[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in sci-visualization/paraview: paraview-2.6.2.ebuild ChangeLog paraview-3.2.1.ebuild
On 19:59 Fri 14 Dec , Markus Dittrich (markusle) wrote:
> Index: paraview-3.2.1.ebuild
> ===
> # Copyright 1999-2007 Gentoo Foundation
> # Distributed under the terms of the GNU General Public License v2
> # $Header:
> /var/cvsroot/gentoo-x86/sci-visualization/paraview/paraview-3.2.1.ebuild,v
> 1.1 2007/12/14 19:59:46 markusle Exp $
>
> inherit distutils eutils flag-o-matic toolchain-funcs versionator python
I'm surprised to see this not using cmake-utils.eclass.
> src_compile() {
> cd "${BUILDDIR}"
> local CMAKE_VARIABLES=""
> CMAKE_VARIABLES="${CMAKE_VARIABLES}
> -DPV_INSTALL_LIB_DIR:PATH=/${PVLIBDIR}"
> CMAKE_VARIABLES="${CMAKE_VARIABLES} -DCMAKE_SKIP_RPATH:BOOL=YES"
> CMAKE_VARIABLES="${CMAKE_VARIABLES} -DVTK_USE_RPATH:BOOL=OFF"
> CMAKE_VARIABLES="${CMAKE_VARIABLES} -DCMAKE_INSTALL_PREFIX:PATH=/usr"
> CMAKE_VARIABLES="${CMAKE_VARIABLES} -DBUILD_SHARED_LIBS:BOOL=ON"
...
Since this seems to go beyond the standard variable naming for cmake,
you could set up custom functions based on cmake-utils _use_me_now() for
e.g. VTK_USE, PARAVIEW_USE, CMAKE_USE. Also seems like this would be a
lot easier to read if it were a multiline variable...
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in media-video/undvd: undvd-0.3.0.ebuild metadata.xml ChangeLog Manifest
On 17:46 Fri 14 Dec , Greg Kroah-Hartman (gregkh) wrote:
> pkg_setup() {
> einfo "Checking mplayer for USE flags we need..."
> for f in "encode dvd x264 mp3"; do
> if ! built_with_use media-video/mplayer $f; then
> eerror "$f"
> die "mplayer merged without $f USE flag"
> fi
> done
Does this actually work as you want? I'd be surprised. Quoting the whole
thing should treat it as a single flag with spaces and only run through
the loop once.
It should only coincidentally work because built_with_use() takes any
number of flags and loops through them, but I doubt it prints the single
missing flag as you intend.
[EMAIL PROTECTED] $ for f in "a b c d"; do echo $f; done
a b c d
[EMAIL PROTECTED] $ for f in a b c d; do echo $f; done
a
b
c
d
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Re: [gentoo-commits] gentoo-x86 commit in www-servers/resin: ChangeLog resin-3.0.25.ebuild resin-3.1.4.ebuild
On 09:42 Fri 14 Dec , Krzysiek Pawlik (nelchael) wrote:
> 1.1 www-servers/resin/resin-3.1.4.ebuild
>
> file :
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/www-servers/resin/resin-3.1.4.ebuild?rev=1.1&view=markup
> plain:
> http://sources.gentoo.org/viewcvs.py/gentoo-x86/www-servers/resin/resin-3.1.4.ebuild?rev=1.1&content-type=text/plain
> src_unpack() {
>
> unpack ${A}
> for i in "${WORKDIR}"/${PV}/resin-${PV}-*; do
> epatch "${i}"
> done;
How about:
epatch "${WORKDIR}"/${PV}/resin-${PV}-*
> java-ant_bsfix_one "${S}/build.xml"
>
> sed -i -e 's/256m/384m/' "${S}/build.xml"
>
> }
> src_compile() {
>
> append-flags -fPIC -DPIC
Somehow it seems wrong to append this for a whole package...
> chmod 755 "${S}/configure"
fperms means you don't need S or quotes.
> src_install() {
>
> make DESTDIR="${D}" install || die
Does emake work? If not, please add a comment to that effect.
> einfo "Fixing permissions..."
> chown -R resin:resin "${D}${RESIN_HOME}"
> chown -R resin:resin "${D}/etc/resin"
> chown -R resin:resin "${D}/var/log/resin"
> chown -R resin:resin "${D}/var/lib/resin"
> chown -R resin:resin "${D}/var/run/resin"
>
> chmod 644 "${D}/etc/conf.d/resin"
> chmod 755 "${D}/etc/init.d/resin"
> chmod 750 "${D}/var/lib/resin"
> chmod 750 "${D}/var/run/resin"
> chmod 750 "${D}/etc/resin"
Again, fowners/fperms can make this a bit easier to read.
Thanks,
Donnie
--
[EMAIL PROTECTED] mailing list
[gentoo-dev] Adding pkg_preinst to java-utils-2 for QA purposes
I wrote a patch to add pkg_preinst to java-utils-2 to check the jars in
the images for problems in the jars but it can later be used for other
stuff too. If there are no objections we will need to modify the
following ebuilds as not to change their behavior. I will do this in
behalf of the maintainers in question so that all the changes go in at
the same time.
[EMAIL PROTECTED] /mnt/checkouts/java/scripts $ bash
find-function-usage-in-inheriting-ebuilds.sh --can-add --no-default
java-pkg-2 pkg_preinst
perl-module.eclass: dev-perl/Inline-Java/Inline-Java-0.50.ebuild
perl-module.eclass: dev-perl/Inline-Java/Inline-Java-0.51.ebuild
perl-module.eclass: dev-perl/Inline-Java/Inline-Java-0.52.ebuild
[EMAIL PROTECTED] /mnt/checkouts/java/scripts $ bash
find-function-usage-in-inheriting-ebuilds.sh --can-add --no-default
java-pkg-opt-2 pkg_preinst
perl-module.eclass: app-pda/pilot-link/pilot-link-0.12.2.ebuild
perl-module.eclass: dev-util/subversion/subversion-1.3.2-r3.ebuild
perl-module.eclass: dev-util/subversion/subversion-1.3.2-r4.ebuild
The script can be found in
https://overlays.gentoo.org/svn/proj/java/scripts/find-function-usage-in-inheriting-ebuilds.sh
Regards,
Petteri
Index: java-utils-2.eclass
===
RCS file: /var/cvsroot/gentoo-x86/eclass/java-utils-2.eclass,v
retrieving revision 1.100
diff -u -r1.100 java-utils-2.eclass
--- java-utils-2.eclass 28 Nov 2007 02:03:48 - 1.100
+++ java-utils-2.eclass 16 Dec 2007 01:57:29 -
@@ -293,7 +295,6 @@
java-pkg_do_write_
}
-
#
--
# @internal-function depend-java-query
#
@@ -833,6 +830,30 @@
}
#
--
+# @eclass-pkg_preinst
+#
+# Checks jars in the image for problems.
+#
+#
--
+
+java-utils-2_pkg_preinst() {
+ debug-print-function ${FUNCNAME} "[EMAIL PROTECTED]"
+ is-java-strict || return
+
+ for jar in $(find "${D}" -name "*.jar"); do
+ for dep in $(unzip -p "${jar}" META-INF/MANIFEST.MF | egrep
"^Class-Path:" | cut -d ":" -f 2); do
+ local dir=$(dirname "${jar}")
+ einfo "${dir}/${dep}" "${ROOT}${dir#${D}}/${dep}"
+ if [[ ! -e "${dir}/${dep}" && ! -e
"${ROOT}${dir#${D}}/${dep}" ]]; then
+ eerror "${jar}"
+ eerror "has Class-Path entry in Manifest
pointing at ${dep}"
+ eerror "but that file does not exist in image
or ROOT=${ROOT}"
+ fi
+ done
+ done
+}
+
+#
--
# @section-end install
#
--
@@ -2729,6 +2760,8 @@
return $?
}
+EXPORT_FUNCTIONS pkg_preinst
+
#
--
# @eclass-end
#
--
signature.asc
Description: OpenPGP digital signature
Re: [gentoo-dev] Ebuilds I lastrited month ago for treecleaners, time to look at them once more.
On Dec 15, 2007 8:00 AM, Samuli Suominen <[EMAIL PROTECTED]> wrote: > x11-wm/flwm (Coldwind promised to take a look at this, it needs a > patched fltk.) Both fltk and flwm fixed. x11-wm/flwm should go out of p.mask when fltk and flwm get proper keywords again. -- Santiago M. Mola Jabber ID: [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
[gentoo-dev] Guidance needed for an Erlang ebuild
Hi, I'm not an official Gentoo dev but I play one in my overlay. For bug #192278 - an ebuild for the very cool RabbitMQ message broker - I'm looking for someone with experience in writing Erlang package ebuilds; I have a few questions regarding fs layout and stuff, so that my contributed ebuild can go into portage. I figure that's faster than me posting an attempt into bugzilla and then not getting any feedback :) would love to hear from someone - Holger PS: dear Gentoo devs: I <3 you all. Thanks for Gentoo. -- [EMAIL PROTECTED] mailing list
[gentoo-dev] MIPS Equipment
Hi team, We have a forums user in the chicago/milwaukee area wanting to dispose of some MIPS equipment. See http://forums.gentoo.org/viewtopic-p-4620101.html#4620101 Regards, Roy Bamford (NeddySeagoon) -- [EMAIL PROTECTED] mailing list
Re: [gentoo-dev] EAPI placement
On Sat, 15 Dec 2007 10:43:35 +0100 Vlastimil Babka <[EMAIL PROTECTED]> wrote: > Ciaran McCreesh wrote: > > On Wed, 12 Dec 2007 23:14:24 +0100 > > Carsten Lohrke <[EMAIL PROTECTED]> wrote: > >> I disagree here. It would be annoying and possibly even hindering > >> in future not being able to use higher EAPI features in eclasses. > >> Point is the eclass has to check, if the author of an ebuild sets > >> another EAPI and throw an error, in this case. > > > > There is no way for an eclass to throw an error. Nor, with the > > current way Portage implements EAPI, is there a way to add such a > > way. > > How bout declaring all supported (possibly with later conditional > checks on EAPI variable etc) EAPIs in eclass via some variable, and > repoman checking that EAPI set in the ebuild is compatible with all > inherited eclasses? And if you need newer EAPI in the ebuild, get > eclasses updated first (even if its just updating the compatibility > declaration). Also, repoman could check that EAPI is not being set in > the eclass. Nice idea, but gets quite complicated when multiple (possibly nested) eclasses are involved as we'd have to generate a complete inheritance graph and get that new variable separately for each eclass. Or alternatively each eclass has to update the variable with the intersection between the old value and it's own set of supported EAPIs. And doesn't really solve the problem, just reduces the scope a fair bit (e.g. if an eclass drops support for an EAPI version repoman will only catch it gradually if the packages using it are bumped, unless you manually check the whole tree). Marius -- [EMAIL PROTECTED] mailing list
Re: [gentoo-dev] EAPI placement
Ciaran McCreesh wrote: > On Wed, 12 Dec 2007 23:14:24 +0100 > Carsten Lohrke <[EMAIL PROTECTED]> wrote: >> I disagree here. It would be annoying and possibly even hindering in >> future not being able to use higher EAPI features in eclasses. Point >> is the eclass has to check, if the author of an ebuild sets another >> EAPI and throw an error, in this case. > > There is no way for an eclass to throw an error. Nor, with the current > way Portage implements EAPI, is there a way to add such a way. How bout declaring all supported (possibly with later conditional checks on EAPI variable etc) EAPIs in eclass via some variable, and repoman checking that EAPI set in the ebuild is compatible with all inherited eclasses? And if you need newer EAPI in the ebuild, get eclasses updated first (even if its just updating the compatibility declaration). Also, repoman could check that EAPI is not being set in the eclass. VB -- [EMAIL PROTECTED] mailing list
Re: [gentoo-dev] Ebuilds I lastrited month ago for treecleaners, time to look at them once more.
On Sat, Dec 15, 2007 at 09:00:28AM +0200, Samuli Suominen wrote: > sys-apps/tcng As the person that added it, and the maintainer: Make it die. Last release was Oct 2004, and it's heavily dependant on specific kernel versions (2.4 kernel sources are used in the build, and 2.6 headers must not be installed). > x11-misc/fluxspace (lack, other fluxbox users/maintainers?) I tried a long time ago, with the 0.4.0_alpha version from upstream: the alpha doesn't compile with GCC4, and it needs fluxbox internal libs. 0.3.0 also has weird graphics bugs - I used to use it, but have since moved to fbpager. -- Robin Hugh Johnson Gentoo Linux Developer & Infra Guy E-Mail : [EMAIL PROTECTED] GnuPG FP : 11AC BA4F 4778 E3F6 E4ED F38E B27B 944E 3488 4E85 pgpJZYaegFNFU.pgp Description: PGP signature
