This setting will use the best quality or latency profiles for BT devices if
available. HSP/HFP profiles will always be ignored. The setting is a string that
only accepts 'quality' and 'latency' strings. Any other value will be treated
the same way as the 'quality' value.
The core object (ID 0) is implicit in the PipeWire connection and never
appears in the permission manager's ObjectManager. Add a
core_permissions field to set explicit permissions on it independently
of default_permissions.
This is suspected to be the reason why the CI pipelines fail on some
of the linking tests. Reverting the commit for now, until the issue
is better understood.
See #934
This reverts commit 529aaa66cb.
The refactoring uses a new 'select-access' event to select the access for each
client with a fallback mechanism. The fallback priority is: configuration,
flatpak, snap, portal, and default.
The access JSON configuration has also been improved so that users can create
their custom permission managers and attach them to any client. See the access
configuration example for more information describing how to do this.
The `properties` argument of `wp_impl_module_load()` is marked
"transfer none", thus the caller's reference remains valid and
must be disposed of.
Fixes: ef29018c55 ("m-lua-scripting: Add WpImplModule bindings")
Split rescan-on-linkable triggering into a separate script file
(linking/rescan-on-linkable.lua) that can be disabled via
wireplumber.profiles by setting hooks.linking.rescan-on-linkable =
disabled.
Since 'autoswitch-*' events have lower priority than 'select-*' events,
this guarantees that the autoswitch hooks will always run after the
'device/apply-profile' hook, avoiding possible race conditions.
These local events have lower priority than the 'create-*' and 'select-*' ones,
and are meant to be used when wireplumber wants to automatically switch profiles
or other things.
We cannot only rely on the input routes to check whether a profile is a headset
profile or not because some BT devices can have A2DP source nodes, causing the
autoswitch logic to not work properly.
This patch fixes the problem by also checking if the profile name matches the
'headset-head-unit*' or 'bap-duplex' patterns. If the profile name does not
match those patterns, the profile is not considered headset profile.
See #926
Setting api.bluez5.internal=true changes media.class to
Audio/(Stream|Sink)/Internal which makes HFP HF output get wrong media
class.
Don't set this for headset-audio-gateway. Fixes HFP HF stream media
class.
This allows us to have per-device configuration for device volumes, via
a `device.routes.default-volume` property. This allows a non-global
override for situations where we want to expose a different out-of-box
volume (say for an internal speaker where we know a better comfortable
default setting, or HDMI where we might want to avoid attenuation by
default).
At the moment the libcamera monitor and "Device" objects are loaded in the
wireplumber process, but the "Node" objects are loaded in the pipewire daemon.
Due to that, both processes have their own separate `libcamera::CameraManager`
objects. These operate independently.
As a consequence of the above, there is an inherent race condition: when a
new camera appears and the wireplumber process detects it and instructs the
pipewire process to create a "Node" for it, at that point, the camera might
not exist in the `CameraManager` of the "pipewire" process.
This can happen during the initial enumeration as well as hotplug. So load
the nodes locally in the wireplumber process so that all libcamera objects
use the same `CameraManager`, thus eliminating the race condition.
This is required in order to allow plugins to use GL as mincore
is used in Mesas `_eglPointerIsDereferenceable()`.
One example for a client wanting to do so is the in-development
libcamera GPUISP, see https://patchwork.libcamera.org/cover/24183/
(cherry picked from commit pipewire@4796b3fb9524c20ac0f5006143b6a13ee50c01ec)
See pipewire/pipewire!2530
Allows installation of systemd services without libsystemd installed.
Useful for Alpine Linux where systemd services are allowed to be subpackaged
(e.g. for postmarketOS) but hasn't systemd in it's repos.
It has no change in existing behavior, but installs services if the unit
directories are explicitly set.
In `wp_interest_event_hook_get_matching_event_types()`, the variable
`res` owns the allocated `GPtrArray`, but there is an early return
in the function. Hitting that will leak the allocation, so use
`g_autoptr` to avoid that.
Fixes: b80a0975c7 ("event-dispatcher: Register hooks for defined events in a hash table")
