April 2009 - Arch-dev-public - lists.archlinux.org

[arch-dev-public] [PATCH] commitpkg: rework for new community layout
by Dan McGee 04 Apr '09

04 Apr '09

Remove the community-specific code that calls tupkg. Make the server variable a bit more dependent on what repo we are uploading to in order to get community working, and finally create the correct symlinks for communitypkg in the Makefile. Note that this change will require the Arch devtools package to get a replace entry for aurtools. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- Makefile | 2 ++ commitpkg | 32 +++++++++++--------------------- 2 files changed, 13 insertions(+), 21 deletions(-) diff --git a/Makefile b/Makefile index 6a99366..f29082a 100644 --- a/Makefile +++ b/Makefile @@ -8,6 +8,7 @@ install: ln -sf commitpkg $(DESTDIR)/usr/bin/extrapkg ln -sf commitpkg $(DESTDIR)/usr/bin/corepkg ln -sf commitpkg $(DESTDIR)/usr/bin/testingpkg + ln -sf commitpkg $(DESTDIR)/usr/bin/communitypkg # arch{co,release,rm} install -m 755 archco $(DESTDIR)/usr/bin install -m 755 archrelease $(DESTDIR)/usr/bin @@ -27,6 +28,7 @@ uninstall: rm $(DESTDIR)/usr/bin/extrapkg rm $(DESTDIR)/usr/bin/corepkg rm $(DESTDIR)/usr/bin/testingpkg + rm $(DESTDIR)/usr/bin/communitypkg rm $(DESTDIR)/usr/sbin/mkarchroot rm $(DESTDIR)/usr/sbin/makechrootpkg rm $(DESTDIR)/usr/bin/lddd diff --git a/commitpkg b/commitpkg index c434974..8f4d37d 100755 --- a/commitpkg +++ b/commitpkg @@ -43,6 +43,8 @@ if [ ! -f $pkgfile ]; then fi fi +# set up repo-specific opts depending on how we were called +server="archlinux.org" if [ "$cmd" == "extrapkg" ]; then repo="extra" elif [ "$cmd" == "corepkg" ]; then @@ -53,6 +55,7 @@ elif [ "$cmd" == "unstablepkg" ]; then repo="unstable" elif [ "$cmd" == "communitypkg" ]; then repo="community" + server="community.archlinux.org" else if [ $# -eq 0 ]; then echo "usage: commitpkg <reponame> [-l limit] [commit message]" @@ -69,29 +72,16 @@ if [ "$1" = "-l" ]; then shift 2 fi -if [ "$repo" != "community" ]; then - # combine what we know into a variable - uploadto="staging/${repo}/$(basename ${pkgfile})" - server="archlinux.org" - scp ${scpopts} "${pkgfile}" "${server}:${uploadto}" - if [ "$(md5sum "${pkgfile}" | cut -d' ' -f1)" != "$(ssh ${server} md5sum "${uploadto}" | cut -d' ' -f1)" ]; then - echo "File got corrupted during upload, cancelled." - exit 1 - else - echo "File integrity okay." - fi +# combine what we know into a variable +uploadto="staging/${repo}/$(basename ${pkgfile})" +scp ${scpopts} "${pkgfile}" "${server}:${uploadto}" +if [ "$(md5sum "${pkgfile}" | cut -d' ' -f1)" != "$(ssh ${server} md5sum "${uploadto}" | cut -d' ' -f1)" ]; then + echo "File got corrupted during upload, cancelled." + exit 1 else - if [ ! -f ~/.tupkg ]; then - echo "Must configure tupkg via ~/.tupkg, cancelled" - exit 1 - fi - if [ "$(basename $pkgfile)" != "$(basename $oldstylepkgfile)" ]; then - echo "Renaming makepkg3 package for compatibility" - mv $pkgfile $oldstylepkgfile - pkgfile=$oldstylepkgfile - fi - tupkg $pkgfile + echo "File integrity okay." fi + if [ $? -ne 0 ]; then echo "Cancelled" exit 1 -- 1.6.2.1

2 1

[arch-dev-public] [signoff] mailx-8.1.1-6
by Allan McRae 04 Apr '09

04 Apr '09

Added smtp-forwarder as optdepends to fix FS#4812. I do not use so no default sign-off from me... Allan

5 7

[arch-dev-public] devtools and specifying a username
by Dan McGee 04 Apr '09

04 Apr '09

Hey guys, especially the new ones, I noticed this commit tonight on devtools: Specify upload server in variable Allan McRae [Fri, 9 Jan 2009 14:11:59 +0000 (00:11 +1000)] This means people with different usernames on their local machine only need to change a single line. http://projects.archlinux.org/?p=devtools.git;a=commitdiff;h=1b178621a5fc1e… While well intentioned, it is completely wrong and means you have to update your script every time a devtools update comes! Use .ssh/config and set things up right by adding something like the following: Host svn.archlinux.org User dan Host archlinux.org User dan -Dan

2 1

[arch-dev-public] dropping Gnash?
by Andreas Radke 04 Apr '09

04 Apr '09

Gnash is in Orphaned state for a long time. Is there any Dev wanting to pick it up? Any TU wants to maintain it? If not we will remove it so our community might pick it up for AUR. I myself will keep maintaining swfdec in extra as an OSS alternative for Adobe flash. -Andy

7 13

[arch-dev-public] Integrity Check x86_64 03-04-2009
by repomaint＠archlinux.org 03 Apr '09

03 Apr '09

======================================== = Integrity Check x86_64 of core,extra = ======================================== Performing integrity checks... ==> parsing pkgbuilds ==> checking mismatches ==> checking archs ==> checking dependencies ==> checking makedepends ==> checking for circular dependencies Missing Dependencies ---------------------- archboot --> 'netkit-telnet>=0.17-8' archboot --> 'wlan-ng26>=0.2.9-2' archboot --> 'atl2>=2.0.5-2' Missing Makedepends --------------------- wpa_supplicant --> 'kernel26<2.6.28' perl-mail-spf --> 'perl-net-dns-resolver-programmable' gnome-speech --> 'espeak' Repo Hierarchy for Makedepends -------------------------------- core/sqlite3 depends on extra/tcl core/crda depends on extra/python-m2crypto core/iputils depends on extra/opensp core/iputils depends on extra/libxslt core/iputils depends on extra/docbook-xsl core/madwifi-utils depends on extra/sharutils core/e2fsprogs depends on extra/bc core/ca-certificates depends on extra/ruby core/madwifi depends on extra/sharutils Circular Dependencies ----------------------- gcc-gcj>eclipse-ecj>gcc-gcj Summary --------- Missing PKGBUILDs: 0 Invalid PKGBUILDs: 0 Mismatching PKGBUILD names: 0 Duplicate PKGBUILDs: 0 Invalid archs: 0 Missing (make)dependencies: 6 Repo hierarchy problems: 9 Circular dependencies: 1

1 0

[arch-dev-public] Integrity Check i686 03-04-2009
by repomaint＠archlinux.org 03 Apr '09

03 Apr '09

====================================== = Integrity Check i686 of core,extra = ====================================== Performing integrity checks... ==> parsing pkgbuilds ==> checking mismatches ==> checking archs ==> checking dependencies ==> checking makedepends ==> checking for circular dependencies Missing Dependencies ---------------------- archboot --> 'netkit-telnet>=0.17-8' archboot --> 'wlan-ng26>=0.2.9-2' archboot --> 'atl2>=2.0.5-2' Missing Makedepends --------------------- wpa_supplicant --> 'kernel26<2.6.28' perl-mail-spf --> 'perl-net-dns-resolver-programmable' gnome-speech --> 'espeak' Repo Hierarchy for Makedepends -------------------------------- core/sqlite3 depends on extra/tcl core/crda depends on extra/python-m2crypto core/iputils depends on extra/opensp core/iputils depends on extra/libxslt core/iputils depends on extra/docbook-xsl core/madwifi-utils depends on extra/sharutils core/e2fsprogs depends on extra/bc core/ca-certificates depends on extra/ruby core/madwifi depends on extra/sharutils Circular Dependencies ----------------------- gcc-gcj>eclipse-ecj>gcc-gcj Summary --------- Missing PKGBUILDs: 0 Invalid PKGBUILDs: 0 Mismatching PKGBUILD names: 0 Duplicate PKGBUILDs: 0 Invalid archs: 0 Missing (make)dependencies: 6 Repo hierarchy problems: 9 Circular dependencies: 1

1 0

Re: [arch-dev-public] [arch-general] Make arch-announce read only?
by Aaron Griffin 03 Apr '09

03 Apr '09

On Tue, Mar 31, 2009 at 8:30 PM, pyther <pyther(a)pyther.net> wrote: > > I was wondering if we could make arch-announce a read only list. Sort of > like the dev mailing list. Anyone who tries to respond to an announcement > could get an email saying something along the lines of "If you have a > comment about an announcement, please post it on the arch-general list. > Thank you." > > I do not filter the arch-announce mailings for obvious reason and I do not > think I am alone. It is very annoying to see replies in my inbox. Forwarding to the dev list - what do you guys think? I think it's a good idea

2 2

Re: [arch-dev-public] KOffice 1.9.99.0 RC1 uploaded
by Ronald van Haren 03 Apr '09

03 Apr '09

On 4/3/09, Pierre Schmitz <pierre(a)archlinux.de> wrote: > Hi Ronald (and others), > > this mail is quite interesting. Especially that we still ship lcms without > the > security patch. Imho we should update to lcms 1.18. I hope it does not > introduce a so name bump...but I think it will. :-( > > Pierre > replied from arch-dev mailing list. I think we're safe just updating lcms without rebuilding anything. Everything seems to remain in working order and lddd also does not find anything lcms related after a local version bump. When nobody bumped the repo version at the end of the weekend I should have time to do so. @Pierre: which mailing list is that you get those messages from so I can subscribe? Ronald

3 2

[arch-dev-public] [signoff] mysql, mysql-clients and libmysqlclient
by Douglas Soares de Andrade 03 Apr '09

03 Apr '09

Hi, I have just updated mysql to 5.1.33 and it has been working great here - thanks for everyone that helped in the process. I want to move the asap to extra. Any cons here ? Thanks a lot =)

3 4

[arch-dev-public] WTF? i686 support?
by Aaron Griffin 02 Apr '09

02 Apr '09

I specifically told you guys that we were NOT dropping the i686 port, and you guys went and voted on it without me? This is the last straw. If you guys don't fix this now, consider this my last email.

11 13