If x11-misc/autorandr-1.14 is installed, the udev rule refers to systemd service instead of autorandr call. cat /lib/udev/rules.d/40-monitor-hotplug.rules ACTION=="change", SUBSYSTEM=="drm", RUN+="/bin/systemctl start --no-block autorandr.service" Therefore autorandr is broken for openrc users in this version. It seems the same issue is back like the second one in https://bugs.gentoo.org/803545.
The issue is still not solved in x11-misc/autorandr-1.15 I thing the cause of comeback the issue is this change: https://github.com/phillipberndt/autorandr/pull/345 that was merged into 1.14
I am seeing this issue as well on an OpenRC system. running the `make install_udev` command gives the following output: mkdir -p ///usr/lib/udev/rules.d/ echo 'ACTION=="change", SUBSYSTEM=="drm", RUN+="/usr//bin/autorandr --batch --change --default default"' > ///usr/lib/udev/rules.d/40-monitor-hotplug.rules and indeed the indicated file has the correct output. so the ebuild must be doing something different. I noticed it runs emake with SYSTEMD_UNIT_DIR="$(systemd_get_systemunitdir)" unconditionally rather than only if the use flag is in effect. from the makefile it would seem the output issue comes from this if statement triggering: # Rules for systemd SYSTEMD_UNIT_DIR:=$(shell pkg-config --variable=systemdsystemunitdir systemd 2>/dev/null) ifneq (,$(SYSTEMD_UNIT_DIR)) DEFAULT_TARGETS+=systemd endif removing the SYSTEMD_UNIT_DIR assignment from the emake call in the ebuild gives the correct output, so I surmise that $(systemd_get_systemunitdir) dies not return an empty stirn on non-systemd installs the way that pkg-config --variable=systemdsystemunitdir systemd does. seems like the fix is either to conditionally set SYSTEMD_UNIT_DIR, or to use pkg-config --variable=systemdsystemunitdir systemd rather than $(systemd_get_systemunitdir), but its possible this also indicates a QA issue with $(systemd_get_systemunitdir)?
this change to the ebuild produces correct results for both systemd and -systemd 61c61 < SYSTEMD_UNIT_DIR="$(systemd_get_systemunitdir)" \ --- > SYSTEMD_UNIT_DIR=$(usex systemd "$(systemd_get_systemunitdir)" "") \
Yep, the problem are those lines in upstream's Makefile ifneq (,$(SYSTEMD_UNIT_DIR)) DEFAULT_TARGETS+=systemd endif https://github.com/phillipberndt/autorandr/blob/770dc86bec9d20bbe22b596aad9aa1520c966720/Makefile#L84-L86
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=0acccf502267e74b578035511adaf2d53e1eaf67 commit 0acccf502267e74b578035511adaf2d53e1eaf67 Author: Florian Schmaus <flow@gentoo.org> AuthorDate: 2024-09-23 06:42:56 +0000 Commit: Florian Schmaus <flow@gentoo.org> CommitDate: 2024-09-23 06:47:26 +0000 x11-misc/autorandr: fix non-systemd udev rule, enable python 3.13 Thanks to wolfwood for digging into the root cause and suggesting the used fix. Closes: https://bugs.gentoo.org/912692 Signed-off-by: Florian Schmaus <flow@gentoo.org> x11-misc/autorandr/autorandr-1.15-r1.ebuild | 76 +++++++++++++++++++++++++++++ x11-misc/autorandr/autorandr-9999.ebuild | 4 +- 2 files changed, 78 insertions(+), 2 deletions(-)
