're not even readable by a user. A correct tor-browser.SlackBuild should look like this: https://beingboiled.info/tmp/tor-browser.SlackBuild (needs a copy of mozicon128.png and a bit more refinement) -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Prosody vs. SSL/TLS
On Sat, 19 Sep 2015, Nicolas Kovacs wrote: > This being said, I think it's safe to add luasec to Prosody's > dependencies, since it's needed for encryption. Technically the *.info is correct as luasec is an optional depedency and Prosody will happily build and run without it. However, unless you're running an isolated server Prosody is pretty much useless without encryption as TLS has become mandatory for S2S connections more than a year ago. Unfortunately the SlackBuild hasn't been maintained for longer than that and I wouldn't recommend to use it as it is. Prosody versions after 0.9.4 have improved encryption and fix several bugs, one interoperability issue and one CVE. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] Github file name issue
Hello there, A user of one of my SlackBuilds discovered an interesting problem: If you follow the download link from the info file, pointing to a release on Github, you'll end up with different filenames depending on whether you use a browser or a tool like 'wget', thanks to a 'Content-Disposition' header. E.g.: 'Release' link on Github: https://github.com/BlackIkeEagle/par2cmdline/archive/v0.6.11.tar.gz As expected, wget will save the archive as 'v0.6.11.tar.gz'. However, Firefox will save it as 'par2cmdline-0.6.11.tar.gz', so anybody following the download link from the slackbuilds.org web site will get a non-working build. Most of the Github links I've checked behave this way. What would be a good way to handle this? Editing the SlackBuild script to handle both names doesn't seem very desirable. Do we have to resort to self hosting? -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] nginx permissions?
On Tue, 5 May 2015, Miguel De Anda wrote: i've installed nginx from the slackbuild and i have it run as a proxy for tomcat with a config something like this... when it tries to load some files, it fails with the following error: /var/lib/nginx/proxy/3/00/03 failed (13: Permission denied) it seems to kind of work if i add a user mdeanda users; to the config and change the permission to /var/lib/nginx/* Nginx uses /var/lib/nginx for various temporary files, especially when proxying. The directory should belong to root:root. Upon start, nginx will then create the various subfolders and asign them to the $NGINXUSER unless they already exist. I ran into problems once when I either reconfigured or rebuilt nginx to use a different user id but forgot about /var/lib/nginx/* still belonging to the old user. I don't know how you ended up with the wrong permissions, but if you stop nginx, remove the contents of /var/lib/nginx and restart you should be fine. If the permission problems remain you should check your config and your nginx package for possible errors (e.g. built in wrong environment, manually specified user doesn't exist, etc.). Regarding the user id... the SlackBuild script follows best practices by using the defaults but still giving the user a choice. Having a dedicated nginx user is IMHO desirable with regards to permissions and security. It's not technically required, though. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] stfl-0.24 and Perl examples
Hello there, The SlackBuild for stfl conveniently relocates example.pl to the better fitting doc directory, but as of stfl 0.24 there's also an example2.pl which gets left in vendor_perl :-) Trivial fix: sed -i 's/example\.pl/example*.pl/' stfl.SlackBuild -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] UID/GID for another Dovecot case
On Sun, 15 Feb 2015, Willy Sudiarto Raharjo wrote: So, please consider this a request for either discussion or simply for an assigned uid/gid for a vmail user. please use 303 for vmail 303 is a bad choice for the vmail uid because then you'll need further adjustments in the Dovecot config to work around a security feature to prevent users from ever logging in as daemons (i.e. with uids 500). A vmail user is not required at build time and not necessarily at runtime. When it is, the choice depends on the environment and on what one is trying to achieve. A static entry in the list of recommended UIDs/GIDs doesn't look sensible to me. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] PostgreSQL /usr/share/postgresql
Hello there, I noticed a minor problem with the PostgreSQL SlackBuild. The script installs all libs and binaries into a path containing the version number, e.g. /usr/lib64/postgresql/9.4. This allows a parallel install of two versions which is required for the pg_upgrade utility. As I now noticed the files in /usr/share/postgresql are version dependent, too. Installing 9.4 overwrote the 'Default' time zone set, but 9.3 couldn't parse the enhanced syntax and pg_upgrade failed. I'd suggest adding --datadir=/usr/share/$PRGNAM/$PG_VERSION to the SlackBuild's configure call. This will keep extensions, time zone sets and auxilliary files for text search separate between versions. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] bitcoin incompatible w/ openssl-1.0.1k
Users of the bitcoin SlackBuild, listen up: The recent upgrade to openssl-1.0.1k is incompatible with the Bitcoin system and could lead to consensus forks. You must either withhold the openssl upgrade or rebuild the bitcoin package including a hotfix. For a more detailed explanation and links to patches see http://thread.gmane.org/gmane.comp.bitcoin.devel/6702 and http://thread.gmane.org/gmane.comp.bitcoin.devel/6702/focus=6704 -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] qemu annotations
On Sat, 15 Nov 2014, Antonio Hernández Blas wrote: would it be possible to have an static version of libusb for qemu? See here: https://beingboiled.info/slackware/14.1/build/twm/qemu/ Simply place the libusb 1.0.19 archive next to the other files and the modified SlackBuild will build qemu including a static libusb so you'll get USB passthrough. The libusb addition to the build script is neither very sophisticated nor very well tested, so use it carefully. Also be aware of the different defaults for TARGETS, AUDIODRIVERS and VERSION :-) -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] qemu annotations
Hello there, I've had a closer look at the qemu SlackBuild and now I'm wondering about some of the constructs in there. The checks for libusb, spice and usbredir seem to be completely redundant because qemu's configure will check for availability of these features and enable them when possible. The check for the device tree compiler seems to be even more than redundant, because this isn't even an optional depedency as the dtc has been bundled with qemu since at least version 2.0.0. Most configure options just state the defaults. The OS_CFLAGS parameter is probably meaningless; it's not used anywhere in the makefiles. A propos make... to get a parallel build (e.g. with MAKEFLAGS=-j4) you have to make config-all-devices.mak config-all-disas.mak first, otherwise make will run a single job only which will take quite a while when building all targets (anybody with a better understanding of make care to explain?). Maybe the qemu SlackBuild should get a cleanup (and an update to 2.1.2). Cheers! -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] mupdf docdir download link
Hello there, The SlackBuild for mupdf doesn't set the docdir when calling make. Thus the resulting package has two docdirs, /usr/share/doc/mupdf /usr/doc/ mupdf-1.5, with roughly the same content. Instead of hosting the source archive on a 3rd party site, the download link should point to the official site. Luckily the current version is already available in the archive folder, providing us with a stable link: http://mupdf.com/downloads/archive/mupdf-1.5-source.tar.gz Cheers! -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] REQUIRES=%README% revisited
Hello there, A while ago there was a discussion about the %README% option that didn't really result in a clear conclusion. Yet, I noticed that the option has been removed from my Exim SlackBuild's info. So, what's the official stance on the purpose of %README%? As I see it, Exim fits either way of interpreting what the %README% is supposed to do. The package conflicts with Slackware's Sendmail, so automatic tools should stop here and prompt the user with the README before building and installing. And there's an optional depedency on PostgreSQL. Some clarification would be appreciated. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] A request to pull from my metacpan branch.
On Sun, 20 Jul 2014, Willy Sudiarto Raharjo wrote: if anyone has comment on this branch, please let us know otherwise, this will be part of the next public update I don't get this. The other thread mentioned a total of six SlackBuilds having non-working download links. Yet this branch will touch and change around 650, completely ignoring the respective maintainers who might have a reason for preferring search.cpan.org to metacpan.org and sticking to the download links either of them provide. Besides... a broken link often is an indicator for a SlackBuild in neglect. It's better when these get reported on the mailing list than having a nicely working download link for a version that's missing important bug or security fixes. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] gtk-engines - wrong md5sum
Hello y'all! There seems to be a mix-up in the .info file of the recently updated desktop/gtk-engines regarding the md5sum. From the diff: -MD5SUM=e0f7ac7eaad3896762804312f03c6f46 +MD5SUM=fe8e0540932b4de8b4b320bfca178c84 My existing gtk-engines-2.20.2.tar.gz archive has the e0f7... md5sum, and so does a recent download from ftp.gnome.org. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] qemu vnc
On Tue, 29 Apr 2014, King Beowulf wrote: On a side note: what's the reason for making VNC support optional? 1. Not everyone wants or needs or uses vnc 2. To use vnc you'll need to set up a bunch of stuff for remote access I don't use vnc...my qemu VMs are just plain clean build environments I don't need vnc to view the VM on my host's monitor. If the consensus is that it should be enable by default, let me know. I'd vote for leave it enabled (and don't even make it an option) for the following reasons: - It's the default. - It doesn't introduce any additional dependencies or links to heavyweight libs. - No extra step needed if you need vnc. - No extra step needed if you don't need vnc, either. If you don't want it, just don't make it your -display when you start the VM. IMHO vnc is quite handy even for local use. I can start a bunch of machines without having a bunch of SDL windows clutter up my desktop, yet unlike -display none I can still access the virtual console when necessary simply by starting the vnc client. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] nginx temp paths
Hello there, I'm using my own build script for nginx, but I just noticed a behaviour that probably affects the SBo SlackBuild as well. If you don't explicitly define the various temp paths when calling configure, nginx will create them in /usr during startup. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] qemu vnc
Hello there, The README for the recently updated qemu says to use the ENABLE_VNC variable for optional vnc support. In the SlackBuild, however, this variable is called VNC_ENABLE. On a side note: what's the reason for making VNC support optional? -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Fw: perl-Locale-Maketext-Lexicon
On Mon, 24 Mar 2014, Willy Sudiarto Raharjo wrote: I've mailed the maintainer of perl-Locale-Maketext-Lexicon (see the forwarded mail at the bottom) and haven't received a response by now, so I suppose either the email address is dead or he isn't interested anymore. Fixed now It's nice having a working download link, but this fix just hides the fact that this SlackBuild isn't maintained any more - or is this also marked as invalidate if a new maintainer doesn't step up somewhere? -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] kyotocabinet CFLAGS
Hello there, I've noticed that kyotocabinet's configure adds -march=native to the CFLAGS in the Makefile. According to the gcc man page, when built on very modern CPU (e.g. i7 Haswell), the resulting package might not work on a less capable CPU (e.g. Pentium-D). I've unknowingly been running a native kyotocabinet for some time without encountering any problems, yet I've decided to rebuild the package with a minor modification to get a generic library (see attached patch). -- --- kyotocabinet.SlackBuild.orig2013-11-26 10:57:41.0 +0100 +++ kyotocabinet.SlackBuild 2014-03-11 01:38:51.308089868 +0100 @@ -80,6 +80,11 @@ --docdir=/usr/doc/$PRGNAM-$VERSION \ --build=$ARCH-slackware-linux +# Don't use 'native' or the resulting package might +# not work on low-spec CPUs of the same architecture. +sed -i -r 's/^CFLAGS = -march=native/CFLAGS =/' Makefile +sed -i -r 's/^CXXFLAGS = -march=native/CXXFLAGS =/' Makefile + make make install DESTDIR=$PKG ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] duplicity tarsnap
Hello there, In search of a backup tool I stumbled across the Slackbuilds for duplicity and tarsnap which have been abandoned by their respective maintainers. The Slackbuild for duplicity hasn't been updated for more than three years, although there have been a few new releases since with lots of improvements and important bug fixes. The Slackbuild for tarsnap hasn't been updated for more than four years and refers to a version with a critical security issue that could expose supposedly encrypted data. I've settled on rsnapshot for different reasons, but if you're a user of either duplicity or tarsnap, please consider stepping in as the new maintainer. On a side note... discovering neglected Slackbuilds has become sort of a pattern for me whenever I look beyond those which I've been using for years. Do I just happen to always pick up that one foul apple or is this a problem on a larger scale? -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Email address rejected
On Tue, 24 Dec 2013, Rob McGee wrote: The function is very simple, as Erik has explained before. When a build is submitted, the PHP code looks up the MX host for the given address. It makes an SMTP connection thereto, and does a simple test: MAIL FROM:slackbuilds-users@slackbuilds.org RCPT TO:your@email.address If it gets a 250 OK [...] response all is well and your build is accepted. If not, it assumes your email address is bad. No attempt to send actual mail is sent; in either case the PHP code will QUIT next. Without knowing the exact implementation, wouldn't it be better to assume a bad e-mail address only when receiving a 5xx permanent error? -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Want to help with work toward 14.1 repo?
On Fri, 8 Nov 2013, Robby Workman wrote: If there are things on that checklist which you can confirm will build properly on both 32bit and 64bit Slackware 14.1, then we'd like to hear from you. Reply to this message with a list of them, and we'll remove them from the checklist. Shorten the list for I've successfully built the following packages on both 64- and 32-bit 14.1: haskell/* (- the complete tree!) development/darcs development/happy desktop/xmonad Downloading and building was done in an automatic fashion. The resulting darcs was able to read, upgrade and work on my (quite old) darcs repositories. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Want to help with work toward 14.1 repo?
On Sun, 10 Nov 2013, Matteo Bernardini wrote: I had a look at this: it seems to me that the new SlackBuild make some changes that I haven't understand completely: - no more manpages The previous man pages were outdated and the current man pages contain Debian-specific changes that would have to be edited out, so I decided to abandon the man pages for now. Full documentation is still included with the spec.txt in the /usr/doc/exim-4.82 directory. - ipv6 enabled by default Though not enabled by default in the previous builds, I've actually been running an IPv6-enabled Exim on all my systems for several months now. I haven't encountered any problems and IPv6 really should be embraced more, so this is now a default. if this is ok for you just whistle and I'll push it in my branch. I can confirm that all changes are intentional :-) -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] tor
I've just the noticed the SlackBuild for tor in the 14.1 master branch... The script is for version 0.2.2.37 which has several known bugs that can be triggered remotely. Also, 0.2.3 has superseeded 0.2.2 as the stable branch about a year ago. Considering the nature of Tor, this looks dangerously outdated and should not be part of the Slackbuilds for 14.1. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Want to help with work toward 14.1 repo?
On Fri, 8 Nov 2013, Robby Workman wrote: If there are things on that checklist which you can confirm will build properly on both 32bit and 64bit Slackware 14.1, then we'd like to hear from you. Reply to this message with a list of them, and we'll remove them from the checklist. Actually, most of the SlackBuilds I feel responsible for have already been taken care of (thanks!). That leaves me with: misc/par2cmdline perl/perl-Template-Toolkit network/proxymini These build fine on 14.1 (tested on i486 and x86_64). Some editing is needed for network/exim: there's been a new Exim release recently, thus the download link for 4.80.1 is no longer valid. You could just change the link to point to the archived 4.80.1 release, but I'd prefer to replace the SlackBuild with an updated one for 4.82. Archived Exim 4.80.1 download link: ftp://ftp.exim.org/pub/exim/exim4/old/exim-4.80.1.tar.bz2 Updated SlackBuild for Exim 4.82: http://beingboiled.info/exim.tar.gz Both versions build fine on 14.1 (i486 x86_64). -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Want to help with work toward 14.1 repo?
On Fri, 8 Nov 2013, Robby Workman wrote: If there are things on that checklist which you can confirm will build properly on both 32bit and 64bit Slackware 14.1, then we'd like to hear from you. Reply to this message with a list of them, and we'll remove them from the checklist. Some more... (they're not mine, I just use them) These build fine on 14.1 (i486 x86_64): desktop/gtk-engines desktop/gtk-nodoka-engine development/bvi development/cryptoki games/hatari_tos_roms libraries/equinox libraries/libotr libraries/libpar2 misc/bsdsfv misc/cksfv misc/keychain homepage has moved to http://www.funtoo.org/index.php?title=Keychain network/bitcoin network/claws-mail network/pidgin-otr office/epdfview homepage seems to be gone system/dosbox system/p7zip These build fine, too. They refer to rather old versions, though. Once submissions are open again, an update from the maintainers would be nice: games/hatari games/mednafen libraries/libetpan Minor issue here: system/kyotocabinet This builds fine on 64-bit systems. On 32-bit systems, however, assuming ARCH=i486 doesn't work. At least i686 is required and the compiler flags have to be set accordingly (i.e. -march=i686). -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] PostgreSQL - multiple installed versions
On Sun, 29 Sep 2013, Audrius Kažukauskas wrote: I have successfully upgraded from 9.2.4 to 9.3.0 and then encountered the problem: while psycopg2 (Python DB adapter for Postgres) builds fine, trying to use it results in the following error: ImportError: libpq.so.5: cannot open shared object file: No such file or directory The setup.py for psycopg2 asks pg_config for config options, but ignores the required LDFLAGS. As a workaround, you could specify the flags while building the package, e.g.: env LDFLAGS=$(pg_config --ldflags) sh psycopg2.SlackBuild It's easy to adapt the SlackBuild accordingly, but maybe it would be better to address this in psycopg2 itself. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] PostgreSQL - multiple installed versions
On Wed, 18 Sep 2013, Adis Nezirovic wrote: The problem with symlinks is we must not use absolute paths (installpkg can be called with --root parameter). Anyway, I've simplified logic for symlinks in doinst.sh, should be more robust now. I've also removed PRGNAM from all scripts except SlackBuild (it was a leftover from an earlier experiment with versioned package names). The new version works almost perfectly. One minor nitpick though: The previous version explicitly set permissions for the DATADIR: # permissions for DATADIR should be u=rwx (0700) chmod 700 $PKG/var/lib/pgsql/$PG_VERSION/data The new version doesn't. That's fine, because the mandatory initdb will set the correct permissions. However, if you upgradepkg --reinstall postgres the DATADIR's permissions will get reset to 0755 and the server won't start any more. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] PostgreSQL - multiple installed versions
On Wed, 11 Sep 2013, Adis Nezirovic wrote:
3) You have PostgreSQL installed (let's say 9.x series, or at least 8.3)
You want to do in-place testing (two running PostgreSQL instances) and
upgrade using pg_upgrade
I've got a small 15 GB Postgres 9.2 DB on a modest Atom server. The
traditional dump/restore cycle takes around 3 hours, so the new approach
to an easier pg_upgrade is very much welcome.
3.3) Install new PostgreSQL 9.3, while keeping the old package installed
as well. Please note that new PostgreSQL will have binaries in
/usr/lib64/postgresql/9.3/bin, and install script create symlinks in
/usr/bin only if respective postgresql binaries are not found in
/usr/bin (i.e. it will _not_ overwrite your PostgreSQL 9.x binaries!)
Minor problem: doinst.sh generates broken symlinks to usr/lib... instead
of /usr/lib... - was easily fixed.
3.4) Make sure new rc.postgresql script is active
Minor problem: Could not find 'postgres' binary. because the PRGNAM
variable is missing in rc.postgresql. Again, easily fixed. Hmm... does
this actually need to be a variable?
Any feedback on all of this and testing the SlackBuild would be warmly
welcomed.
Besides those two minor issues (see attached diff for my changes) your
test release of the SlackBuild has been working fine for me. Adding the
--link option to pg_upgrade allowed upgrading from 9.2 to 9.3 in about a
minute with no need for a dump/restore-cycle.
-- diff -u -r postgresql-an/doinst.sh postgresql-tm/doinst.sh
--- postgresql-an/doinst.sh 2013-09-11 14:39:41.0 +0200
+++ postgresql-tm/doinst.sh 2013-09-17 20:44:40.822303331 +0200
@@ -30,10 +30,10 @@
if [ -e usr/bin/$pg_prog ]; then
# make sure we overwrite only symlinks and not actual binaries
if [ -L usr/bin/$pg_prog ]; then
- ln -sf $pg_binary usr/bin/$pg_prog
+ ln -sf /$pg_binary usr/bin/$pg_prog
fi
else
-ln -s $pg_binary usr/bin/$pg_prog
+ln -s /$pg_binary usr/bin/$pg_prog
fi
done
diff -u -r postgresql-an/rc.postgresql.new postgresql-tm/rc.postgresql.new
--- postgresql-an/rc.postgresql.new 2013-09-11 14:39:41.0 +0200
+++ postgresql-tm/rc.postgresql.new 2013-09-17 22:14:27.446243178 +0200
@@ -23,6 +23,7 @@
PG_VERSION=${PG_VERSION:-@PG_VERSION@}
PG_PORT=${PG_PORT:-@PG_PORT@}
+PRGNAM=${PRGNAM:-@PRGNAM@}
LIBDIRSUFFIX=@LIBDIRSUFFIX@
if [ x$PG_VERSION != x9.2 ];then
___
SlackBuilds-users mailing list
SlackBuilds-users@slackbuilds.org
http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users
Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/
FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] lighttpd nginx
On Tue, 25 Jun 2013, Martin Matějek wrote: Lighttpd user here. I have manually rebuild packages on every servers from slackbuild, so I'd like to see upstream slackbuild updated. If maintainer won't respond, I'm willing to maintain lighty because I use it everywhere I can. Since Paul still hasn't replied yet I'd say: please go ahead, submit an updated SlackBuild, and take over future maintenance. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] lighttpd nginx
On Mon, 29 Apr 2013, Thomas Morper wrote: In November 2012 lighttpd 1.4.32 was released, containing a fix for a critical bug. The SlackBuild, however, is still at 1.4.31. It's been two month and there has been neither an update nor a reply from Paul (I tried contacting him off-list, too). I don't want to take over maintenance as I'm in the process of migrating everything to nginx (which fits my needs better), but lighttpd is a nice webserver (easy to set up and well-documented) and shouldn't go unmaintained. Any lighty-users willing to step up? -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] lighttpd nginx
Hello *, I've noticed that the SlackBuilds for lighttpd and nginx lag a bit behind. In November 2012 lighttpd 1.4.32 was released, containing a fix for a critical bug. The SlackBuild, however, is still at 1.4.31. The SlackBuild for nginx is still at 1.2.26 (December 2012), having missed numerous bugfixes from two more stable releases in the 1.2-branch (superseeded by the new stable 1.4 a few days ago). I've tried contacting the respective maintainers but haven't received a reply yet (considering the eagerness of Google's spam filter I'm not sure if they even got my mails). -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] hitting a problem with a SlackBuild
On Sun, 30 Dec 2012, Glenn Becker wrote: Is it acceptable, in this case, to create a diff for this makefile that simply skips over the interactive bit and auto-supplies the y? I am not sure how else to accomplish this. No need for a patch, just set PERL_MM_USE_DEFAULT to a true value, then MakeMaker will assume defaults without waiting for user input, e.g.: PERL_MM_USE_DEFAULT=y perl Makefile.PL Cheers, Thomas -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] Exim 4.80 security issue
Hello all, There's a critical security issue in Exim from versions 4.70 to 4.80: http://article.gmane.org/gmane.mail.exim.announce/147 The exim.org-site seems to be unreachable right now, but you can get the bugfixed 4.80.1 from one of the mirrors, e.g. ftp://ftp.easynet.be/exim/exim/exim4/exim-4.80.1.tar.bz2 ...and use the existing SlackBuild for an upgrade: root@darkstar:~/exim# VERSION=4.80.1 BUILD=1 sh exim.Slackbuild Cheers! -- tho...@beingboiled.info ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] [PATCH]
Here's a patch for games/openttd... (did I get it right this time?)
There seems to be a bug in 14.0's rename utility that causes file name
substitutions with leading dashes to be interpreted as non-existing
parameters. Using underscores instead, openttd builds and works fine.
Disclaimer: I'm not the maintainer, but I really like this game and
wouldn't want to see it missing in 14.0 :-)
Cheers
-- From cf6cc8e9d00548201ac65f914ddf206ee4ab1261 Mon Sep 17 00:00:00 2001
From: Thomas Morper tho...@beingboiled.info
Date: Sat, 15 Sep 2012 18:11:33 +0200
Subject: [PATCH] games/openttd: changed filenames to avoid bug in rename
---
games/openttd/openttd.SlackBuild |8
1 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/games/openttd/openttd.SlackBuild b/games/openttd/openttd.SlackBuild
index 6fffcd1..5b5d284 100644
--- a/games/openttd/openttd.SlackBuild
+++ b/games/openttd/openttd.SlackBuild
@@ -24,7 +24,7 @@
PRGNAM=openttd
VERSION=${VERSION:-1.2.0}
-BUILD=${BUILD:-1}
+BUILD=${BUILD:-2}
TAG=${TAG:-_SBo}
OPENGFX=0.4.4
@@ -103,9 +103,9 @@ make install DEST_DIR=$PKG
# Copy opengfx, opensfx, and openmsx files if OPENDATA=YES
if [ $OPENDATA == YES ]; then
# First rename some text files to avoid name collisions when copied
- rename .txt -opengfx.txt opengfx-$OPENGFX/*
- rename .txt -opensfx.txt opensfx-$OPENSFX/*
- rename .txt -openmsx.txt openmsx-$OPENMSX/*
+ rename .txt _opengfx.txt opengfx-$OPENGFX/*
+ rename .txt _opensfx.txt opensfx-$OPENSFX/*
+ rename .txt _openmsx.txt openmsx-$OPENMSX/*
# Now copy the data files
mkdir $PKG/usr/share/games/openttd/data/
mkdir $PKG/usr/share/games/openttd/gm/
--
1.7.4.4
___
SlackBuilds-users mailing list
SlackBuilds-users@slackbuilds.org
http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users
Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/
FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] [PATCH] network/exim: fixed README, link with liblber
Fixed README to contain the correct argument for enabling the PostgreSQL lookup; fixed exim.Makefile to link with liblber when using the LDAP lookup. Tested on 14.0RC4 32- 64-bit; seems to build and work fine. Sorry, git-n00b here. Was this the correct way to send a patch? --- network/exim/README|2 +- network/exim/exim.Makefile |2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/network/exim/README b/network/exim/README index 42cf5ce..58c1319 100644 --- a/network/exim/README +++ b/network/exim/README @@ -32,5 +32,5 @@ but you can enable them easily by passing the directives on the command line when calling this script, e.g. LOOKUP_LDAP=yes LOOKUP_SQLITE=yes \ - LOOKUP_MYSQL=yes LOOKUP_POSTGRES=yes \ + LOOKUP_MYSQL=yes LOOKUP_PGSQL=yes \ sh exim.SlackBuild diff --git a/network/exim/exim.Makefile b/network/exim/exim.Makefile index 97c90e5..132ee5e 100644 --- a/network/exim/exim.Makefile +++ b/network/exim/exim.Makefile @@ -367,7 +367,7 @@ PCRE_CONFIG=yes # ...or just enable your favourite lookups and let GNUmake handle the rest ifeq ($(LOOKUP_LDAP),yes) -LOOKUP_LIBS+=-lldap +LOOKUP_LIBS+=-lldap -llber endif ifeq ($(LOOKUP_MYSQL),yes) -- 1.7.4.4 ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] perl-xml-sax doinst.sh
Hello there,
While playing with some Perl modules I noticed that the doinst.sh for
perl-xml-sax won't work as expected because of a version- and platform-
dependent path that is hardcoded for Slackware 13.0 32-bit.
Fortunately, Perl itself knows the correct path, so the problem is easy
to fix with a doinst.sh like this:
---8---
chroot . /usr/bin/perl -MConfig -MXML::SAX -e '
XML::SAX-add_parser(q(XML::SAX::PurePerl))-save_parsers
unless (-f $Config{installvendorlib} . /XML/SAX/ParserDetails.ini);'
---8---
Cheers!
--
http://beingboiled.info/slackware/
___
SlackBuilds-users mailing list
SlackBuilds-users@slackbuilds.org
http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users
Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/
FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] nvidia-kernel vs. xorg.conf
On Fri, Jul 29, 2011 at 09:26:51AM +0200, emmel wrote: Remember that /etc/X11/xorg.conf.d overrides /usr/share/X11/xorg.conf.d. Duh, stupid me, that's how I configured my keyboard layout! I completely forgot about /usr/share and the usual copy to /etc, then modify Now that I remember: Forget what I wrote. Keep the config, but make it /usr/share/X11/xorg.conf.d/10-nvidia.conf instead of /etc/X11/xorg.conf.d/driver.conf - should work out of the box then, but is easy to customize and updates won't break user-made changes. Right? -- http://beingboiled.info/slackware/ ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] mupdf manpages
Hello there, I just built the mupdf PDF reader and noticed that the manpages end up in the wrong place (/usr/man instead of /usr/man/man1) where man won't find them. It's trivial to fix, maintainer is Cc:'ed, just thought potential users of this nice little PDF reader should know. -- http://beingboiled.info/slackware/ ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] nvidia-kernel vs. xorg.conf
Hello there, Recently, a new bit of code, that will put an xorg.conf.d/device.conf to the resulting package, has been added to the nvidia-kernel SlackBuild. While this is convenient for most first time installs, it's less convenient when updating, because the hardcoded device.conf will overwrite any user-made changes and will clash with certain multi-head setups. I think I'd rather have the device.conf in the doc-directory and a hint in the README. Still convenient for first-time installs, but won't break existing setups. What do you think? -- http://beingboiled.info/slackware/ ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] 13.37 SlackBuild glitches
Hello fellow Slackbuilders, While preparing packages for my upgrade to 13.37, I noticed minor glitches with some of the Slackbuilds, namely: - hatari: is this still being maintained? The script is for 1.1.0 from 2008, current version is 1.4.0 from 2010. - lighttpd: is this still being maintained? The script is for 1.4.26, current version is 1.4.28 from 2010. - virtualbox: needs jdk from /extra to build. - vobcopy: the README still mentions a dependency on libdvdread from SBo, but this is a part of Slackware now. - wireshark: download link is broken. As I said... details, mostly. But maintainers and admins propably want to have a look at it. -- http://beingboiled.info/slackware/ ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] Exim 4.74 download link
Exim 4.75 has just been released which means that the download link to 4.74 is broken now. Can one of the admins please edit the exim.info file and make it point to the source's new location at: ftp://ftp.exim.org/pub/exim/exim4/old/exim-4.74.tar.bz2 Exim 4.74 is safe to use and will happily run on slackware-current. I'll submit an update for Exim 4.75 once Slackware 13.37 is ready. -- http://beingboiled.info/slackware/ ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] request for removal: exim
Hello there, Can one of the admins please remove exim from the ready-queue? Exim 4.74 was released just a few days ago and I'd rather have that than 4.73 and a broken download link in the next update. Please do not remove exim-html, the version in ready is still fine. Cheers! -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Exim Sbo
On Mon, Jan 17, 2011 at 11:06:59AM +0100, TuxaneMedia wrote: I just tried to build network/exim but the download failed. The Sbo referes to ftp://ftp.exim.org/pub/exim/exim4/exim-4.72.tar.bz2 but exim has been updated to v4.73. An update for 4.73 is in the works and will be ready soon. Till then you can safely build and run 4.72 using the download link you provided in your follow-up. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Exim Sbo
On Mon, Jan 17, 2011 at 03:23:16PM +0200, V'yacheslav Stetskevych wrote: No, do not run 4.72, it has a ROOT HOLE in it. No, it doesn't. The recently discovered security hole was based on a flaw that has been fixed since 4.70 (released in 2008) and only affected distributions still shipping older versions. Exim 4.73 contains additional security measures. An update is desirable, but not required immediately. At least for now, Exim 4.72 on Slackware is perfectly safe. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Updates - 20101122.1
On Mon, 22 Nov 2010 21:23:13 +0100 Heinz Wiesinger ppr...@slackbuilds.org wrote: network/broadcom-sta: Updated for version 5.60.246.2. This driver version has already been superseded by 5.60.246.6 and unfortunately the previous versions are no longer available for download. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Updates - 20100726.1
On Tue, 10 Aug 2010 23:55:00 +0300 Mikko Varri v...@linuxbox.fi wrote: libraries/perl-HTTP-Proxy: Added (a pure Perl HTTP proxy) The .info file seems to have wrong version number (0.03, should be 0.24). Can one of the admins please fix this or should I resubmit a fixed package? -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] Perl-modules and CFLAGS
Dear Slackbuilders, I'm in the process of preparing some Perl-modules for SBo but I'm unsure how to handle CFLAGS the right way when C is involved. A MakeMaker-based Makefile.PL allows you to pass the CCFLAGS and OPTIMIZE variables to the compiler, however... 1.) 'perl Makefile.PL CCFLAGS=$SLKCFLAGS' isn't a very good idea, since it will a overwrite the CCFLAGS defaults and something like a missing -D_LARGEFILE_SOURCE will haunt you later on. 2.) 'perl Makefile.PL OPTIMIZE=$SLKCFLAGS' isn't perfect, either. So far I've got three modules and in each one OPTIMIZE behaves differently: - module1 only has -O2 as a default but will accept OPTIMIZE - module2 uses what Perl was compiled with as a default, e.g. -O2 -march=i486 -mtune=i686 and thus doesn't need OPTIMIZE, but will use it if it's set - module3 is similar to module2, however for some of its C-sources it will use the Perl defaults instead of whatever you set OPTIMIZE to My conclusion: there are 18000 modules on cpan.org - don't waste your time trying to find the über-perfect settings for each one, just go with the defaults unless there's a good reason for manual intervention (e.g. doesn't work otherwise). In SBo-terms: set ARCH, SLKCFLAGS and LIBDIRSUFFIX as we always do, but don't care about it; the result will be fine for most users. Those who want more from their Perl modules still only have to add a single line to the module's SlackBuild script. Interestingly, this is what the current template does :-) So, is this the way to go or am I missing something and the CFLAGS actually do need more attention? -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] small fix for perl-dbd-sqlite
Am Sonntag, 16. Mai 2010 04:27:20 schrieben Sie: There's a small bug in my SlackBuild for perl-dbd-sqlite. On anything else than Slackware 13.0 with Perl 5.10.0 the hardcoded Perl version number prevents the $PKG dir from being cleaned up prior to makepkg. Could one of the admins please fix this? Thanks! Done in local branch ; will be in next public update. Whoops... looks like this got un-fixed again when the patch for bash4- compliance was added two days later. I like gitweb :-) Here's a proper version: - bash4 compliant - version independent cleanup of $PKG - removes the unwanted files only once (stupid me!) Can one of the admins enter this into the repository, please? Thanks! -- perl-dbd-sqlite.SlackBuild Description: application/shellscript ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] small fix for perl-dbd-sqlite
Hello *, There's a small bug in my SlackBuild for perl-dbd-sqlite. On anything else than Slackware 13.0 with Perl 5.10.0 the hardcoded Perl version number prevents the $PKG dir from being cleaned up prior to makepkg. The problem is in line 76: rm -rf $PKG/usr/lib$LIBDIRSUFFIX/perl5/5.10.0 Quick and dirty solution for the upcoming Slackware 13.1: rm -rf $PKG/usr/lib$LIBDIRSUFFIX/perl5/5.10.1 A more generic approach: find $PKG \( -name perllocal.pod -o -name .packlist \ -o -name '*.bs' -empty \) -delete find $PKG -depth -type d -empty -delete Could one of the admins please fix this? Thanks! -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] pdksh weirdness, very strange bug
Am Mittwoch 24 Februar 2010 19:14:47 schrieb LEVAI Daniel: If a file's size is exactly 2GiB or more, pdksh fails to complete its name during filename completion (eg.: hitting TAB or CTRL+e or whatever). If the file's size is just one byte short of the 2 gigs, then the completion works fine. I can't really test it right now, but my guess is that the stat()-call in edit.c line 595 fails and that you should add something like - D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE to the compiler flags when you're compiling for i486. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] critical vulnerability in flash-player-plugin
A critical vulnerability has been identified in Adobe Flash Player version 10.0.42.34 and earlier. This vulnerability (CVE-2010-0186) could subvert the domain sandbox and make unauthorized cross-domain requests. http://www.adobe.com/support/security/bulletins/apsb10-06.html Heise Online has some more details: http://tinyurl.com/ygbvmxk It'd be nice to have 10.0.45.2 in the next SBo update. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Updates - 20091123
Am Montag, 23. November 2009 18:33:56 schrieb Robby Workman: multimedia/GoogleEarth: Updated for version 5.1.3533.1731. The new SlackBuild will always build a .tgz-Package regardless of the PKGTYPE setting. Cheers, Thomas -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] ulogd database driver selection doesn't work
The ulogd Slackbuild allows to disable the MySQL- and SQLite-plugins during build-time, but actually this doesn't work because the corresponding configure-parameters don't work as expected. SQLite is always included, and trying to disable MySQL will cause the SlackBuild to fail. The simple fix: don't use --without-mysql or --without-sqlite3; just set do_mysql and do_sqlite to an empty string instead. Cheers, Thomas -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] QtCurve source lost and found
Hello, The download-links for both QtCurve-Gtk2 and QtCurve-KDE4 are gone. It seems like this is another project where they don't keep an archive of previous releases. Luckily both files still can be found in the FreeBSD distfiles with md5sums matching those in the respective SBo *.info files: ftp://ftp.de.freebsd.org/pub/FreeBSD/ports/distfiles/QtCurve-Gtk2-0.65.3.tar.bz2 ftp://ftp.de.freebsd.org/pub/FreeBSD/ports/distfiles/QtCurve-KDE4-0.65.4.tar.bz2 I don't know... would it be wise for future releases to contain download-links pointing to e.g. FreeBSD or Gentoo distfiles instead of the official home.freeuk.com link or would that just introduce other problems? -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Updates - 20090510
Again, some minor permission problems have sneaked into this update for those who use rsync to keep up to date: rsync: opendir 12.2/audio/mp3gain (in slackbuilds) failed: Permission denied (13) rsync: send_files failed to open 12.2/games/gnubg/gnubg.info (in slackbuilds): Permission denied (13) Thanks in advance for fixing :-) -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] Flash Player security update
Just in case this hasn't been addressed yet and is already in the queue waiting for a push... Adobe has fixed a vulnerability in the Flash Player: http://www.adobe.com/support/security/bulletins/apsb09-01.html The download-link from the .info-File now points to version 10.0r22, md5sum is 23e4c2b844db0f87ff62084178aa2b1f. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
[Slackbuilds-users] cpan2tgz 0.6.2 no good on Slackware 12.2
Hello there, The SlackBuild in the 12.2 repository for cpan2tgz is still for version 0.6.2 which itself builds fine, but the resulting program actually isn't able to build packages due to changes in the Slackware-supplied CPAN.pm. This problem is fixed from 0.6.3 onwards, but I wonder if it's really worth updating the SBo script since cpan2tgz is a tool exclusively for Slackware and the author himself offers ready-to-be-installed noarch packages on his homepage. What do you think? -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] fontforge.SlackBuild writes outside of $PKG
Am Sunday 18 January 2009 09:58:53 schrieb Robby Workman: Anyway, I'll look into this and try to get a fix in asap; at a glance, it looks like the --mandir usage plus DESTDIR target at make install should handle it. I tried it that way (leaving configure as it is and using the common make install DESTDIR=$PKG without any other tweaks) and the result looks good to me. I checked with make -n install ... before and didn't notice any files being written where they weren't supposed to be. Thanks for the heads-up on this, Thomas. You're welcome. -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] flash-player-plugin md5sum wrong in info file
Am Friday 19 December 2008 02:45:10 schrieb João Felipe Santos: Just a quick note: i think the package for installing Adobe Flash Player was changed upstream. Checksum is failing here but after changing it to the new value the installation works fine with new package. The upstream package definetely has been changed to fix a critical vulnerabiliy. An update is highly recommended. http://www.heise-online.co.uk/news/Critical-hole-in-Linux-Flash-Player--/112282 -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
Re: [Slackbuilds-users] Updates - 20081210
Am Monday 15 December 2008 21:11:33 schrieb Robby Workman: I'd be willing to accept a build based on the snapshots, but since I no longer have a need for madwifi, I'm effectively abandoning maintenance of those scripts. If you or someone else wants to take what's in our 12.1 repo and make the changes needed for a snapshot *and* take over maintenance of them, feel free. Just as a hint for the future maintainer: There are no changes necessary. Well, at least no major ones. I used the existing scripts to build packages from the 20081204 snapshot in the hal-0.10.5.6 branch. The driver worked out of the box, and the tools just needed two additional sed calls to adjust a path. sed -i 's|/usr/local|/usr|g' tools/ath_info/Makefile sed -i 's|/share/man|/man|g' tools/ath_info/Makefile Unfortunately, the kernel-based ath5k is not without flaws, making madwifi the better choice even for systems supported by ath5k (e.g. Asus EEE 701: incredibly slow with ath5k, but works fine with madwifi) -- ___ SlackBuilds-users mailing list SlackBuilds-users@slackbuilds.org http://lists.slackbuilds.org/mailman/listinfo/slackbuilds-users Archives - http://lists.slackbuilds.org/pipermail/slackbuilds-users/ FAQ - http://slackbuilds.org/faq/
