[gentoo-dev] eselect_zenity: alpha eselect GUI
Hey all, I've been wanting a GUI for eselect lately, so tonight I hacked up the start of one called eselect_zenity [1]. It only works for the most trivial modules so far -- it has to parse eselect output, so special parsers need to be written for each type. eselect_zenity uses (surprise!) Zenity, a shell-scripting interface to GTK+. I'd appreciate any patches you'd like to contribute to add functionality or fix bugs. Thanks, Donnie 1. http://dev.gentoo.org/~dberkholz/eselect_zenity/ -- [EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-firewall/ipset: ChangeLog metadata.xml ipset-2.3.0.20070828.ebuild
On 12:06 Wed 07 Nov , Peter Volkov wrote:
> On Mon, 05/11/2007 в 10:03 -0800, Donnie Berkholz wrote:
>
> BTW, is it possible to force portage to fetch sources if they do not
> exist in ${DISTDIR}? Also does there exist a better way to find
> iptables sources than find version without revision:
>
> IPTVERINS=`echo $(best_version net-firewall/iptables) | \
> sed -n
> 's:^[^/]*/[[:alpha:]]*-\([0-9]\+\([.][0-9]\+\)*[a-z]\?\(_\(pre\|p\|beta\|alpha\|rc\)[0-9]*\)*\)\(-r[0-9]\+\)\?$:\1:p'
>
> and construct package name (iptables-${IPTVERINS}}.tar.bz2)?
Not that I know of for fetching, but for the version, that sed seems
awfully complex. I'd probably use bash substitution like this instead:
IPTVERINS=$(best_version net-firewall/iptables)
# Strip revision (safe, since nothing else has a hyphen followed by 'r')
IPTVERINS=${IPTVERINS%-r*}
# Strip category
IPTVERINS=${IPTVERINS#*/}
I previously brought up the idea of a way to access portage's
CATEGORY/PN/PV parser from within ebuilds, but there didn't seem to be a
whole lot of interest.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-apps/paludis: ChangeLog paludis-0.26.0_alpha3.ebuild
On 14:58 Mon 05 Nov , Piotr Jaroszynski (peper) wrote:
> 1.1 sys-apps/paludis/paludis-0.26.0_alpha3.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-apps/paludis/paludis-0.26.0_alpha3.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-apps/paludis/paludis-0.26.0_alpha3.ebuild?rev=1.1&content-type=text/plain
> create-paludis-user() {
> enewgroup "paludisbuild"
> enewuser "paludisbuild" -1 -1 "/var/tmp/paludis" "paludisbuild"
> }
>
> pkg_setup() {
> replace-flags -Os -O2
> create-paludis-user
> }
> pkg_preinst() {
> create-paludis-user
> }
Shouldn't need to create a user twice, and that quoting makes this a bit
harder to read than it needs to be.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-firewall/ipset: ChangeLog metadata.xml ipset-2.3.0.20070828.ebuild
On 14:23 Mon 05 Nov , Peter Volkov (pva) wrote:
> 1.1 net-firewall/ipset/ipset-2.3.0.20070828.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-firewall/ipset/ipset-2.3.0.20070828.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-firewall/ipset/ipset-2.3.0.20070828.ebuild?rev=1.1&content-type=text/plain
> src_defs() {
> # these are used in both of src_compile and src_install
> myconf="${myconf} PREFIX="
> myconf="${myconf} LIBDIR=/lib"
Should this be get_libdir()?
Also seems like instead of having a function to do this, you could just
define the variables once in pkg_setup() or src_compile() as globals.
> pkg_config() {
> cd "${ROOT}${PORTAGE_TMPDIR}"
> mkdir patch-o-matic-ng
> cd patch-o-matic-ng
>
> IPTVERINS=`echo $(best_version net-firewall/iptables) | \
> sed -n
> 's:^[^/]*/[[:alpha:]]*-\([0-9]\+\([.][0-9]\+\)*[a-z]\?\(_\(pre\|p\|beta\|alpha\|rc\)[0-9]*\)*\)\(-r[0-9]\+\)\?$:\1:p'`
>
> einfo "Unpacking patch-o-matic-ng-${POM_PV}"
> tar -jxf "${DISTDIR}"/patch-o-matic-ng-${POM_PV}.tar.bz2 || \
> die "Unable to unpack
> patch-o-matic-ng-${POM_PV}"
> einfo "Unpacking iptables-${IPTVERINS}"
> tar -jxf "${DISTDIR}"/iptables-${IPTVERINS}.tar.bz2 || \
> die "Unable to unpack
> iptables-${IPTVERINS}.tar.bz2"
Could you just call unpack() here?
> einfo "Enter path to your kernel sources, relative to ${ROOT}"
> echo -n "[/usr/src/linux]: "
> read K_DIR
> [ "${ROOT}${K_DIR}" == "/" ] && K_DIR=${ROOT}/usr/src/linux
/usr/src/linux should be what's used here, and if they want to change
the kernel to affect, they should change where the symlink points
instead of having a secondary config mechanism.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sci-mathematics/mathomatic: ChangeLog mathomatic-12.7.9.ebuild
On 11:17 Mon 05 Nov , Marcus D. Hanwell wrote: > On Friday 02 November 2007 02:59:36 Donnie Berkholz wrote: > > Could you drop the icc USE flag? All you need to do is make sure this > > package respects CC, CFLAGS and LDFLAGS, and users can take care of the > > rest in make.conf. > > > I never put it in and just left it alone as I don't use ICC. I have dropped > it > in the testing version but would rather leave the stable version alone. So if > nobody objects it will go from the tree once I stabilise this version. > > Is this policy that we are getting rid of icc stuff now? I think kugelfang > added it but didn't check to be honest. Just so I know when doing future > bumps. icc should be treated as a compiler, not as a USE flag. So if all you need to do is set up CC, CFLAGS, CXX and the like, just make sure the package respects the environment. If there's something else you need to do, such as set up a special config file for icc, you can do it based on tc-getCC() or tc-getCXX() rather than USE=icc. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-libs/db: ChangeLog db-4.6.21.ebuild
On 18:58 Sun 04 Nov , Caleb Tennis (caleb) wrote:
> 1.1 sys-libs/db/db-4.6.21.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-libs/db/db-4.6.21.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-libs/db/db-4.6.21.ebuild?rev=1.1&content-type=text/plain
> for (( i=1 ; i<=${PATCHNO} ; i++ )) ; do
> for (( i=1 ; i<=${PATCHNO} ; i++ ))
It reads strangely to mix ways of referencing variables like this. If
it's gonna be a C-style loop, then go all out:
for (( i=1 ; i<=PATCHNO ; i++ ))
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] RFC: cmake.eclass
On 12:51 Sun 04 Nov , Krzysiek Pawlik wrote:
> A little introduction: cmake is an alternative for autotools, more and more
> packages are using it (and some new big ones are on the way, KDE4 for
> example).
>
> I've wrote an eclass that makes writing ebuilds for such packages a little
> easier - it provides an ecmake function that takes care of few needed
> variables,
> prefix and such.
Great! When's the scons one coming? =)
> # Copyright 1999-2007 Gentoo Foundation
> # Distributed under the terms of the GNU General Public License v2
> # $Header: $
>
> #
> # Original Author: nelchael
> # Purpose: Automate src_install and src_compile for packages using cmake
> #
This would be a great opportunity to start using real eclass
documentation (e.g. that found in eutils.eclass) so we can autogenerate
a useful manpage on how to use it.
> # If you want to build in source tree set CMAKE_IN_SOURCE_BUILD to anything:
> [[ -n "${CMAKE_IN_SOURCE_BUILD}" ]] && CMAKE_BUILD_DIR="${S}"
Why would I want to do that? Some documentation would help.
> function ecmake() {
Why are some functions declared with 'function' and others not?
> -DCMAKE_CXX_COMPILER="$(type -P $(tc-getCXX))" \
Why the 'type -P' bit?
> make DESTDIR="${D}" install || die "make install failed"
Does emake work?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-proxy/polipo: ChangeLog polipo-1.0.3.ebuild polipo-1.0.1.ebuild polipo-1.0.0.ebuild polipo-0.9.12-r1.ebuild
On 10:37 Sat 03 Nov , Alin Năstac wrote: > You have to create users and groups in both pkg_ functions: > - pkg_setup : when installation is performed the usual way, through > compilation > - pkg_preinst: when installed from binary package (-k) You shouldn't, all pkg_* functions should always be run regardless of binary or source package. File a portage bug if pkg_setup() isn't getting run, don't work around it. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-proxy/polipo: ChangeLog polipo-1.0.3.ebuild polipo-1.0.1.ebuild polipo-1.0.0.ebuild polipo-0.9.12-r1.ebuild
On 15:38 Fri 02 Nov , Alin Nastac (mrness) wrote:
> 1.1 net-proxy/polipo/polipo-1.0.3.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-proxy/polipo/polipo-1.0.3.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-proxy/polipo/polipo-1.0.3.ebuild?rev=1.1&content-type=text/plain
> pkg_setup() {
> enewgroup polipo
> enewuser polipo -1 -1 /var/cache/polipo polipo
> }
...
> pkg_preinst() {
> pkg_setup
> }
What's up with this?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sci-mathematics/mathomatic: ChangeLog mathomatic-12.7.9.ebuild
On 02:37 Fri 02 Nov , Marcus Hanwell (cryos) wrote:
> 1.1 sci-mathematics/mathomatic/mathomatic-12.7.9.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sci-mathematics/mathomatic/mathomatic-12.7.9.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sci-mathematics/mathomatic/mathomatic-12.7.9.ebuild?rev=1.1&content-type=text/plain
> IUSE="doc icc"
>
> DEPEND="sys-libs/readline
> sys-libs/ncurses
> icc? ( dev-lang/icc )"
>
> src_compile() {
> if use icc; then
> CC="icc" CFLAGS="-O3 -axKWNBP -ipo" LDFLAGS="-O3 -axKWNBP -ipo
> -limf" emake READLINE=1 || die "emake failed"
> else
> emake READLINE=1 || die "emake failed"
> fi
Could you drop the icc USE flag? All you need to do is make sure this
package respects CC, CFLAGS and LDFLAGS, and users can take care of the
rest in make.conf.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-p2p/deluge: ChangeLog deluge-0.5.6.2.ebuild deluge-0.5.6.1.ebuild
On 02:22 Thu 01 Nov , Mike Frysinger wrote: > On Wednesday 31 October 2007, Donnie Berkholz wrote: > > On 15:38 Wed 31 Oct , Raul Porcel (armin76) wrote: > > > filter-ldflags -Wl,--as-needed > > > > If you moved the filter-ldflags() call up to pkg_setup(), you could drop > > src_compile() altogether to clean up the ebuild a little. > > a better idea is to fix the problem instead of ignoring it with filter-ldflags Brilliant, my dear Watson! For anyone who doesn't know how to fix --as-needed problems, we've got a guide: http://www.gentoo.org/proj/en/qa/asneeded.xml Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-misc/nxserver-freenx: nxserver-freenx-0.7.0-r1.ebuild ChangeLog nxserver-freenx-0.7.1.ebuild
On 19:49 Wed 31 Oct , Bernard Cafarelli (voyageur) wrote:
> 1.1 net-misc/nxserver-freenx/nxserver-freenx-0.7.1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-misc/nxserver-freenx/nxserver-freenx-0.7.1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-misc/nxserver-freenx/nxserver-freenx-0.7.1.ebuild?rev=1.1&content-type=text/plain
> pkg_postinst () {
> usermod -s /usr/bin/nxserver nx || die "Unable to set login shell of nx
> user!!"
> usermod -d ${NX_HOME_DIR} nx || die "Unable to set home directory of nx
> user!!"
This isn't safe with ROOT != / and it looks wrong too, you oughta be
using enewuser for this stuff.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-p2p/deluge: ChangeLog deluge-0.5.6.2.ebuild deluge-0.5.6.1.ebuild
On 19:25 Wed 31 Oct , Marijn Schouten (hkBst) wrote: > Donnie Berkholz wrote: > > If you moved the filter-ldflags() call up to pkg_setup(), you could drop > > src_compile() altogether to clean up the ebuild a little. > > Wouldn't that make binary packages cry? Binary packages don't do any linking, so they don't pay any attention to LDFLAGS. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-p2p/deluge: ChangeLog deluge-0.5.6.2.ebuild deluge-0.5.6.1.ebuild
On 15:38 Wed 31 Oct , Raul Porcel (armin76) wrote:
> 1.1 net-p2p/deluge/deluge-0.5.6.2.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-p2p/deluge/deluge-0.5.6.2.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-p2p/deluge/deluge-0.5.6.2.ebuild?rev=1.1&content-type=text/plain
> pkg_setup() {
> if has_version " ! built_with_use "dev-libs/boost" threads; then
> eerror "dev-libs/boost has to be built with threads USE-flag."
> die "Missing threads USE-flag for dev-libs/boost"
> fi
> }
>
> src_compile() {
> filter-ldflags -Wl,--as-needed
>
> distutils_src_compile
> }
If you moved the filter-ldflags() call up to pkg_setup(), you could drop
src_compile() altogether to clean up the ebuild a little.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-admin/webmin: ChangeLog webmin-1.370-r1.ebuild
On 10:36 Tue 30 Oct , Roy Marples (uberlord) wrote:
> 1.1 app-admin/webmin/webmin-1.370-r1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-admin/webmin/webmin-1.370-r1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-admin/webmin/webmin-1.370-r1.ebuild?rev=1.1&content-type=text/plain
> (find . -name '*.cgi' ; find . -name '*.pl') | perl perlpath.pl
> /usr/bin/perl -
Please don't tell me the -or find flag isn't POSIX so we can't use it...
> cp -rp * ${D}/usr/libexec/webmin
Lots of quote stuff here, repoman should be catching that.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-emulation/xen: ChangeLog xen-3.1.1.ebuild
On 00:31 Wed 31 Oct , Mike Frysinger wrote:
> On Monday 15 October 2007, Donnie Berkholz wrote:
> > On 18:33 Mon 15 Oct , Micheal Marineau (marineam) wrote:
> > > 1.1 app-emulation/xen/xen-3.1.1.ebuild
> > >
> > > file :
> > > http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-emulation/xen/xen-3.1
> > >.1.ebuild?rev=1.1&view=markup plain:
> > > http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-emulation/xen/xen-3.1
> > >.1.ebuild?rev=1.1&content-type=text/plain
> > >
> > > src_compile() {
> > > local myopt
> > > use debug && myopt="${myopt} debug=y"
> > > use pae && myopt="${myopt} pae=y"
> > >
> > > if use custom-cflags; then
> > > filter-flags -fPIE -fstack-protector
> > > else
> > > unset CFLAGS
> > > fi
> > >
> > > # Send raw LDFLAGS so that --as-needed works
> > > emake LDFLAGS="$(raw-ldflags)" -C xen ${myopt} || die "compile failed"
> > > }
> > >
> > > src_install() {
> > > local myopt
> > > use debug && myopt="${myopt} debug=y"
> > > use pae && myopt="${myopt} pae=y"
> >
> > The environment should be preserved across phases, so you shouldn't have
> > to repeat this stuff.
>
> "myopt" is declared "local"
Sure, but did you notice it's set to the exact thing in both phases?
Seems a bit pointless.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-irc/mistbot: mistbot-0.9.ebuild metadata.xml Manifest ChangeLog
On 22:45 Mon 29 Oct , Dawid Weglinski (cla) wrote:
> Revision ChangesPath
> 1.1 net-irc/mistbot/mistbot-0.9.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-irc/mistbot/mistbot-0.9.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-irc/mistbot/mistbot-0.9.ebuild?rev=1.1&content-type=text/plain
> src_compile() {
> ebegin "Change build.conf to fit needs"
>
> if ! use xml; then
> sed -e '/^MODS\ +=\ rss$/d' -i build.conf || die "sed failed"
> fi
>
> if ! use nls; then
> sed -e '/^NLS = nls$/d' -i build.conf || die "sed failed"
> fi
>
> if ! use debug; then
> sed -e '/^DEBUG = debug$/d' -i build.conf || die "sed failed"
> fi
>
> if ! use ssl; then
> sed -e '/^SSL = ssl$/d' -i build.conf || die "sed failed"
> fi
>
> sed -e 's/^#ONCE\ =\ yes$/ONCE\ =\ yes/' -i build.conf || die "sed
> failed"
What's up with some of these escaping spaces, and others not?
> echo "CXXFLAGS=${CXXFLAGS}" >> build.conf
> echo "CXX=$(tc-getCXX)" >> build.conf
> echo "PREFIX=/usr" >> build.conf
>
> ebegin "compiling source"
>
> emake all-oneGo || die "emake failed"
>
> if use doc; then
> ebegin "generate documentation"
> make doc || die "make doc failed"
> fi
> }
>
> src_install() {
> make install DESTDIR="${D}" || die "make install failed"
Any particular reason 'emake' isn't used here or in the 'make doc' call?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-analyzer/nagios-plugins: ChangeLog nagios-plugins-1.4.10-r1.ebuild
On 19:51 Mon 29 Oct , Tobias Scherbaum (dertobi123) wrote:
> 1.1
> net-analyzer/nagios-plugins/nagios-plugins-1.4.10-r1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-analyzer/nagios-plugins/nagios-plugins-1.4.10-r1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-analyzer/nagios-plugins/nagios-plugins-1.4.10-r1.ebuild?rev=1.1&content-type=text/plain
> src_install() {
> mv "${S}"/contrib/check_compaq_insight.pl
> "${S}"/contrib/check_compaq_insight.pl.msg
> chmod +x "${S}"/contrib/*.pl
>
> sed -i -e '1s;#!.*;#!/usr/bin/perl -w;' "${S}"/contrib/*.pl || die "sed
> failed"
> sed -i -e '30s/use lib utils.pm;/use utils;/' \
> "${S}"/plugins-scripts/check_file_age.pl || die "sed failed"
>
> dodoc ABOUT-NLS ACKNOWLEDGEMENTS AUTHORS BUGS CODING \
> ChangeLog FAQ NEWS README REQUIREMENTS SUPPORT THANKS
>
> emake DESTDIR="${D}" install || die "make install failed"
>
> if use mysql || use postgres; then
> dodir /usr/nagios/libexec
> exeinto /usr/nagios/libexec
> doexe "${S}"/contrib/check_nagios_db.pl
> fi
>
> dodir /usr/nagios/libexec/
> mv "${S}"/contrib "${D}"/usr/nagios/libexec/contrib
>
> chown root:nagios "${D}"/usr/nagios || die "Failed Chown of
> ${D}usr/nagios"
> chown -R root:nagios "${D}"/usr/nagios/libexec || die "Failed Chown of
> ${D}usr/nagios/libexec"
>
> chmod -R o-rwx "${D}"/usr/nagios/libexec || die "Failed Chmod of
> ${D}usr/nagios/libexec"
>
> chmod 04710 "${D}"/usr/nagios/libexec/check_icmp || die "Failed Chmod
> of ${D}usr/nagios/libexec/check_icmp"
This may not be worth changing, but if you're ever working on it,
fperms/fowners could be nice to use instead of these. It lets you clean
out all those references to $D.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-util/bless: ChangeLog bless-0.5.2.ebuild
On 18:56 Sun 28 Oct , Saleem Abdulrasool (compnerd) wrote:
> 1.1 dev-util/bless/bless-0.5.2.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-util/bless/bless-0.5.2.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-util/bless/bless-0.5.2.ebuild?rev=1.1&content-type=text/plain
> src_unpack() {
> gnome2_src_unpack
>
> einfo "Running gettextize -f --no-changelog..."
> ( "${T}/gettextize" -f --no-changelog > /dev/null ) || die "gettexize
> failed"
> eautoreconf
Why the subshell?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Opinions Wanted - Arrays again :)
On 22:49 Thu 25 Oct , Roy Marples wrote: > On Thu, 2007-10-25 at 14:31 -0700, Donnie Berkholz wrote: > > Is there any way we could avoid these altogether, and instead use > > separate variables for each array element? > > Well, we could prefix with numbers > > array="1.2.3.4 netmask 5.6.7.8; > \* > 'host.name' netmask 1.2.3.4 > -I 'option; $FOO with spaces'" > > Would become > > 0_config_eth0="1.2.3.4 netmask 5.6.7.8;" > 1_config_eth0="\*" > 2_config_eth0="'host.name' netmask 1.2.3.4" > 3_config_eth0="-I 'option; $FOO with spaces'" I was hoping for some sort of meaningfully named separate variables, not an even messier fake array. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Opinions Wanted - Arrays again :)
On 16:40 Thu 25 Oct , Roy Marples wrote:
> Hello List
>
> It's your favourite posix shell lover here, asking for your honest
> opinions.
>
> array="1.2.3.4 netmask 5.6.7.8;
> \*
> 'host.name' netmask 1.2.3.4
> -I 'option; $FOO with spaces'
> "
>
> array=("1.2.3.4 netmask 5.6.7.8;"
> "\*"
> "'host.name' netmask 1.2.3.4"
> "-I 'option; $FOO with spaces'"
> )
>
> array="'1.2.3.4 netmask 5.6.7.8;' \
> '\*' \
> \"'host.name' netmask 1.2.3.4\" \
> \"-I 'option; $FOO with spaces'\"
> "
>
> The first and last are of course posix constructs whilst the middle is
> bash.
>
> The bash one is purely there for reference, and of course you can still
> use it if /bin/sh is bash. The last one is what baselayout-2 currently
> uses and I'm wondering if we should switch to the first one before we
> come out of package.mask.
>
> I'm asking which you think are the most readable of the first and last
> ones and if you see any issues with either.
Is there any way we could avoid these altogether, and instead use
separate variables for each array element?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sci-mathematics/calc: calc-2.12.1.5.ebuild ChangeLog calc-2.11.9.3.ebuild calc-2.11.10.1.ebuild calc-2.12.2.2.ebuild
On 12:50 Wed 24 Oct , Markus Dittrich (markusle) wrote:
> 1.1 sci-mathematics/calc/calc-2.12.2.2.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sci-mathematics/calc/calc-2.12.2.2.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sci-mathematics/calc/calc-2.12.2.2.ebuild?rev=1.1&content-type=text/plain
> src_compile() {
> make \
If emake doesn't work, please add a comment to that effect.
> T="${D}" \
> DEBUG="${CFLAGS}" \
> CALCPAGER=less \
> USE_READLINE="-DUSE_READLINE" \
> READLINE_LIB="-lreadline -lhistory -lncurses" \
> all \
> || die
> if echo "${LD_PRELOAD}" | grep -q "sandbox"; then
> ewarn "Can't run check when running in sandbox - see bug #59676"
> else
> make chk || die "Check failed"
Shouldn't this 'make chk' bit be in src_test() ?
> fi
> }
>
> src_install() {
> make T="${D}" install || die
If emake doesn't work, please add a comment to that effect.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in x11-libs/compiz-bcop: metadata.xml Manifest compiz-bcop-0.6.0.ebuild ChangeLog
On 22:28 Tue 23 Oct , Hanno Boeck (hanno) wrote:
> 1.1 x11-libs/compiz-bcop/compiz-bcop-0.6.0.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/x11-libs/compiz-bcop/compiz-bcop-0.6.0.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/x11-libs/compiz-bcop/compiz-bcop-0.6.0.ebuild?rev=1.1&content-type=text/plain
> S="${WORKDIR}/${P}"
>
> src_compile() {
> econf || die "econf failed"
> emake || die "make failed"
> }
Both S and src_compile() are the defaults, and this is not the only
compiz ebuild committed today where that is the case.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-freebsd/freebsd-lib: ChangeLog freebsd-lib-6.2-r3.ebuild
On 12:02 Tue 23 Oct , Roy Marples (uberlord) wrote:
> 1.1 sys-freebsd/freebsd-lib/freebsd-lib-6.2-r3.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-freebsd/freebsd-lib/freebsd-lib-6.2-r3.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-freebsd/freebsd-lib/freebsd-lib-6.2-r3.ebuild?rev=1.1&content-type=text/plain
> PATCHES="${FILESDIR}/${PN}-bsdxml.patch
> ${FILESDIR}/${PN}-6.0-pmc.patch
> ${FILESDIR}/${PN}-6.0-gccfloat.patch
> ${FILESDIR}/${PN}-6.0-flex-2.5.31.patch
> ${FILESDIR}/${PN}-6.0-binutils-asm.patch
> ${FILESDIR}/${PN}-6.0-ssp.patch
> ${FILESDIR}/${PN}-6.1-csu.patch
> ${FILESDIR}/${PN}-6.2-bluetooth.patch
> ${FILESDIR}/${PN}-6.2-gcc41.patch
> ${FILESDIR}/${PN}-6.2-dl_iterate_phdr.patch
> ${FILESDIR}/${PN}-6.2-as-needed.patch
> ${FILESDIR}/${PN}-6.2-libthr.patch"
Would it make this work with space-filled paths to single-quote each
patch? I tried a quick test, and it looks like it would.
> # Add symlinks (-> libthr) for legacy threading libraries, since these
> are
> # not built by us (they are disabled in FreeBSD-7 anyway).
> dosym libthr.a /usr/lib/libpthread.a
> dosym libthr.so /usr/lib/libpthread.so
> dosym libthr.a /usr/lib/libc_r.a
> dosym libthr.so /usr/lib/libc_r.so
>
> # Add symlink (-> libthr) so previously built binaries still work.
> dosym libthr.so.2 /lib/libpthread.so.2
> dosym libthr.so.2 /lib/libc_r.so.6
>
> # Compatibility symlinks to run FreeBSD 5.x binaries (ABI is mostly
> # identical, remove when problems will actually happen)
> dosym /lib/libc.so.6 /usr/lib/libc.so.5
> dosym /lib/libm.so.4 /usr/lib/libm.so.3
FreeBSD doesn't use get_libdir() ?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-mail/mailgraph: ChangeLog mailgraph-1.14.ebuild
On 18:24 Mon 22 Oct , Tobias Scherbaum (dertobi123) wrote:
> 1.1 net-mail/mailgraph/mailgraph-1.14.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-mail/mailgraph/mailgraph-1.14.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-mail/mailgraph/mailgraph-1.14.ebuild?rev=1.1&content-type=text/plain
> user_group_setup() {
> # add user and group for mailgraph daemon
> # also add mgraph to the group adm so it's able to
> # read syslog logfile /var/log/messages (should be owned by
> # root:adm with permission 0640)
> enewgroup mgraph
> enewuser mgraph -1 -1 /var/empty mgraph,adm
> }
>
> pkg_setup() {
> webapp_pkg_setup
> built_with_use net-analyzer/rrdtool perl \
> || die "net-analyzer/rrdtool must be built with USE=perl"
> user_group_setup
> }
...
> pkg_preinst() {
> user_group_setup
> }
Why is user_group_setup() run twice?
> if [[ has_version <=net-mail/mailgraph-1.12 ]] ; then
Does this actually work?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-portage/portato: ChangeLog portato-0.8.6.ebuild portato-0.8.5.ebuild
On 17:03 Sat 20 Oct , Markus Ullmann (jokey) wrote:
> 1.1 app-portage/portato/portato-0.8.6.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-portage/portato/portato-0.8.6.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-portage/portato/portato-0.8.6.ebuild?rev=1.1&content-type=text/plain
> apply_sed ()
> {
> cd "${S}/${PN}"
>
> }
Uhh, what's this doing? It's not called anywhere in the ebuild. I'd
guess it's an artifact from an older version.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sci-biology/seaview: ChangeLog seaview-20060213.ebuild seaview-20041220.ebuild seaview-20070919.ebuild
On 21:20 Sun 21 Oct , Alec Warner wrote: > You told me on IRC 'people should quote all the time, as there is no > harm in overquoting' (aside from readability). Yet here you tell him > to 'not blindly quote everything'. So make up your mind ;) I was referring to the variables that actually need quoting -- just quoting them every time they're used. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
Re: [gentoo-dev] RFC: new project gentoo-extreme-security
On 01:42 Mon 22 Oct , Alexander Gabert wrote: > this is a request for comments on a new project: > > http://www.gentoo.org/proj/en/extreme-security/ > > When the webservers have finished mirroring the xml file, you can view the > description and the goals of the preliminary version of the new project > page for the Gentoo Extreme Security Project. The new project is about > sharing a new vision and building a development platform for new ideas and > collecting input about making Gentoo Linux even more attractive for > professional server environments and increasing the security and resilience > factor of the distribution beyond the currently possible achievements! I'm curious whether this would be better-placed as a subproject of either the security or hardened projects. Why do you think it would be better off independent? Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sci-biology/seaview: ChangeLog seaview-20060213.ebuild seaview-20041220.ebuild seaview-20070919.ebuild
On 04:34 Sun 21 Oct , Jeffrey Gardner (je_fro) wrote:
> 1.3 sci-biology/seaview/seaview-20060213.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sci-biology/seaview/seaview-20060213.ebuild?rev=1.3&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sci-biology/seaview/seaview-20060213.ebuild?rev=1.3&content-type=text/plain
> diff :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sci-biology/seaview/seaview-20060213.ebuild?r1=1.2&r2=1.3
> @@ -41,6 +41,6 @@
>
> src_install() {
> dobin seaview seaview_align.sh
> - insinto /usr/share/${PN}
> + insinto "/usr/share/${PN}"
You only need to do this if $PN could have spaces in it (which should
never be the case). Don't just blindly quote everything; by
understanding why it's done, you can understand in which cases it should
be done.
I hope that repoman did not suggest this.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-print/hplip: ChangeLog hplip-2.7.9-r1.ebuild hplip-2.7.7-r3.ebuild hplip-2.7.9.ebuild
On 13:08 Thu 18 Oct , Denis Dupeyron (calchan) wrote:
> 1.1 net-print/hplip/hplip-2.7.9-r1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-print/hplip/hplip-2.7.9-r1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-print/hplip/hplip-2.7.9-r1.ebuild?rev=1.1&content-type=text/plain
> pkg_setup() {
> if ! use ppds ; then
> ewarn "Not installing built-in PPD files, which is probably not
> what you want."
> ewarn "You need USE=ppds if you want to install them."
> fi
> if use minimal ; then
> ewarn "Installing hpijs driver only, make sure you know what
> you are doing."
If this is probably not what you want, is it ever what you want?
> [ -e /etc/sane.d/dll.conf ] && cp /etc/sane.d/dll.conf .
> [ -e "${ROOT}"/etc/sane.d/dll.conf ] && cp
> "${ROOT}"/etc/sane.d/dll.conf .
I'm a bit confused about why you need the first of these lines.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-apps/sandbox: ChangeLog sandbox-1.2.18.1-r1.ebuild
On 15:55 Wed 17 Oct , Daniel Drake (dsd) wrote:
> 1.1 sys-apps/sandbox/sandbox-1.2.18.1-r1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-apps/sandbox/sandbox-1.2.18.1-r1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-apps/sandbox/sandbox-1.2.18.1-r1.ebuild?rev=1.1&content-type=text/plain
> keepdir /var/log/sandbox
> fowners root:portage /var/log/sandbox
> fperms 0770 /var/log/sandbox
>
> cd ${S}
> dodoc AUTHORS ChangeLog NEWS README
> }
>
> pkg_preinst() {
> chown root:portage ${D}/var/log/sandbox
> chmod 0770 ${D}/var/log/sandbox
> }
How come you need to repeat the permissions like this?
Also, quoting. =)
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: elibtoolize and when to drop it from an ebuild
On 19:51 Wed 17 Oct , Duncan wrote: > What about putting it in an fbsd conditional? That way Linux users don't > have to deal with the step if it's unnecessary for them, while the fbsd > users still get it when they need it. Seems to me this should be a > workable compromise, where people might otherwise be inclined to remove > it. I prefer to avoid conditionals wherever reasonable, because then you just have more code paths to test. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: Re: [gentoo-commits] gentoo-x86 commit in dev-php5/onphp: ChangeLog onphp-0.10.6.ebuild onphp-0.10.4.ebuild onphp-0.10.3.ebuild
On 08:19 Tue 16 Oct , Steve Long wrote:
> stdDocs=(ChangeLog AUTHORS FEATURES THANKS README CodingStyle TODO FAQ)
>
> stdDoc() {
> local d l=()
> for d in "[EMAIL PROTECTED]"; do
> [[ -f $d ]] && l+=("$d")
> done
> (([EMAIL PROTECTED])) && dodoc "[EMAIL PROTECTED]"
> }
>
> (assuming [[ $PWD = $S ]] or whichever is the right one for docs.)
Is it just me, or is this code way more complex than it needs to be? It
looks like bashturbation to me. =P
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-office/gnotime: ChangeLog gnotime-2.2.3.ebuild
On 21:34 Mon 15 Oct , Christian Faulhammer (opfer) wrote:
> 1.1 app-office/gnotime/gnotime-2.2.3.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-office/gnotime/gnotime-2.2.3.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-office/gnotime/gnotime-2.2.3.ebuild?rev=1.1&content-type=text/plain
> pkg_setup() {
> # upstream knows about the fix and has promised to incorporate it
> if has_version ">=dev-scheme/guile-1.8" && ! built_with_use
> dev-scheme/guile deprecated;then
> eerror "rebuild dev-scheme/guile with USE=deprecated"
> die
> fi
> }
You could drop the has_version check and pass '--missing true' to
built_with_use().
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-libs/qof: qof-0.7.2.ebuild metadata.xml ChangeLog Manifest
On 21:28 Mon 15 Oct , Christian Faulhammer (opfer) wrote:
> src_compile() {
> econf || die "econf failed"
> emake || die "emake failed"
> }
This is the default, you can get rid of it.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-emulation/xen: ChangeLog xen-3.1.1.ebuild
On 18:33 Mon 15 Oct , Micheal Marineau (marineam) wrote:
> 1.1 app-emulation/xen/xen-3.1.1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-emulation/xen/xen-3.1.1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-emulation/xen/xen-3.1.1.ebuild?rev=1.1&content-type=text/plain
> src_compile() {
> local myopt
> use debug && myopt="${myopt} debug=y"
> use pae && myopt="${myopt} pae=y"
>
> if use custom-cflags; then
> filter-flags -fPIE -fstack-protector
> else
> unset CFLAGS
> fi
>
> # Send raw LDFLAGS so that --as-needed works
> emake LDFLAGS="$(raw-ldflags)" -C xen ${myopt} || die "compile failed"
> }
>
> src_install() {
> local myopt
> use debug && myopt="${myopt} debug=y"
> use pae && myopt="${myopt} pae=y"
The environment should be preserved across phases, so you shouldn't have
to repeat this stuff.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in media-tv/mythtv: ChangeLog mythtv-0.20.2_p14668.ebuild mythtv-0.21_pre14666.ebuild mythtv-0.21_pre14480-r1.ebuild
On 16:57 Mon 15 Oct , Doug Goldstein wrote: > I still stand by my original feeling that we'd better the community NOT > only the developers doing the commits by updating the devmanual, which > is accessible to all developers and all users in the Gentoo community. > In addition to updating and cleaning up repoman checks, which is a tool > that everyone in the community can use. This is versus individual > examples in random ebuilds in random e-mails that all have almost an > identical subject on the mailing list. > > The commits review is flawed because if we're not documenting this stuff > in one central place, then when new developers join. The same lessons > have to be learned over and over again. Doing the review helps us figure out what the common issues are, so we actually know what's worth documenting. I've had in mind all along getting common issues into docs and into repoman, which is why I wrote the original quoting patch. Also, just because something is documented doesn't mean people do it. Furthermore, just because a doc says "write code without bugs" doesn't mean people are physically capable of doing so. More eyes on the code can never hurt. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-misc/note: ChangeLog note-1.3.3.ebuild
On 09:56 Mon 15 Oct , Bo Ørsted Andresen wrote:
> Written like this u certainly can't since the output of the find expression is
> subjected to word splitting before u gets set. Hence this will fail if ${D}
> contains spaces even if ${u} gets quoted. The best way to solve this is to use
> find ... -print0 | xargs -0 rm. Unfortunately that solution means using GNU
> extensions that I'm not sure are guaranteed to be available in the ebuild
> environment. If they aren't maybe they should be.
Cool, I hadn't dealt with this before. Thanks for pointing that out! We
were just talking about whether to require GNU find the other day, but I
don't recall the conclusion. Does anyone else?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-misc/note: ChangeLog note-1.3.3.ebuild
On 07:40 Mon 15 Oct , Christian Faulhammer (opfer) wrote:
> 1.1 app-misc/note/note-1.3.3.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-misc/note/note-1.3.3.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-misc/note/note-1.3.3.ebuild?rev=1.1&content-type=text/plain
> for v in mysql text dbm general; do
> if ! use ${v}; then
> for u in `find "${D}" -type f -name *${v}.*pm`; do
> rm ${u}
Looks like 'u' could have spaces in it, since it's based on D.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-php5/onphp: ChangeLog onphp-0.10.6.ebuild onphp-0.10.4.ebuild onphp-0.10.3.ebuild
On 16:52 Sun 14 Oct , Drake Wyrm wrote:
> Donnie Berkholz <[EMAIL PROTECTED]> wrote:
> > On 18:05 Sun 14 Oct , Konstantin Arkhipov (voxus) wrote:
> > > 1.1 dev-php5/onphp/onphp-0.10.6.ebuild
> > >
> > > file :
> > > http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-php5/onphp/onphp-0.10.6.ebuild?rev=1.1&view=markup
> > > plain:
> > > http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-php5/onphp/onphp-0.10.6.ebuild?rev=1.1&content-type=text/plain
> >
> > > if use doc ; then
> > > for doc in `find doc -maxdepth 1 -type f -print` ; do
> > > dodoc ${doc}
> > > done
> >
> > You could avoid all these calls to dodoc by saving the docs in a
> > variable, then calling dodoc on it. Another option could be to send the
> > loop output to 'xargs dodoc'.
>
> Or, since you already have the command substitution there, how about:
>
> dodoc `find doc -maxdepth 1 -type f`
Much better than what I said.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-p2p/btg: btg-0.9.6.ebuild Manifest metadata.xml ChangeLog
On 23:37 Sun 14 Oct , Christoph Mende (angelos) wrote: > Revision ChangesPath > 1.1 net-p2p/btg/btg-0.9.6.ebuild > > file : > http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-p2p/btg/btg-0.9.6.ebuild?rev=1.1&view=markup > plain: > http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-p2p/btg/btg-0.9.6.ebuild?rev=1.1&content-type=text/plain > einfo > einfo "Compile dev-libs/boost with USE=threads or > USE=threads-only" > einfo "if you want threading support for btg" > einfo Might be nice to switch the middle two to elog. > einfo > einfo "BTG needs a daemon.ini and client.ini, to create them run > btg-config" > einfo "and put them in the user running btg (/home/p2p by default)" > einfo Same.. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-php5/onphp: ChangeLog onphp-0.10.6.ebuild onphp-0.10.4.ebuild onphp-0.10.3.ebuild
On 18:05 Sun 14 Oct , Konstantin Arkhipov (voxus) wrote:
> 1.1 dev-php5/onphp/onphp-0.10.6.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-php5/onphp/onphp-0.10.6.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-php5/onphp/onphp-0.10.6.ebuild?rev=1.1&content-type=text/plain
> if use doc ; then
> for doc in `find doc -maxdepth 1 -type f -print` ; do
> dodoc ${doc}
> done
You could avoid all these calls to dodoc by saving the docs in a
variable, then calling dodoc on it. Another option could be to send the
loop output to 'xargs dodoc'.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] texlive cyrillic font : restrictive license
On 12:49 Sun 14 Oct , Alexis Ballier wrote: > while adding cyrillic packages for texlive I've come across a > restrictive license for the literat package. > You are allowed: ... > - to distribute fonts together with RFBR TeX distribution. > You are not allowed: ... > - to distribute this fonts as an independent package; This is the one part that concerns me. If you aren't doing this literally within one of the packages, rather than as a standalone package of its own, don't do it. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-fs/openafs: ChangeLog openafs-1.4.5_pre1.ebuild
On 10:53 Sun 14 Oct , Bo Ørsted Andresen wrote:
> On Sunday 14 October 2007 10:39:53 Donnie Berkholz wrote:
> > On 14:25 Sat 13 Oct , Stefaan De Roeck (stefaan) wrote:
> > > 1.1 net-fs/openafs/openafs-1.4.5_pre1.ebuild
> > >
> > > file :
> > > http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-fs/openafs/openafs-1.4.5_pre1.ebuild?rev=1.1&view=markup
> > > plain:
> > > http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-fs/openafs/openafs-1.4.5_pre1.ebuild?rev=1.1&content-type=text/plain
> > >
> > > PATCHDIR=${WORKDIR}/gentoo/patches/$(get_version_component_range 1-2)
> > > CONFDIR=${WORKDIR}/gentoo/configs
> > > SCRIPTDIR=${WORKDIR}/gentoo/scripts
> >
> > Repoman won't catch these, but they still need quotes.
>
> Actually they don't in assignments like these. In the rest of the
> ebuild ${PATCHDIR}, ${CONFDIR} and ${SCRIPTDIR} need quotes though
> (the same applies to variables such as ${oldafsconfdir},
> ${newafsconfdir} etc.).
Yes, that is what I meant. Sorry if it was unclear, since I was trying
to avoid repeating quotes of almost the same lines 20 times. References
to the variables require quotes, but the initial assignments do not. But
I don't see any particular problem if folks think it's easier to just
always quote than to remember the details.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in media-libs/libquicktime: ChangeLog libquicktime-1.0.1.ebuild
On 23:37 Sat 13 Oct , Alexis Ballier (aballier) wrote:
> 1.1 media-libs/libquicktime/libquicktime-1.0.1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-libs/libquicktime/libquicktime-1.0.1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-libs/libquicktime/libquicktime-1.0.1.ebuild?rev=1.1&content-type=text/plain
> pkg_setup() {
> if has_version '=x11-base/xorg-x11-6*' && ! built_with_use
> x11-base/xorg-x11 xv; then
> die "You need xv support to compile ${PN}."
> fi
> }
This can definitely drop, XOrg 6.x has been gone for ages.
> if $(has_version =media-libs/libquicktime-0.9.4); then
You don't need to use command substitution for this, just run it.
> elog "It seems this directory belongs to
> libquicktime-0.9.4."
> elog "We'll delete that directory now."
> rm -rvf /usr/include/quicktime
ROOT?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-misc/i810switch: ChangeLog metadata.xml i810switch-0.6.5-r2.ebuild
On 18:39 Sat 13 Oct , Cedric Krier (cedk) wrote: > cedk07/10/13 18:39:19 > > Modified: ChangeLog metadata.xml > Added:i810switch-0.6.5-r2.ebuild > Log: > Add support for i945, bug #184144 > (Portage version: 2.1.3.9) > DESCRIPTION="A utility for switching the LCD and external VGA displays on and > off" Is this thing still needed with the 2.0 intel driver, or can we drop it? Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-fs/openafs: ChangeLog openafs-1.4.5_pre1.ebuild
On 14:25 Sat 13 Oct , Stefaan De Roeck (stefaan) wrote:
> 1.1 net-fs/openafs/openafs-1.4.5_pre1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-fs/openafs/openafs-1.4.5_pre1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-fs/openafs/openafs-1.4.5_pre1.ebuild?rev=1.1&content-type=text/plain
> PATCHDIR=${WORKDIR}/gentoo/patches/$(get_version_component_range 1-2)
> CONFDIR=${WORKDIR}/gentoo/configs
> SCRIPTDIR=${WORKDIR}/gentoo/scripts
Repoman won't catch these, but they still need quotes.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-cluster/pvfs2: ChangeLog pvfs2-2.6.3-r1.ebuild
On 23:45 Sat 13 Oct , Drake Wyrm wrote:
> Donnie Berkholz <[EMAIL PROTECTED]> wrote:
> > On 13:36 Sat 13 Oct , Matti Bickel (mabi) wrote:
> > > if kernel_is gt 2 6 20 ; then
> > > epatch "${FILESDIR}"/${PV}-register_sysctl_table.patch
> > > fi
> > >
> > > if kernel_is ge 2 6 22 ; then
> > > epatch "${FILESDIR}"/${PV}-kmem-and-dtor-fix.patch
> > > fi
> >
> > Mixing 'gt' and 'ge' is a bad idea.
>
> Just outa curiosity, why?
Because it's inconsistent and one generally assumes that people will be
consistent with the way they test numbers. That way you only need to
read the number rather than continually checking every single line to
see how exactly it's tested for.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-cluster/pvfs2: ChangeLog pvfs2-2.6.3-r1.ebuild
On 13:36 Sat 13 Oct , Matti Bickel (mabi) wrote:
> 1.1 sys-cluster/pvfs2/pvfs2-2.6.3-r1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-cluster/pvfs2/pvfs2-2.6.3-r1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-cluster/pvfs2/pvfs2-2.6.3-r1.ebuild?rev=1.1&content-type=text/plain
> if kernel_is 2 4; then
> BUILD_TARGETS="just_kmod24"
> ECONF_PARAMS="--with-kernel24=${KV_DIR}"
> MODULE_NAMES="pvfs2(fs::src/kernel/linux-2.4)"
Feel free to drop 2.4 support.
> if kernel_is gt 2 6 20 ; then
> epatch "${FILESDIR}"/${PV}-register_sysctl_table.patch
> fi
>
> if kernel_is ge 2 6 22 ; then
> epatch "${FILESDIR}"/${PV}-kmem-and-dtor-fix.patch
> fi
Mixing 'gt' and 'ge' is a bad idea.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in media-tv/mythtv: ChangeLog mythtv-0.20.2_p14668.ebuild mythtv-0.21_pre14666.ebuild mythtv-0.21_pre14480-r1.ebuild
On 00:12 Sun 14 Oct , Doug Goldstein wrote: > Because if you pass the inverse the script blows up. It's ffmpeg's > configure script that's a hand written script and modified by the MythTV > developers. Sigh. Any chance of getting things to move to autotools? Thanks, Donnie -- [EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-python/gnome-python-extras: ChangeLog gnome-python-extras-2.19.1-r1.ebuild
On 11:11 Sat 13 Oct , Torsten Veller wrote:
> * Donnie Berkholz <[EMAIL PROTECTED]>:
> > On 10:44 Sat 13 Oct , Torsten Veller wrote:
> > > * Donnie Berkholz <[EMAIL PROTECTED]>:
> > > > python_mod_optimize() and python_mod_cleanup() already tag ROOT on,
> > >
> > > python_mod_optimize() does not.
> >
> > Mine does:
> >
> > python_mod_optimize() {
> > local myroot
> > # strip trailing slash
> > myroot="${ROOT%/}"
> >
> > ...
> >
> > ebegin "Byte compiling python modules for python-${PYVER} .."
> python${PYVER} ${myroot}/usr/$(get_libdir)/python${PYVER}/compileall.py
> ${compileopts} $@
> python${PYVER} -O
> ${myroot}/usr/$(get_libdir)/python${PYVER}/compileall.py ${compileopts} $@
> > eend $?
> > }
>
> No. The arguments of python_mod_optimize are just passed to the ROOTed
> compileall.py.
It took me about five minutes to figure out what you meant: this means
that the filenames do not get passed with ROOT prepended to them, so
compileall.py in ROOT compiles the files outside of ROOT.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-python/gnome-python-extras: ChangeLog gnome-python-extras-2.19.1-r1.ebuild
On 10:44 Sat 13 Oct , Torsten Veller wrote:
> * Donnie Berkholz <[EMAIL PROTECTED]>:
> > python_mod_optimize() and python_mod_cleanup() already tag ROOT on,
>
> python_mod_optimize() does not.
Mine does:
python_mod_optimize() {
local myroot
# strip trailing slash
myroot="${ROOT%/}"
...
ebegin "Byte compiling python modules for python-${PYVER} .."
python${PYVER} ${myroot}/usr/$(get_libdir)/python${PYVER}/compileall.py ${c$
python${PYVER} -O ${myroot}/usr/$(get_libdir)/python${PYVER}/compileall.py $
eend $?
}
> > so for ROOT != / this is broken.
>
> But does python_mod_optimize do anything useful with ROOT !=/ and
> cross-compilation? What if python versions differ?
That's a good question, and it looks like python_version() could use a
fix for ROOT != / on same-arch systems. On cross-compiled, it'll
probably be more work and I'm not sure how to deal with it.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in www-apps/tikiwiki: ChangeLog tikiwiki-1.9.8.1.ebuild
On 10:10 Sat 13 Oct , Gunnar Wrobel wrote: > > Could this respect ROOT? > > For a mysql db? I'm not sure I follow but the pkg_config was something > I should have removed anyhow. Sure. I could envision mounting / of a database server onto a build server and building with ROOT, to avoid sucking up much-needed CPU/memory on the DB server. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in media-tv/mythtv: ChangeLog mythtv-0.20.2_p14668.ebuild mythtv-0.21_pre14666.ebuild mythtv-0.21_pre14480-r1.ebuild
On 18:55 Fri 12 Oct , Doug Goldstein (cardoe) wrote:
> 1.1 media-tv/mythtv/mythtv-0.20.2_p14668.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-tv/mythtv/mythtv-0.20.2_p14668.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-tv/mythtv/mythtv-0.20.2_p14668.ebuild?rev=1.1&content-type=text/plain
> use alsa || myconf="${myconf} --disable-audio-alsa"
> use jack || myconf="${myconf} --disable-audio-jack"
> use dts || myconf="${myconf} --disable-dts"
> use freebox || myconf="${myconf} --disable-freebox"
> use dbox2 || myconf="${myconf} --disable-dbox2"
> use hdhomerun || myconf="${myconf} --disable-hdhomerun"
> use crciprec || myconf="${myconf} --disable-crciprec"
> use altivec || myconf="${myconf} --disable-altivec"
> use xvmc && myconf="${myconf} --enable-xvmc"
> use xvmc && use video_cards_via && myconf="${myconf} --enable-xvmc-pro"
> use perl && myconf="${myconf} --with-bindings=perl"
> myconf="${myconf}
> --disable-audio-arts
> $(use_enable lirc)
> $(use_enable joystick joystick-menu)
> $(use_enable dvb)
> --dvb-path=/usr/include
> $(use_enable opengl opengl-vsync)
> $(use_enable ieee1394 firewire)
> --enable-xrandr
> --enable-xv
> --disable-directfb
> --enable-x11
> --enable-proc-opt"
How come some of these don't use use_enable()?
> cd ${S}/contrib/channel_changers
Some quoting issues like this one.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in www-apps/zina: ChangeLog zina-1.0_rc3.ebuild
On 13:26 Fri 12 Oct , Gunnar Wrobel (wrobel) wrote:
> 1.1 www-apps/zina/zina-1.0_rc3.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/www-apps/zina/zina-1.0_rc3.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/www-apps/zina/zina-1.0_rc3.ebuild?rev=1.1&content-type=text/plain
Quoting issues again...
> src_install() {
> webapp_src_preinst
>
> touch ${S}/zina.ini.php
>
> cp -R . ${D}${MY_HTDOCSDIR}
>
> webapp_configfile ${MY_HTDOCSDIR}/zina.ini.php
> webapp_serverowned ${MY_HTDOCSDIR}/zina.ini.php
>
> keepdir ${MY_HTDOCSDIR}/_zina/cache
> webapp_serverowned ${MY_HTDOCSDIR}/_zina/cache
>
> webapp_postinst_txt en ${FILESDIR}/postinstall-en.txt
> webapp_src_install
> }
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-python/gnome-python-extras: ChangeLog gnome-python-extras-2.19.1-r1.ebuild
On 13:25 Fri 12 Oct , Remi Cardona (remi) wrote:
> 1.1
> dev-python/gnome-python-extras/gnome-python-extras-2.19.1-r1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-python/gnome-python-extras/gnome-python-extras-2.19.1-r1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-python/gnome-python-extras/gnome-python-extras-2.19.1-r1.ebuild?rev=1.1&content-type=text/plain
> pkg_postinst() {
> python_version
> python_mod_optimize
> "${ROOT}usr/$(get_libdir)/python${PYVER}/site-packages/gtk-2.0"
> }
>
> pkg_postrm() {
> python_version
> python_mod_cleanup
> "${ROOT}usr/$(get_libdir)/python${PYVER}/site-packages/gtk-2.0"
> }
python_mod_optimize() and python_mod_cleanup() already tag ROOT on, so
for ROOT != / this is broken.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-python/gnome-python: ChangeLog gnome-python-2.20.0.ebuild
On 12:46 Fri 12 Oct , Remi Cardona (remi) wrote:
> 1.1 dev-python/gnome-python/gnome-python-2.20.0.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-python/gnome-python/gnome-python-2.20.0.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-python/gnome-python/gnome-python-2.20.0.ebuild?rev=1.1&content-type=text/plain
> pkg_postinst() {
> python_version
> python_mod_optimize /usr/lib/python${PYVER}/site-packages/gtk-2.0
> }
>
> pkg_postrm() {
> python_version
> python_mod_cleanup
> }
Looks like you need get_libdir(), and you don't need to run
python_version() in postrm.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in www-apps/tikiwiki: ChangeLog tikiwiki-1.9.8.1.ebuild
On 12:35 Fri 12 Oct , Gunnar Wrobel (wrobel) wrote:
> 1.1 www-apps/tikiwiki/tikiwiki-1.9.8.1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/www-apps/tikiwiki/tikiwiki-1.9.8.1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/www-apps/tikiwiki/tikiwiki-1.9.8.1.ebuild?rev=1.1&content-type=text/plain
You've got quoting issues in here, are you using current portage?
> pkg_config() {
> elog "Type in your MySQL root password to create an empty tiki
> database:"
> mysqladmin -u root -p create tikiwiki
> }
Could this respect ROOT?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in eclass: gnatbuild.eclass
On 07:28 Fri 12 Oct , Steve Long wrote: > Donnie Berkholz wrote: > > > On 18:17 Wed 10 Oct , George Shapovalov (george) wrote: > >> george 07/10/10 18:17:58 > >> > >> Modified: gnatbuild.eclass > >> > >> Log: fixed src_install issue, no longer relies on portage leaking > >> env vars between functions > > > > It's really sad that you have to add this workaround. > > > If it's an env var that needs to be set in different phases, it doesn't > sound so bad to me? export or setting vars in the global ebuild scope is > fine imo. Portage is supposed to preserve the environment across functions. This workaround is added because that seems to break on a regular basis. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-misc/lirc: ChangeLog lirc-0.8.2-r2.ebuild
On 19:35 Thu 11 Oct , Matthias Schwarzott (zzam) wrote:
> 1.1 app-misc/lirc/lirc-0.8.2-r2.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-misc/lirc/lirc-0.8.2-r2.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-misc/lirc/lirc-0.8.2-r2.ebuild?rev=1.1&content-type=text/plain
> add_device() {
> ((lirc_device_count++))
[skip lots of code]
> lirc_driver_count=0
"driver" != "device"
Might be useful to initialize it in add_device() if it's unset, so code
being this far apart won't get out of sync.
> make DESTDIR="${D}" install || die "make install failed"
If emake doesn't work, please add a comment to that effect.
> newinitd ${FILESDIR}/lircd lircd
> newinitd ${FILESDIR}/lircmd lircmd
> newconfd ${FILESDIR}/lircd.conf lircd
>
> insinto /etc/modules.d/
> newins ${FILESDIR}/modulesd.lirc lirc
>
> newinitd ${FILESDIR}/irexec-initd irexec
> newconfd ${FILESDIR}/irexec-confd irexec
Quoting.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-apps/baselayout: ChangeLog baselayout-2.0.0_rc5.ebuild
On 15:56 Thu 11 Oct , Roy Marples (uberlord) wrote:
> 1.1 sys-apps/baselayout/baselayout-2.0.0_rc5.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-apps/baselayout/baselayout-2.0.0_rc5.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-apps/baselayout/baselayout-2.0.0_rc5.ebuild?rev=1.1&content-type=text/plain
> cp -p "${ROOT}usr/share/baselayout/${x}" ${ROOT}etc
Two ROOT's to quote here, not just one. =)
> # We need to copy svcdir if upgrading
> if has_version " (
> . "${ROOT}etc/conf.d/rc"
> svcdir="${svcdir:-/var/lib/init.d}"
> einfo "Moving state from ${ROOT}${svcdir} to
> ${ROOT}lib/rcscripts/init.d"
> cp -RPp "${ROOT}${svcdir}"/* "${ROOT}"lib/rcscripts/init.d
> rm -rf "${ROOT}"lib/rcscripts/init.d/daemons \
> "${ROOT}"lib/rcscripts/init.d/console
> umount "${ROOT}${svcdir}" 2>/dev/null
> rm -rf "${ROOT}${svcdir}"
> )
Can this be done in a code block instead, or do svdir and /etc/conf.d/rc
sourcing pollute things too badly?
> if has_version " (
> . "${ROOT}etc/conf.d/rc"
> svcdir="${svcdir:-/var/lib/init.d}"
> einfo "Moving state from ${ROOT}lib/rcscripts/init.d to
> ${ROOT}${svcdir}"
> mkdir -p "${ROOT}${svcdir}"
> cp -RPp "${ROOT}lib/rcscripts/init.d"/* "${ROOT}${svcdir}"
> rm -rf "${ROOT}${svcdir}"/daemons
> umount "${ROOT}lib/rcscripts/init.d" 2>/dev/null
> rm -rf "${ROOT}lib/rcscripts/init.d"
> )
Same question. Also, could this code, as well as the other cases I
cropped out, simply be abstracted into a function instead?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Council meeting summary for 11 October 2007
Hi all, Here is the summary from today's council meeting. The complete log will show up at http://www.gentoo.org/proj/en/council/ shortly. Thanks, Donnie Summary of 11 October 2007 council meeting Present: all members, josejx proxying for lu_zero = "What are not clear are (1) whether the Code of Conduct is in effect; (2) if so, how we enforce it." - The CoC is in effect, but it needs a new enforcement section since the proctors were disbanded. The council is sending discussion of this to the gentoo-project list, to come up with proposals for three points: - who enforces it - musikc said devrel could - tsunam said userrel could - how to enforce it - whether it's active or passive enforcement - which actions are appropriate - If the -project list does not come up with a draft, dberkholz will write one based on -project discussion to vote upon at the November council meeting. = packages.gentoo.org: https://bugs.gentoo.org/show_bug.cgi?id=187971 comment #86 from marduk on rewrite comment #90 re jokey's rewrite (comment #85) - The infrastructure team will decide the future of packages.gentoo.org. - KingTaco informed us that the current code will probably not return. Alternatives include: - http://packages.gentooext.net/ (written by jokey) - http://spaceparanoids.org/gentoo/gpnl/ (written by beandog) - http://gentoo-portage.com/ - Until we get a replacement, packages.gentoo.org will link to alternatives. It now links to a forums thread describing them. = GLEP 39: project RFC addition - We will amend the GLEP rather than writing a new one, and a note will be added saying the GLEP was amended. = When does the new council term end? - Voting will always take place the same month, as mentioned in http://thread.gmane.org/gmane.linux.gentoo.devel/52081/focus=52143 - If the new council is delayed, it gets a slightly shortened term. = Open floor - Where is the PMS repo? - robbat2 has imported it. It's in git. Need to ping him for details. - The channel was not moderated during the meeting and it went well. Let's try that again next time.
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-fs/evms: ChangeLog evms-2.5.5-r8.ebuild
On 11:13 Thu 11 Oct , Tiziano Müller wrote:
> Am Dienstag, 9. Oktober 2007 14.56:24 schrieb Doug Goldstein:
> > Donnie Berkholz wrote:
> > > On 22:01 Mon 08 Oct , Doug Goldstein (cardoe) wrote:
> > >> 1.1 sys-fs/evms/evms-2.5.5-r8.ebuild
> > >>
> > >> file :
> > >> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-fs/evms/evms-2.5.5-r
> > >>8.ebuild?rev=1.1&view=markup plain:
> > >> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-fs/evms/evms-2.5.5-r
> > >>8.ebuild?rev=1.1&content-type=text/plain
> > >>
> > >> epatch "${FILESDIR}/${PV}/md_super_fix.patch"
> > >> epatch "${FILESDIR}/${PV}/ntfs_unmkfs.patch"
> > >> epatch "${FILESDIR}/${PV}/raid5_degrade_fix_v2.patch"
> > >> epatch "${FILESDIR}/${PV}/raid5_remove_spare_fix.patch"
> > >> epatch "${FILESDIR}/${PV}/raid5_remove_spare_fix_2.patch"
> > >> epatch "${FILESDIR}/${PV}/raid5_algorithm.patch"
> > >> epatch "${FILESDIR}/${PV}/cli_reload_options.patch"
> > >> epatch "${FILESDIR}/${PV}/cli_query_segfault.patch"
> > >> epatch "${FILESDIR}/${PV}/get_geometry.patch"
> > >> epatch "${FILESDIR}/${PV}/BaseName.patch"
> > >> epatch "${FILESDIR}/${PV}/disk_cache.patch"
> > >>
> > >> epatch "${FILESDIR}/${P}-as-needed.patch"
> > >> epatch "${FILESDIR}/${P}-glib_dep.patch"
> > >> epatch "${FILESDIR}/${P}-ocfs2.patch"
> > >> epatch "${FILESDIR}/${P}-use_disk_group.patch"
> > >> epatch "${FILESDIR}/${P}-pagesize.patch"
> > >
> > > This would be another good candidate for using epatch's bulk patching,
> > > particularly if you moved the last group of patches into the PV
> > > directory.
> >
> > dev-zero?
> Nope. The stuff in ${PV} are the patches upstream has on their servers, while
> the others are mostly Gentoo-specific or something else. So I want them to be
> separated.
> And I don't see any reason to start renaming now.
Oh, I would've expected upstream patches to be in SRC_URI ...
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-python/pygtksourceview: metadata.xml pygtksourceview-2.0.0.ebuild Manifest ChangeLog
On 21:22 Wed 10 Oct , Remi Cardona (remi) wrote:
> 1.1 dev-python/pygtksourceview/pygtksourceview-2.0.0.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-python/pygtksourceview/pygtksourceview-2.0.0.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-python/pygtksourceview/pygtksourceview-2.0.0.ebuild?rev=1.1&content-type=text/plain
> pkg_postinst() {
> python_version
> python_mod_optimize /usr/$(get_libdir)/python${PYVER}/site-packages/
> }
>
> pkg_postrm() {
> python_version
> python_mod_optimize /usr/$(get_libdir)/python${PYVER}/site-packages/
> }
I think the python_mod_ functions take care of running python_version()
... also have you noticed that you're optimizing in postrm instead of
cleaning up?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in eclass: gnatbuild.eclass
On 18:17 Wed 10 Oct , George Shapovalov (george) wrote: > george 07/10/10 18:17:58 > > Modified: gnatbuild.eclass > > Log: fixed src_install issue, no longer relies on portage leaking > env vars between functions It's really sad that you have to add this workaround. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in media-video/mplayer: ChangeLog mplayer-1.0_rc2.ebuild
On 04:22 Wed 10 Oct , Steve Dibb (beandog) wrote:
> 1.1 media-video/mplayer/mplayer-1.0_rc2.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-video/mplayer/mplayer-1.0_rc2.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-video/mplayer/mplayer-1.0_rc2.ebuild?rev=1.1&content-type=text/plain
> if use encode; then
> use aac || myconf="${myconf} --disable-faac"
> use dv || myconf="${myconf} --disable-libdv"
> use x264 || myconf="${myconf} --disable-x264"
> else
> myconf="${myconf} --disable-mencoder --disable-libdv
> --disable-x264 \
> --disable-faac"
> fi
This could be:
if use encode; then
myconf="${myconf}
--enable-mencoder
$(use_enable aac faac)
$(use_enable dv libdv)
$(use_enable x264)"
fi
> use lirc || myconf="${myconf} --disable-lirc --disable-lircc"
> myconf="${myconf} $(use_enable joystick)"
> use ipv6 || myconf="${myconf} --disable-inet6"
> use rar || myconf="${myconf} --disable-unrarlib"
> use rtc || myconf="${myconf} --disable-rtc"
> use samba || myconf="${myconf} --disable-smb"
This could be more use_enable()'s. Does mplayer's build die if you pass
a redundant --enable-foo or something?
> # DVB / Video4Linux / Radio support
> if { use dvb || use v4l || use v4l2 || use radio; }; then
> use dvb || myconf="${myconf} --disable-dvb --disable-dvbhead"
> use v4l || myconf="${myconf} --disable-tv-v4l1"
> use v4l2 || myconf="${myconf} --disable-tv-v4l2"
> use teletext || myconf="${myconf} --disable-tv-teletext"
> use pvr || myconf="${myconf} --disable-pvr"
> if use radio && { use dvb || use v4l || use v4l2; }; then
> myconf="${myconf} --enable-radio $(use_enable encode
> radio-capture)"
> else
> myconf="${myconf} --disable-radio-v4l2
> --disable-radio-bsdbt848"
> fi
> else
> myconf="${myconf} --disable-tv --disable-tv-v4l1
> --disable-tv-v4l2 \
> --disable-radio --disable-radio-v4l2
> --disable-radio-bsdbt848 \
> --disable-dvb --disable-dvbhead --disable-tv-teletext \
> --disable-pvr"
> fi
Here's one other way to do this (leaving out myconf etc):
$(use_enable dvb)
$(use_enable dvb dvbhead)
$(use_enable v4l tv-v4l1)
$(use_enable v4l2 tv-v4l2)
$(use_enable teletext tv-teletext)
$(use_enable pvr)
if use dvb || use v4l || use v4l2; then
$(use_enable radio)
$(use_enable radio radio-bsdbt848)
$(use_enable radio radio-v4l2)
# etc...
fi
> use aac || myconf="${myconf} --disable-faad-internal"
> use amrnb || myconf="${myconf} --disable-libamr_nb"
> use amrwb || myconf="${myconf} --disable-libamr_wb"
> use dts || myconf="${myconf} --disable-libdca"
> ! use png && ! use gtk && myconf="${myconf} --disable-png"
> use lzo || myconf="${myconf} --disable-liblzo"
> use encode && use mp2 || myconf="${myconf} --disable-twolame \
> --disable-toolame"
> use mp3 || myconf="${myconf} --disable-mp3lib"
> use quicktime || myconf="${myconf} --disable-qtx"
> use vorbis || myconf="${myconf} --disable-libvorbis"
> use xanim && myconf="${myconf} --xanimcodecsdir=/usr/lib/xanim/mods"
Again, use_enable() would be cleaner in many cases.
> use aalib || myconf="${myconf} --disable-aa"
> use dga || myconf="${myconf} --disable-dga1 --disable-dga2"
> use fbcon || myconf="${myconf} --disable-fbdev"
> use fbcon && use video_cards_s3virge && myconf="${myconf} --enable-s3fb"
> use libcaca || myconf="${myconf} --disable-caca"
> use opengl || myconf="${myconf} --disable-gl"
> use video_cards_vesa || myconf="${myconf} --disable-vesa"
> use vidix || myconf="${myconf} --disable-vidix-internal \
> --disable-vidix-external"
> use zoran || myconf="${myconf} --disable-zr"
And again
> if use xv; then
> if use xvmc; then
> myconf="${myconf} --enable-xvmc --with-xvmclib=XvMCW"
> else
> myconf="${myconf} --disable-xvmc"
> fi
> else
> myconf="${myconf} --disable-xv --disable-xvmc"
> fi
Something along these lines might do:
$(use_enable xv)
$(use xv && $(use_enable xvmc))
$(use xv && $(use_with xvmc with-xvmclib XvMCW))
Would this kind of cleanup work?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in media-video/vlc: ChangeLog vlc-0.9.0_alpha20071009.ebuild
On 22:23 Tue 09 Oct , Alexis Ballier (aballier) wrote: > 1.1 media-video/vlc/vlc-0.9.0_alpha20071009.ebuild > > file : > http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-video/vlc/vlc-0.9.0_alpha20071009.ebuild?rev=1.1&view=markup > plain: > http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-video/vlc/vlc-0.9.0_alpha20071009.ebuild?rev=1.1&content-type=text/plain > IUSE="a52 3dfx debug altivec httpd vlm gnutls live v4l v4l2 cdda ogg matroska > dvb dvd vcd vcdx dts flac mpeg vorbis theora X opengl truetype svg fbcon svga > oss aalib ggi libcaca esd arts alsa wxwindows ncurses xosd lirc stream > mp3 xv bidi sdl sdl-image png xml samba daap mod speex shout rtsp > win32codecs skins hal avahi xinerama cddb directfb upnp nsplugin seamonkey > optimisememory libnotify jack musepack x264 dc1394 lua gnome pvr taglib > musicbrainz dbus libgcrypt id3tag cdio ffmpeg twolame" > > RDEPEND=" > ffmpeg? ( >=media-video/ffmpeg-0.4.9_p20050226-r1 ) > cdda? ( >=dev-libs/libcdio-0.72 > cddb? ( >=media-libs/libcddb-1.2.0 ) ) > live? ( >=media-plugins/live-2007.02.20 ) > dvd? ( media-libs/libdvdread > media-libs/libdvdcss > >=media-libs/libdvdnav-0.1.9 > media-libs/libdvdplay ) > esd? ( media-sound/esound ) > ogg? ( media-libs/libogg ) > matroska? ( > >=dev-libs/libebml-0.7.6 > >=media-libs/libmatroska-0.8.0 ) > mp3? ( media-libs/libmad ) > a52? ( >=media-libs/a52dec-0.7.4-r3 ) > dts? ( media-libs/libdca ) > flac? ( media-libs/libogg > >=media-libs/flac-1.1.2 ) > mpeg? ( >=media-libs/libmpeg2-0.3.2 ) > vorbis? ( media-libs/libvorbis ) > theora? ( media-libs/libtheora ) > truetype? ( media-libs/freetype > media-fonts/ttf-bitstream-vera ) [crop another 30 or so] Both the IUSE list and this dep list would be vastly easier to parse if they were ordered alphabetically. > if use nsplugin; then > if use seamonkey; then > XPIDL=/usr/lib/seamonkey > MOZILLA_CONFIG=/usr/lib/seamonkey/seamonkey-config > else > XPIDL=/usr/lib/mozilla-firefox > MOZILLA_CONFIG=/usr/lib/mozilla-firefox/firefox-config > fi > fi Should this be get_libdir() ? > > econf \ > $(use_enable altivec) \ > $(use_enable stream sout) \ > $(use_enable httpd) \ > $(use_enable gnutls) \ > $(use_enable v4l) \ > $(use_enable v4l2) \ > $(use_enable cdda) $(use_enable cdda cddax)\ > $(use_enable cddb libcddb) \ [crop another 30 or so] Another place where ordering would help. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-analyzer/tcpdump: ChangeLog tcpdump-3.9.8.ebuild
On 18:04 Tue 09 Oct , Markus Ullmann (jokey) wrote:
> 1.1 net-analyzer/tcpdump/tcpdump-3.9.8.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-analyzer/tcpdump/tcpdump-3.9.8.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-analyzer/tcpdump/tcpdump-3.9.8.ebuild?rev=1.1&content-type=text/plain
> src_compile() {
> # tcpdump needs some optymalization. see bug #108391
> ( ! is-flag -O? || is-flag -O0 ) && append-flags -O
You don't need a subshell here. If you feel that it improves
readability, use a code block instead.
> # Fix wrt bug #48747
> if [[ $(gcc-major-version) -gt 3 ]] || \
> [[ $(gcc-major-version) -eq 3 && $(gcc-minor-version) -ge 4 ]]
> then
> filter-flags -funit-at-a-time
> append-flags -fno-unit-at-a-time
> fi
This kind of stuff might be worth verifying whether it's still needed on
every bump. The last comment on the bug refers to gcc 4.0.*.
> make CCOPT="$CFLAGS" || die "make failed"
If emake doesn't work, please add a comment to that effect.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-libs/libpcap: ChangeLog libpcap-0.9.8.ebuild libpcap-0.9.4.ebuild
On 17:57 Tue 09 Oct , Markus Ullmann (jokey) wrote: > 1.1 net-libs/libpcap/libpcap-0.9.8.ebuild > > file : > http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-libs/libpcap/libpcap-0.9.8.ebuild?rev=1.1&view=markup > plain: > http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-libs/libpcap/libpcap-0.9.8.ebuild?rev=1.1&content-type=text/plain > DEPEND="!virtual/libpcap" > PROVIDE="virtual/libpcap" Does this still work properly with some of the new portage versions that consider installed packages? Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in media-plugins/vdr-radio: ChangeLog vdr-radio-0.2.4.ebuild
On 17:38 Tue 09 Oct , Matthias Schwarzott (zzam) wrote:
> 1.1 media-plugins/vdr-radio/vdr-radio-0.2.4.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-plugins/vdr-radio/vdr-radio-0.2.4.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-plugins/vdr-radio/vdr-radio-0.2.4.ebuild?rev=1.1&content-type=text/plain
> chown -R vdr:vdr "${D}"/var/cache/vdr-radio
You could use fowners here, since the rest of the ebuild uses the
portage functions.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in x11-wm/fluxbox: ChangeLog fluxbox-1.0.0.ebuild
On 14:45 Tue 09 Oct , Jim Ramsay (lack) wrote:
> 1.1 x11-wm/fluxbox/fluxbox-1.0.0.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/x11-wm/fluxbox/fluxbox-1.0.0.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/x11-wm/fluxbox/fluxbox-1.0.0.ebuild?rev=1.1&content-type=text/plain
> pkg_setup() {
> if use imlib ; then
> if ! built_with_use media-libs/imlib2 X ; then
> eerror "To build fluxbox with imlib in USE, you need an
> X enabled"
> eerror "media-libs/imlib2 . Either recompile imlib2
> with the X"
> eerror "USE flag turned on or disable the imlib USE
> flag for fluxbox."
> die "USE=imlib requires imlib2 with USE=X"
> fi
> fi
> }
There's only one case you're interested in here, so you could just
combine the two things into a single statement:
if use imlib && ! built_with_use media-libs/imlib2 X; then
> src_compile() {
> export PKG_CONFIG_PATH=/usr/X11R6/lib/pkgconfig:/usr/lib/pkgconfig
This shouldn't be necessary anymore.
> src_install() {
> dodir /usr/share/fluxbox
> make DESTDIR="${D}" install || die "make install failed"
If emake doesn't work, please add a comment to that effect.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in x11-base/xorg-x11: xorg-x11-7.2.ebuild xorg-x11-7.3.ebuild ChangeLog
On 13:50 Tue 09 Oct , Peter Volkov wrote:
> > XORGCONF="/etc/X11/xorg.conf"
> [...]
> > - sed -i "/RgbPath/d" "${D}"${XORGCONF}
> > + sed -i "/RgbPath/d" "${D}${XORGCONF}"
>
> Sorry, but why? What is the difference? Seems $XORGCONF does not contain
> any spaces inside and both forms are acceptable.
Oh, that's a leftover from when I was playing around with putting
${ROOT} in ${XORGCONF}, but it turned out to be the wrong way to go. I
do kinda like quoting the whole path, though. Haven't decided which way
I like better yet.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-fs/evms: ChangeLog evms-2.5.5-r8.ebuild
On 22:01 Mon 08 Oct , Doug Goldstein (cardoe) wrote:
> 1.1 sys-fs/evms/evms-2.5.5-r8.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-fs/evms/evms-2.5.5-r8.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-fs/evms/evms-2.5.5-r8.ebuild?rev=1.1&content-type=text/plain
> epatch "${FILESDIR}/${PV}/md_super_fix.patch"
> epatch "${FILESDIR}/${PV}/ntfs_unmkfs.patch"
> epatch "${FILESDIR}/${PV}/raid5_degrade_fix_v2.patch"
> epatch "${FILESDIR}/${PV}/raid5_remove_spare_fix.patch"
> epatch "${FILESDIR}/${PV}/raid5_remove_spare_fix_2.patch"
> epatch "${FILESDIR}/${PV}/raid5_algorithm.patch"
> epatch "${FILESDIR}/${PV}/cli_reload_options.patch"
> epatch "${FILESDIR}/${PV}/cli_query_segfault.patch"
> epatch "${FILESDIR}/${PV}/get_geometry.patch"
> epatch "${FILESDIR}/${PV}/BaseName.patch"
> epatch "${FILESDIR}/${PV}/disk_cache.patch"
>
> epatch "${FILESDIR}/${P}-as-needed.patch"
> epatch "${FILESDIR}/${P}-glib_dep.patch"
> epatch "${FILESDIR}/${P}-ocfs2.patch"
> epatch "${FILESDIR}/${P}-use_disk_group.patch"
> epatch "${FILESDIR}/${P}-pagesize.patch"
This would be another good candidate for using epatch's bulk patching,
particularly if you moved the last group of patches into the PV
directory.
> mv -f ${D}/$(get_libdir)/*.a "${D}/usr/$(get_libdir)"
> mv -f ${D}/sbin/evmsgui ${D}/usr/sbin
Quoting
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-emacs/lua-mode: ChangeLog lua-mode-20070708.ebuild
On 15:22 Mon 08 Oct , Christian Faulhammer (opfer) wrote:
> 1.1 app-emacs/lua-mode/lua-mode-20070708.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-emacs/lua-mode/lua-mode-20070708.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-emacs/lua-mode/lua-mode-20070708.ebuild?rev=1.1&content-type=text/plain
> src_compile() {
> /usr/bin/emacs -batch --no-site-file --no-init-file \
> --load "${S}/lua-mode.el" -f batch-byte-compile lua-mode.el
> }
Shouldn't this be using the new eclass functions?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in media-libs/win32codecs: ChangeLog win32codecs-20071007.ebuild
On 11:13 Mon 08 Oct , Steve Dibb (beandog) wrote:
> 1.1 media-libs/win32codecs/win32codecs-20071007.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-libs/win32codecs/win32codecs-20071007.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-libs/win32codecs/win32codecs-20071007.ebuild?rev=1.1&content-type=text/plain
> cd ${S}
A polite reminder that the quoting patch is in released portage.
> QA_TEXTRELS="usr/$(get_libdir)/real/drvc.so
> usr/$(get_libdir)/real/drv[34].so.6.0
> usr/$(get_libdir)/win32/vid_*.xa"
> dodir /usr/$(get_libdir)/win32
> dodir /usr/$(get_libdir)/real
> insinto /usr/$(get_libdir)/real
> ln -s "${D}/usr/$(get_libdir)/real/*"
> "${D}/usr/$(get_libdir)/win32/"
> insinto /usr/$(get_libdir)/win32
> SEARCH_DIRS_MASK="/usr/$(get_libdir)/real /usr/$(get_libdir)/win32"
With this many calls to get_libdir(), it might make sense to just call
it once and save it in a global variable.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in games-kids/gcompris: ChangeLog gcompris-8.4.ebuild
On 00:46 Mon 08 Oct , Michael Sterrett (mr_bones_) wrote:
> 1.1 games-kids/gcompris/gcompris-8.4.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/games-kids/gcompris/gcompris-8.4.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/games-kids/gcompris/gcompris-8.4.ebuild?rev=1.1&content-type=text/plain
> src_unpack() {
> unpack ${A}
> cd "${S}"
> epatch "${FILESDIR}"/${P}-build.patch
> cp /usr/share/gettext/config.rpath .
Shouldn't this respect ROOT != / ? I can see how that would be a bit of
an unusual use case for games, though.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-games/crystalspace: ChangeLog crystalspace-1.2.ebuild
On 21:51 Sun 07 Oct , Michael Sterrett (mr_bones_) wrote:
> 1.1 dev-games/crystalspace/crystalspace-1.2.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-games/crystalspace/crystalspace-1.2.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-games/crystalspace/crystalspace-1.2.ebuild?rev=1.1&content-type=text/plain
> src_compile() {
> if useq wxwindows; then
> WX_GTK_VER=2.6
> need-wxwidgets gtk2
> if [[ ! -e /usr/bin/wx-config ]]; then
> sed -i -e "s:wx-config:${WX_CONFIG_NAME}:g" configure
Is this correct, to ignore WX_CONFIG_NAME in this case? It looks a
little odd.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-lang/dmd-bin: ChangeLog dmd-bin-1.022.ebuild dmd-bin-1.020.ebuild
On 21:47 Sun 07 Oct , Anant Narayanan (anant) wrote:
> 1.1 dev-lang/dmd-bin/dmd-bin-1.022.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-lang/dmd-bin/dmd-bin-1.022.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-lang/dmd-bin/dmd-bin-1.022.ebuild?rev=1.1&content-type=text/plain
> # Remove unneccessary files
> mv ${S}/dmd/lib/libphobos.a ${S}/dmd
> rm -r ${S}/dmd/lib
> mkdir ${S}/dmd/lib
> mv ${S}/dmd/libphobos.a ${S}/dmd/lib
>
> rm -r ${S}/dm
> rm ${S}/dmd/bin/*.dll ${S}/dmd/bin/*.exe ${S}/dmd/bin/readme.txt
> rm ${S}/dmd/bin/sc.ini ${S}/dmd/bin/windbg.hlp
>
> # Cleanup line endings
> cd ${S}/dmd
> src_compile() {
> cd ${S}/dmd/src/phobos
> sed -i -e "s:DMD=.*:DMD=${S}/dmd/bin/dmd -I${S}/dmd/src/phobos
> -L${S}/dmd/lib/libphobos.a:" linux.mak internal/gc/linux.mak
> edos2unix linux.mak internal/gc/linux.mak
> make -f linux.mak
> cp libphobos.a ${S}/dmd/lib
>
> # Clean up
> make -f linux.mak clean
> rm internal/gc/*.o
> }
>
> src_install() {
> cd ${S}/dmd
Quote variables that can have spaces in them: D, S, T, WORKDIR,
FILESDIR, DESTDIR, ROOT. The latest portage release now contains a
repoman supporting this, so that's no longer an excuse.
If emake doesn't work, please add a comment to that effect.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in media-plugins/vdr-extrecmenu: ChangeLog vdr-extrecmenu-1.0.ebuild
On 20:23 Sun 07 Oct , Joerg Bornkessel (hd_brummy) wrote:
> 1.1 media-plugins/vdr-extrecmenu/vdr-extrecmenu-1.0.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-plugins/vdr-extrecmenu/vdr-extrecmenu-1.0.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-plugins/vdr-extrecmenu/vdr-extrecmenu-1.0.ebuild?rev=1.1&content-type=text/plain
> src_unpack() {
> vdr-plugin_src_unpack
>
> if grep -q fskProtection /usr/include/vdr/timers.h; then
> sed -i "s:#WITHPINPLUGIN:WITHPINPLUGIN:" Makefile
This doesn't respect ROOT != / and it's also dependent on the build
system.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-libs/ace: ChangeLog ace-5.6.1.ebuild
On 19:52 Sun 07 Oct , Daniel Black (dragonheart) wrote:
> 1.1 dev-libs/ace/ace-5.6.1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-libs/ace/ace-5.6.1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-libs/ace/ace-5.6.1.ebuild?rev=1.1&content-type=text/plain
> pkg_postinst() {
> # This is required, as anything trying to compile against ACE will have
> # problems with conflicting OS.h files if this is not done.
>
> local CC_MACHINE=$($(tc-getCC) -dumpmachine)
> if [ -d "/usr/lib/gcc-lib/${CC_MACHINE}/$(gcc-fullversion)/include/ace"
> ]; then
> mv
> "/usr/lib/gcc-lib/${CC_MACHINE}/$(gcc-fullversion)/include/ace" \
>
> "/usr/lib/gcc-lib/${CC_MACHINE}/$(gcc-fullversion)/include/ace.old"
Does gcc not use get_libdir() ? I also find it a little unusual that
you're not informing people when you're renaming directories on their
live systems.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in media-sound/alsaplayer: ChangeLog alsaplayer-0.99.80_rc3.ebuild alsaplayer-0.99.77-r1.ebuild alsaplayer-0.99.80_rc1.ebuild
On 05:12 Sun 07 Oct , Samuli Suominen (drac) wrote:
> 1.1 media-sound/alsaplayer/alsaplayer-0.99.80_rc3.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-sound/alsaplayer/alsaplayer-0.99.80_rc3.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/media-sound/alsaplayer/alsaplayer-0.99.80_rc3.ebuild?rev=1.1&content-type=text/plain
> if ! use alsa && ! use oss && ! use esd && ! use jack && ! use nas && !
> use sparc; then
> ewarn "You've disabled alsa, oss, esd, jack and nas. Enabling
> oss for you."
> myconf="${myconf} --enable-oss"
> fi
How come oss is the default instead of alsa?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-lang/swi-prolog: ChangeLog swi-prolog-5.6.43.ebuild
On 23:57 Sat 06 Oct , Keri Harris (keri) wrote:
> 1.1 dev-lang/swi-prolog/swi-prolog-5.6.43.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-lang/swi-prolog/swi-prolog-5.6.43.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-lang/swi-prolog/swi-prolog-5.6.43.ebuild?rev=1.1&content-type=text/plain
> epatch "${WORKDIR}"/${PV}/1000-cflags.patch
> epatch "${WORKDIR}"/${PV}/1001-multilib.patch
> epatch "${WORKDIR}"/${PV}/1002-parallel-make.patch
> epatch "${WORKDIR}"/${PV}/2100-thread-sandbox.patch
> epatch "${WORKDIR}"/${PV}/2105-pl-profile-va-debug.patch
> epatch "${WORKDIR}"/${PV}/2800-mandir.patch
> epatch "${WORKDIR}"/${PV}/2900-thread-library.patch
> epatch "${WORKDIR}"/${PV}/3029-clib-test.patch
> epatch "${WORKDIR}"/${PV}/3059-ssl-test.patch
> epatch "${WORKDIR}"/${PV}/3069-semweb-test.patch
> epatch "${WORKDIR}"/${PV}/3150-jpl-env.patch
> epatch "${WORKDIR}"/${PV}/3151-jpl-ppc-arch.patch
> epatch "${WORKDIR}"/${PV}/3152-jpl-junit.patch
> epatch "${WORKDIR}"/${PV}/3159-jpl-test.patch
> epatch "${WORKDIR}"/${PV}/3160-xpce-parallel-build.patch
> epatch "${WORKDIR}"/${PV}/3139-sgml-test.patch
> epatch "${WORKDIR}"/${PV}/3149-sgml-RDF-test.patch
> epatch "${WORKDIR}"/${PV}/3169-xpce-test.patch
> epatch "${WORKDIR}"/${PV}/3199-ltx2htm.patch
> epatch "${WORKDIR}"/${PV}/3229-zlib-test.patch
You might want to just use epatch's bulk patching capability. Pass in
the directory containing all the patches instead of specifying each one
individually. Set EPATCH_FORCE="yes" if you don't want to switch the
patch naming scheme.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in gnome-base/gnome-panel: ChangeLog gnome-panel-2.20.0.1.ebuild
On 21:28 Sat 06 Oct , Saleem Abdulrasool (compnerd) wrote:
> 1.1 gnome-base/gnome-panel/gnome-panel-2.20.0.1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/gnome-base/gnome-panel/gnome-panel-2.20.0.1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/gnome-base/gnome-panel/gnome-panel-2.20.0.1.ebuild?rev=1.1&content-type=text/plain
> pkg_postinst() {
> local entries="/etc/gconf/schemas/panel-default-setup.entries"
>
> if [ -e "$entries" ]; then
> einfo "setting panel gconf defaults..."
> GCONF_CONFIG_SOURCE=`"${ROOT}usr/bin/gconftool-2"
> --get-default-source`
> "${ROOT}usr/bin/gconftool-2" --direct --config-source \
> "${GCONF_CONFIG_SOURCE}" --load="${entries}"
This bit is safe for ROOT != /, but look at the check made for whether to run
it.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-lang/R: ChangeLog R-2.6.0.ebuild R-2.4.1.ebuild R-2.5.0.ebuild
On 14:21 Sat 06 Oct , Markus Dittrich (markusle) wrote:
> 1.1 dev-lang/R/R-2.6.0.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-lang/R/R-2.6.0.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-lang/R/R-2.6.0.ebuild?rev=1.1&content-type=text/plain
> pkg_setup() {
> # Test for a 64 bit architecture - f2c won't work on 64 bit archs with
> R.
> # Thanks to vapier for providing the test.
> cd "${T}"
> echo 'int main(){}' > test.c
> $(tc-getCC) -c test.c -o test.o
> if file test.o | grep -qs 64-bit ; then
> einfo "64 bit architecture detected, using g77 or gfortran."
... or ifc?
> FORTRAN="gfortran g77 ifc"
> else
> FORTRAN="gfortran g77 f2c"
How come ifc isn't down here, too?
> # R-2.5.0 started to have issues with parallel make
> emake -j1 || die "emake failed"
Got an upstream bug filed?
> make \
If emake doesn't work, please add a comment to that effect.
> prefix=${D}/usr \
> mandir=${D}/usr/share/man \
> infodir=${D}/usr/share/info \
> libdir=${D}/usr/$(get_libdir) \
> rhome=${D}/usr/$(get_libdir)/R \
> -i ${D}/usr/$(get_libdir)/R/bin/R \
> -i ${D}/usr/$(get_libdir)/R/bin/R \
> sed -e "s:${D}::" -i ${D}/usr/$(get_libdir)/pkgconfig/libR.pc \
> cd ${D}/usr/bin/
> > ${D}/etc/env.d/99R
> cd ${S}
Quote variables that can have spaces in them: D, S, T, WORKDIR,
FILESDIR, DESTDIR, ROOT.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-libs/linux-fusion: ChangeLog linux-fusion-7.0.1.ebuild
On 13:15 Sat 06 Oct , Mike Frysinger (vapier) wrote:
> 1.1 dev-libs/linux-fusion/linux-fusion-7.0.1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-libs/linux-fusion/linux-fusion-7.0.1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-libs/linux-fusion/linux-fusion-7.0.1.ebuild?rev=1.1&content-type=text/plain
> pkg_setup() {
> linux-mod_pkg_setup
> }
The eclass exports pkg_setup()..
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-im/skype: ChangeLog skype-1.4.0.118.ebuild
On 11:32 Sat 06 Oct , Gustavo Felisberto (humpback) wrote:
> 1.1 net-im/skype/skype-1.4.0.118.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-im/skype/skype-1.4.0.118.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-im/skype/skype-1.4.0.118.ebuild?rev=1.1&content-type=text/plain
> SFILENAME=${PN}_static-${PV}.tar.bz2
> DFILENAME=${P}.tar.bz2
> SRC_URI="!qt-static? ( http://download.skype.com/linux/${DFILENAME} )
> qt-static? ( http://download.skype.com/linux/${SFILENAME}
>
> http://felisberto.net/~humpback/${PN}-lang-${PV}.tar.bz2 )
> amd64? (
> http://felisberto.net/~humpback/libsigc++20-2.0.17-1-from-fc5.rf.i386.tar.gz
> )"
> src_unpack() {
> if use qt-static ; then
> unpack $SFILENAME
> cd $S
> unpack ${PN}-lang-${PV}.tar.bz2
> else
> unpack $DFILENAME
> fi
> }
Does the default 'unpack ${A}' not do the right thing for some reason?
The only possibility I see is that you don't like where the language
tarball unpacks, but that appears to be something you created anyhow.
Quote variables that can have spaces in them: D, S, T, WORKDIR,
FILESDIR, DESTDIR, ROOT.
> newins ${FILESDIR}/skype.debus.config skype.conf
> newins ${S}/icons/SkypeBlue_${X}x${X}.png ${PN}.png
More quoting.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in net-proxy/ziproxy: ChangeLog ziproxy-2.3.0.ebuild
On 23:08 Fri 05 Oct , Stefan Briesenick (sbriesen) wrote:
> 1.1 net-proxy/ziproxy/ziproxy-2.3.0.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-proxy/ziproxy/ziproxy-2.3.0.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/net-proxy/ziproxy/ziproxy-2.3.0.ebuild?rev=1.1&content-type=text/plain
> src_install() {
> make DESTDIR="${D}" install || die "make install failed"
If emake doesn't work, please add a comment to that effect.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in eclass: gnustep-base.eclass
On 17:31 Fri 05 Oct , Ryan Hill wrote:
> If there aren't any variables that you actually need expanded in the
> script (i didn't see any but could have easily missed it), just escape
> the termination marker, ie.
>
> cat << \EOF > "${T}"/${cfile}
>
> or
> cat << 'EOF' > "${T}"/${cfile}
>
> This turns off parameter and arithmetic expansion and command
> substitution.
That is so cool. Just another reason I love this reviewing.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: Re: Re: [gentoo-commits] gentoo-x86 commit in x11-libs/qt: ChangeLog qt-4.3.2.ebuild
On 00:41 Sat 06 Oct , Steve Long wrote:
> Donnie Berkholz wrote:
> > On 13:34 Fri 05 Oct , Steve Long wrote:
> >> In actual fact, I'd be more likely to use parameter expansion than set,
> >> eg: spec=${CHOST#*-*-} # chop first two fields off so spec is fields 3 on
> >> spec=${spec%%-*} # chop all but first off so left with just field 3
> >> ..which I believe works in sh[1] as well. The point for me, however, is
> >> not whether sh can be kludged to do something, it's what the most
> >> efficient ways to do something in scripts are. Whether you use pe, read
> >> or an array, avoiding externals leads to quicker scripts.[2]
> >
> > That parameter workaround wouldn't work, because CHOST may also contain
> > just 3 fields, as it does on Gentoo/BSD. Unpack a portage tarball and
> > look at the make.conf's if you're curious.
> >
> Then the second expansion would have no effect.
Good point.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] new-style virtual/editor
On 20:42 Fri 05 Oct , Christian Faulhammer wrote: > about 26 ebuilds have a PROVIDE=virtual/editor. Those could be > transformed to a new-style virtual, which is really simple. According > to zmedico and genone the impact of just commiting the virtual would > be low. But I'd like to hear some comments on it. If noone objects I > will commit it next week (Monday probably) and remove all PROVIDE > lines. Eventually I will check profiles, too, and file bugs when > unsure what the intended behaviour? Or anyone objections about me > touching his profiles. How many packages depend on virtual/editor? Should it be a virtual at all? Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in eclass: gnustep-base.eclass
On 10:51 Fri 05 Oct , Bernard Cafarelli (voyageur) wrote:
> 1.4 eclass/gnustep-base.eclass
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/eclass/gnustep-base.eclass?rev=1.4&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/eclass/gnustep-base.eclass?rev=1.4&content-type=text/plain
> diff :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/eclass/gnustep-base.eclass?r1=1.3&r2=1.4
> @@ -166,8 +166,43 @@
>
> local cfile=config-${PN}.sh
>
> - echo '#!/usr/bin/env bash' > "${T}"/${cfile}
> - echo "echo Applying ${P} default configuration ..." >> "${T}"/${cfile}
> + cat << EOF > "${T}"/${cfile}
> +#!/usr/bin/env bash
> +gnustep_append_default() {
> + if [[ -z \$1 || -z \$2 || -z \$3 ]]; then
> + echo "warning: invalid script invocation"
> + return
> + fi
> + dom=\$1
> + key=\$2
> + val=\$3
> + cur=\$(defaults read \${dom} \${key}) 2> /dev/null
> + if [[ -z \$cur ]] ; then
> + echo " * setting \${dom} \${key}"
> + defaults write \${dom} \${key} "( \${val} )"
> + elif [[ \${cur} != *\${val}* ]] ; then
> + echo " * adding \${val} to \${dom} \${key}"
> + echo "\${cur%)\'}, \"\${val}\" )'" | defaults write
> + else
> + echo " * \${val} already present in \${dom} \${key}"
> + fi
> +}
> +
> +gnustep_set_default() {
> + if [[ -z \$1 || -z \$2 || -z \$3 ]]; then
> + echo "warning: invalid script invocation"
> + return
> + fi
> + dom=\$1
> + key=\$2
> + val=\$3
> + echo " * setting \${dom} \${key}"
> + defaults write \${dom} \${key} \${val}
> +}
> +
> +echo "Applying ${P} default configuration ..."
> +EOF
> +
There's gotta be a better way of doing this. All those escapes really
start to obfuscate the code. Anyone got a better idea?
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-cluster/pvfs2: ChangeLog pvfs2-2.6.3.ebuild
On 14:03 Fri 05 Oct , Matti Bickel (mabi) wrote: > 1.1 sys-cluster/pvfs2/pvfs2-2.6.3.ebuild > > file : > http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-cluster/pvfs2/pvfs2-2.6.3.ebuild?rev=1.1&view=markup > plain: > http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-cluster/pvfs2/pvfs2-2.6.3.ebuild?rev=1.1&content-type=text/plain > IUSE="gtk static doc" No server vs client USE flag? It seems a bit odd to build support for both server and client of a cluster filesystem, since almost all systems will never play both roles. Thanks, Donnie -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sys-cluster/drbd-kernel: drbd-kernel-0.7.24.ebuild drbd-kernel-8.0.6.ebuild metadata.xml ChangeLog Manifest
On 14:01 Fri 05 Oct , Christian Zoffoli (xmerlin) wrote:
> 1.1 sys-cluster/drbd-kernel/drbd-kernel-0.7.24.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-cluster/drbd-kernel/drbd-kernel-0.7.24.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-cluster/drbd-kernel/drbd-kernel-0.7.24.ebuild?rev=1.1&content-type=text/plain
> cd ${S}
> epatch ${FILESDIR}/${MY_PN}-0.7.22-nodevfs.patch || die
> epatch
> ${FILESDIR}/${MY_PN}-0.7.22-scripts.adjust_drbd_config_h.sh.patch || die
Quote variables that can have spaces in them: D, S, T, WORKDIR,
FILESDIR, DESTDIR, ROOT.
> 1.1 sys-cluster/drbd-kernel/drbd-kernel-8.0.6.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-cluster/drbd-kernel/drbd-kernel-8.0.6.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/sys-cluster/drbd-kernel/drbd-kernel-8.0.6.ebuild?rev=1.1&content-type=text/plain
> cd ${S}
> epatch ${FILESDIR}/${PN}-8.0.0-scripts.adjust_drbd_config_h.sh.patch ||
> die
Quote variables that can have spaces in them: D, S, T, WORKDIR,
FILESDIR, DESTDIR, ROOT.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-admin/sysstat: ChangeLog sysstat-8.0.1.ebuild
On 13:52 Fri 05 Oct , Jeroen Roovers (jer) wrote:
> 1.1 app-admin/sysstat/sysstat-8.0.1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-admin/sysstat/sysstat-8.0.1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-admin/sysstat/sysstat-8.0.1.ebuild?rev=1.1&content-type=text/plain
> emake DESTDIR=${D} DOC_DIR=/usr/share/doc/${PF} \
> install || die "make install failed"
Quote variables that can have spaces in them: D, S, T, WORKDIR,
FILESDIR, DESTDIR, ROOT.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Re: Re: [gentoo-commits] gentoo-x86 commit in x11-libs/qt: ChangeLog qt-4.3.2.ebuild
On 13:34 Fri 05 Oct , Steve Long wrote:
> Roy Marples wrote:
> > IFS=-
> > set -- ${CHOST}
> > spec=$2
> >
> > Works fine in bash - and other shells.
> >
> Yeah fine, there are kludgy workarounds; so what? Doesn't mean I want to use
> them. ;)
> In actual fact, I'd be more likely to use parameter expansion than set, eg:
> spec=${CHOST#*-*-} # chop first two fields off so spec is fields 3 on
> spec=${spec%%-*} # chop all but first off so left with just field 3
> ..which I believe works in sh[1] as well. The point for me, however, is not
> whether sh can be kludged to do something, it's what the most efficient
> ways to do something in scripts are. Whether you use pe, read or an array,
> avoiding externals leads to quicker scripts.[2]
That parameter workaround wouldn't work, because CHOST may also contain
just 3 fields, as it does on Gentoo/BSD. Unpack a portage tarball and
look at the make.conf's if you're curious.
Avoiding externals at the cost of added complexity on something that
only gets called one time certainly qualifies as premature optimization.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in app-portage/eix: ChangeLog eix-0.10.1.ebuild
On 18:24 Thu 04 Oct , Stefan Schweizer (genstef) wrote:
> 1.1 app-portage/eix/eix-0.10.1.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-portage/eix/eix-0.10.1.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/app-portage/eix/eix-0.10.1.ebuild?rev=1.1&content-type=text/plain
> src_compile() {
> econf --with-bzip2 $(use_with sqlite) || die "econf failed"
> emake || die "emake failed"
> src/eix --dump-defaults >eixrc || die "generating eixrc failed"
> }
Running the built binary in src_* isn't cross-compile safe.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-java/sun-jre-bin: ChangeLog sun-jre-bin-1.4.2.16.ebuild
On 14:26 Thu 04 Oct , Petteri Raty (betelgeuse) wrote:
> 1.1 dev-java/sun-jre-bin/sun-jre-bin-1.4.2.16.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-java/sun-jre-bin/sun-jre-bin-1.4.2.16.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-java/sun-jre-bin/sun-jre-bin-1.4.2.16.ebuild?rev=1.1&content-type=text/plain
> if [ ! -r ${DISTDIR}/${At} ]; then
> tail -n +${startAt} ${DISTDIR}/${At} > install.sfx
> if [ -f ${S}/lib/unpack ]; then
> UNPACK_CMD=${S}/lib/unpack
> chmod +x $UNPACK_CMD
> sed -i 's#/tmp/unpack.log#/dev/null\x00\x00\x00\x00\x00\x00#g'
> $UNPACK_CMD
> PACK_FILE=${S}/`dirname $i`/`basename $i .jar`.pack
> if [ -f ${PACK_FILE} ]; then
> $UNPACK_CMD ${PACK_FILE} ${S}/$i
> rm -f ${PACK_FILE}
> touch ${D}/opt/${P}/.systemPrefs/.system.lock
> chmod 644 ${D}/opt/${P}/.systemPrefs/.system.lock
> touch ${D}/opt/${P}/.systemPrefs/.systemRootModFile
> chmod 644 ${D}/opt/${P}/.systemPrefs/.systemRootModFile
> ${D}/opt/${P}/plugin/desktop/sun_java.desktop > \
> ${T}/sun_java-jre.desktop
> domenu ${T}/sun_java-jre.desktop
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-java/sun-jdk: ChangeLog sun-jdk-1.4.2.16.ebuild
On 14:25 Thu 04 Oct , Petteri Raty (betelgeuse) wrote:
> 1.1 dev-java/sun-jdk/sun-jdk-1.4.2.16.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-java/sun-jdk/sun-jdk-1.4.2.16.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-java/sun-jdk/sun-jdk-1.4.2.16.ebuild?rev=1.1&content-type=text/plain
> if [ ! -r ${DISTDIR}/${At} ]; then
> if [ ! -r ${DISTDIR}/${jce_policy} ]; then
> tail -n +${startAt} ${DISTDIR}/${At} > install.sfx
> if [ -f ${S}/lib/unpack ]; then
> UNPACK_CMD=${S}/lib/unpack
> PACK_FILE=${S}/`dirname $i`/`basename $i .jar`.pack
> if [ -f ${PACK_FILE} ]; then
> echo " unpacking: $i"
> $UNPACK_CMD ${PACK_FILE} ${S}/$i
> rm -f ${PACK_FILE}
> cp -pPR demo ${D}/opt/${P}/ || die
> cd ${D}/opt/${P}/jre/lib/security
> unzip ${DISTDIR}/${jce_policy}
> mv ${D}/opt/${P}/jre/lib/security/US_export_policy.jar
> ${D}/opt/${P}/jre/lib/security/strong-jce
> mv ${D}/opt/${P}/jre/lib/security/local_policy.jar
> ${D}/opt/${P}/jre/lib/security/strong-jce
> touch ${D}/opt/${P}/.systemPrefs/.system.lock
> chmod 644 ${D}/opt/${P}/.systemPrefs/.system.lock
> touch ${D}/opt/${P}/.systemPrefs/.systemRootModFile
> chmod 644 ${D}/opt/${P}/.systemPrefs/.systemRootModFile
> ${D}/opt/${P}/jre/plugin/desktop/sun_java.desktop > \
> ${T}/sun_java-${SLOT}.desktop
> domenu ${T}/sun_java-${SLOT}.desktop
Quote variables that can have spaces in them: D, S, T, WORKDIR,
FILESDIR, DESTDIR, ROOT.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in dev-java/sun-jre-bin: ChangeLog sun-jre-bin-1.6.0.03.ebuild sun-jre-bin-1.5.0.13.ebuild
On 13:19 Thu 04 Oct , Petteri Raty (betelgeuse) wrote:
> 1.1 dev-java/sun-jre-bin/sun-jre-bin-1.6.0.03.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-java/sun-jre-bin/sun-jre-bin-1.6.0.03.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-java/sun-jre-bin/sun-jre-bin-1.6.0.03.ebuild?rev=1.1&content-type=text/plain
> # install control panel for Gnome/KDE
> if [[ -e "${D}/opt/${P}/plugin/desktop/sun_java.desktop" ]]; then
> sed -e "s/\(Name=Java\)/\1 Control Panel for Sun JRE ${SLOT}/" \
> -e "s#Exec=.*#Exec=/opt/${P}/bin/ControlPanel#" \
> -e
> "s#Icon=.*#Icon=/opt/${P}/plugin/desktop/sun_java.png#" \
> ${D}/opt/${P}/plugin/desktop/sun_java.desktop > \
> ${T}/sun_jre-${SLOT}.desktop || die
> domenu ${T}/sun_jre-${SLOT}.desktop || die
Quote variables that can have spaces in them: D, S, T, WORKDIR,
FILESDIR, DESTDIR, ROOT.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in eclass: java-utils-2.eclass java-virtuals-2.eclass
On 09:59 Thu 04 Oct , Alistair Bush (ali_bush) wrote:
> 1.1 eclass/java-virtuals-2.eclass
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/eclass/java-virtuals-2.eclass?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/eclass/java-virtuals-2.eclass?rev=1.1&content-type=text/plain
> dodir "${JAVA_PKG_VIRTUALS_PATH}"
> if [[ -n "${JAVA_VIRTUAL_PROVIDES}" ]]; then
> echo "PROVIDERS=\"${JAVA_VIRTUAL_PROVIDES}\"" \
> >> ${JAVA_PKG_VIRTUAL_PROVIDER}
> fi
>
> if [[ -n "${JAVA_VIRTUAL_VM}" ]]; then
> echo "VM=\"${JAVA_VIRTUAL_VM}\"" \
> >> ${JAVA_PKG_VIRTUAL_PROVIDER}
> fi
>
> if [[ -n "${JAVA_VIRTUAL_VM_CLASSPATH}" ]]; then
> echo "VM_CLASSPATH=\"${JAVA_VIRTUAL_VM_CLASSPATH}\"" \
> >> ${JAVA_PKG_VIRTUAL_PROVIDER}
> fi
You can use a neat trick with a code block here, along these lines:
{
echo foo
echo bar
echo blah
} >> file
It avoids the possibility of a typo when you're retyping the same
filename multiple times, plus it's just cool.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
