- Arch-general - lists.archlinux.org

[arch-general] Fluidsynth Plugin for VLC media player
by 山田ハヤオ 26 Aug '21

26 Aug '21

Add --enable-fluidsynth to PKGBUILD in the vlc package in the official repository. Consider adding the FluidSynth plugin to support Midi playback in the official repository. If you can't do that, please tell me why.

3 4

[arch-general] PyJWT 2.0+
by Marcus Hoffmann 21 Aug '21

21 Aug '21

Hi, I wanted to ask if anyone knows what's up with updating https://archlinux.org/packages/?q=python-pyjwt to v2.0+. v2.0 has been released last December, and there's been a bunch of minor versions since. I realize this has a bunch of dependent packages, is there anything in particular blocking this? Mostly asking because I adopted python-social-auth-core in the AUR which depends on pyjwt 2.0+ since January this year. (others say it works fine for them with the older version, but I'm hesitant to ignore upstreams version range suggestion.) Best, Marcus

2 1

[arch-general] File conflict between official packages
by 山田ハヤオ 20 Aug '21

20 Aug '21

I tried to install cups-filters and pantheon-default-settings. As a result, Pacman reported a "/usr/share/cups/data/default-testpage.pdf" conflict. This issue should be fixed. PS Since I used the mailing list for the first time, I apologize for sending it to some wrong places.

2 1

[arch-general] about pipewire
by Javier 02 Aug '21

02 Aug '21

Hello ! I do understand pipewire is required on wayland to be able to share the desktop. Is its audio component required as well for desktop sharing on wayland? I'm talking future wise, and mainly about some applications which allows audio and video conferencing as well as desktop sharing, like zoom, slack and perhaps webrtc. If its audio component is not required, is it possible at all to let pipewire without its audio components? Would it be enough to avoid installing its alsa and pulse plugins? Any additional configuration required? I have avoided using pulse, and I really don't want to use pipewire audio component either. I might have to use pipewire if moving to wayland in the future for whatever reason. I have hybrid used cases for audio. On the desktop of course, but also headless, since I also left boxes playing music without anyone logged in (I log in, make mpd play a list, and log out), and I times go back and forth from being logged in and logged out, and not necessarily I log to the desktop, just console or ssh connections might be enough to start playing music on mpd. Plain alsa does great for these hybrid use model. There are as well other non common gnu+linux midi frameworks which do well, like sndio, but pulse-audio and pipewire don't. Thanks ! -- Javier

2 2

[arch-general] Why does cx_freeze console create a core dump?
by Peter Nabbefeld 01 Aug '21

01 Aug '21

Hello, am I doing sth. wrong, when calling "/usr/lib/python3.9/site-packages/cx_Freeze/bases/Console-cp39-linux-x86_64"? The console seems to be some standard piece of software, so in general it shouldn't create a core dump without some missing or corrupt installation dependency - any idea? Well, I've heard I'm not the only one having problems with this, and I'm not used to work with core dumps (not a C programmer, sorry), so I would be very lucky if someone else could fix that. Kind regards, Peter

2 1

[arch-general] [aur] cctag build error
by Peter Nabbefeld 31 Jul '21

31 Jul '21

Hello, when I try to build "cctag" as a new dependency of alice-vision-git, I get "~/.cache/pikaur/build/cctag/src" does not appear to contain CMakeLists.txt". I could find it in ~/.cache/pikaur/build/cctag/src/CCTag-1.0.0/src, so the repository snapshot is probably handled incorrectly. I'm not sure if cctag should reside in it's own project, as it seems to be part of and only used by alice-vision. However, even when I modify PKGBUILD to point to ~/.cache/pikaur/build/cctag/src/CCTag-1.0.0/src, CMakeLists.txt isn't found, with the same incorrect path shown in the error message; it seems, the -S option is ignored. :-O After insertion of "cd" I could start the build, but now the following error message is provided: CMake Error at CMakeLists.txt:172 (CUDA_ADD_LIBRARY): Unknown CMake command "CUDA_ADD_LIBRARY". How can I fix this? Kind regards P.

2 3

[arch-general] udev complaint and USB malfunction
by set＠nmset.info 31 Jul '21

31 Jul '21

This message appears in /var/log/errors.log (using syslog-ng) (don't know since when): systemd-udevd[424]: /usr/lib/udev/rules.d/99-indi_auxiliary.rules:5 Invalid value "/bin/sh -c 'test -f /sys/module/usbcore/parameters/usbfs_memory_mb && test $(cat /sys/module/usbcore/parameters/usbfs_memory_mb) -lt 256 && echo 256 > /sys/module/usbcore/parameters/usbfs_memory_mb'" for RUN (char 76: invalid substitution type), ignoring, but please fix it. I don't have a clue about an appropriate fix, apart from deleting 99- indi_auxiliary.rules, and don't know how valid it would be to delete this file. I'm having this strange USB behaviour on 2 machines : Plugging a USB stick may not trigger the 'New device' popup in KDE, while lsblk does list the USB stick. The latter may be connected directly to the machine, or via extension cable or a USB hub. It works half of the time as expected. An this one on one recent host : A USB 3 hub (tried 2 hubs) may not always list its devices, with these errors on resume from a *long* suspend, or on reboot : Jul 31 12:24:22 localhost.localdomain kernel: usb 5-3: device descriptor read/ 64, error -71 Jul 31 12:24:22 localhost.localdomain kernel: usb 5-3: device descriptor read/ 64, error -71 Jul 31 12:24:22 localhost.localdomain systemd-udevd[408]: /usr/lib/udev/ rules.d/99-indi_auxiliary.rules:5 Invalid value "/bin/sh -c 'test -f /sys/ module/usbcore/parameters/usbfs_memory_mb && test $(cat /sys/module/usbcore/ parameters/usbfs_memory_mb) -lt 256 && echo 256 > /sys/module/usbcore/ parameters/usbfs_memory_mb'" for RUN (char 76: invalid substitution type), ignoring. Jul 31 12:24:22 localhost.localdomain kernel: usb 5-3: device descriptor read/ 64, error -71 Jul 31 12:24:22 localhost.localdomain kernel: usb 5-3: device descriptor read/ 64, error -71 Jul 31 12:24:23 localhost.localdomain kernel: usb 5-3: device not accepting address 5, error -71 Jul 31 12:24:24 localhost.localdomain kernel: usb 5-3: device not accepting address 6, error -71 Jul 31 12:24:24 localhost.localdomain kernel: usb usb5-port3: unable to enumerate USB device I can't know if all these are related. I am requesting help in search of an easy fix. Thanks and regards.

3 2

[arch-general] pacman error
by Jude DaShiell 27 Jul '21

27 Jul '21

Installation path was archinstall along with some post-install configuration needed to enable accessibility among other things. Successful boot with speech coming up. Then as root pacman -Syu I get back an error saying pacman cannot lock data bases and pacman fails. pacman-key --populate ran before the pacman command and finished with no problems. What did I do wrong?

2 1

[arch-general] Rescue mode stopped working
by David Rosenstrauch 27 Jul '21

27 Jul '21

I'm not sure what changed, but for some reason I'm no longer able to drop into rescue mode on my laptop (i.e. using "systemctl rescue") - or to boot directly into single user mode. When I try either of these, the kernel loads, and then looks like it's entering single user mode, prompting me to enter my root password to proceed (as it should). But when I do enter the root password at the prompt - or even just press enter - the system quickly displays "Reloading system manager configuration", followed by something about switching to default target, and then switches into full interactive mode complete with graphical login prompt. I tried downgrading down to several earlier versions of systemd (back to 248), as well as trying to boot using the LTS kernel in single user mode, but I still get the issue. This behavior has definitely changed very recently (i.e., within a few days to a week) as I use single user mode often, and only just noticed this today. I'd file a bug about this but I have no idea what package might be at fault. Any pointers appreciated! Thanks, DR

1 1

[arch-general] Angband package flagged out-of-date since months
by Tim Schumacher 26 Jul '21

26 Jul '21

Hello there, I just wanted to ask if anyone knows where the holdup on updating the angband package [1] comes from. The newest version, 4.2.2, has been released on the 7th of March [2], and the package was subsequently flagged out-of-date, but there hasn't been any activity since then. Thanks, Tim [1] https://archlinux.org/packages/community/x86_64/angband/ [2] https://angband.github.io/angband/release/4.2.2

2 1