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

[arch-dev-public] dropping flashplugin x86_64
by Ionuț Bîru 15 Jun '10

15 Jun '10

Hi, the current x86_64 version of flashplugin has security problems and adobe dropped/temporally closed x86_64 releases. How do we manage this situation? Only dropping from repo? Should we submit an empty package? Suggest our users to remove their package if they consider security important? -- Ionuț

3 3

[arch-dev-public] [PATCH] Detect more usages of $startdir
by Allan McRae 15 Jun '10

15 Jun '10

The detection of $startdir usage ignored any potential brackets or quotes. Now the following are all detected: $startdir ${startdir} "$startdir" "${startdir}" Signed-off-by: Allan McRae <allan(a)archlinux.org> --- Namcap/invalidstartdir.py | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/Namcap/invalidstartdir.py b/Namcap/invalidstartdir.py index f323b76..720e5cd 100644 --- a/Namcap/invalidstartdir.py +++ b/Namcap/invalidstartdir.py @@ -29,7 +29,7 @@ class package: def analyze(self, pkginfo, tar): ret = [[], [], []] for i in pkginfo.pkgbuild: - startdirs = re.split('\$startdir', i) + startdirs = re.split('\${?startdir}?"?', i) for j in startdirs[1:]: if j[:4] != '/pkg' and j[:4] != '/src': ret[0].append(("file-referred-in-startdir", ())) -- 1.7.1

2 1

[arch-dev-public] [signoff] pcmciautils-017-1
by Tobias Powalowski 15 Jun '10

15 Jun '10

Hi guys, bump to latest version, please signoff both arches greetings tpowa -- Tobias Powalowski Archlinux Developer & Package Maintainer (tpowa) http://www.archlinux.org tpowa(a)archlinux.org

2 2

[arch-dev-public] Namcap 2.6 in [testing]
by Dan McGee 15 Jun '10

15 Jun '10

Hopefully it works for everyone! Changes since 2.5 from the NEWS file: $ head NEWS namcap (2.6) - Add new 'anyelf' rule; should catch ELF files that end up in 'any' arch packages and cannot be arch-independent - Allow checking of all checksum types (FS#17168) - Code style and import cleanups - Handle optdepends descriptions correctly (FS#18259) - Make it an error to use $startdir in PKGBUILDs -Dan

2 1

[arch-dev-public] Poor man's mirror redirector
by Pierre Schmitz 15 Jun '10

15 Jun '10

Hi all, I wrote a simply redirector script for fun*. It uses geoip to redirect you to a random Arch mirror in your country; but only if that mirror has synced within the last 24h. It is located at https://downloads.archlinux.de/ So you can for example download an iso using this URL: https://downloads.archlinux.de/iso/2010.05/archlinux-2010.05-netinstall-i68… or simply redirect to a dir like https://downloads.archlinux.de/iso/ It's probably not that useful and don't even think about using that url in pacman. It will redirect you to a random mirror on every request no matter if that mirror is up or has this file. (also pacman doesn't seem to handle https; but once could use the curl xfer command (wget has poor https support too)) However, a possible usage might be to use this for simpler iso download. Greetings, Pierre *) https://git.archlinux.de/www.archlinux.de.git/tree/pages/GetFileFromMirror.… -- Pierre Schmitz, https://users.archlinux.de/~pierre

2 1

[arch-dev-public] When to remove provides/conflicts/replaces?
by Allan McRae 13 Jun '10

13 Jun '10

Hi, Do we have any guidelines on when to remove the provides/conflicts/replaces arrays from a PKGBUILD? In other words, what is the oldest system we support updating from? Allan

2 4

[arch-dev-public] [signoff] gettext-0.18.1.1-1
by Tobias Powalowski 13 Jun '10

13 Jun '10

Hi bump to latest version, please signoff both arches, thanks greetings tpowa -- Tobias Powalowski Archlinux Developer & Package Maintainer (tpowa) http://www.archlinux.org tpowa(a)archlinux.org

3 3

[arch-dev-public] [signoff] mpfr-3.0.0-1, gcc-4.5.0-6 and libmpc-0.8.2-2
by Allan McRae 12 Jun '10

12 Jun '10

mpfr-3.0.0-1: upstream update, passes entire testsuite. gcc-4.5.0-6: update to 20100610 snapshot, patch to fix khtml/chromium issues (please test). libmpc-0.8.2-2: mpfr rebuild only

3 4

[arch-dev-public] coreutils - remove i18n patch?
by Allan McRae 12 Jun '10

12 Jun '10

Hi, The coreutils-i18n patch we currently use adds some extra support for multibyte characters. However, the patch is entirely unsupported upstream. I currently rely on the Fedora maintainer to fix any issues that are found and to update it for each release. He also finds that patch a nightmare. So the question is, should we remove that patch? It is not acceptable by the usual Arch patching rules given it is not supported upstream, but it does provide better multibytle locale support (subject to it working...). Removing the patch will probably result in a few regressions with multibyte locales, but we can then take those issues upstream. Any opinions on this? Allan

3 4

[arch-dev-public] [signoff] xz 4.999.9beta-4
by Pierre Schmitz 11 Jun '10

11 Jun '10

new upstream snapshot. Please sign off. -- Pierre Schmitz, https://users.archlinux.de/~pierre

3 3