Reset back to the session item's "configured" state instead of clearing
the si properties completely. This allows the "session-item-removed"
event to be dispatched with all the original properties of the node
intact, for constraint matching purposes.
Fixes#588
Currently, if the adapter node already has its ports configured when wireplumber
starts (For example a virtual node), the async _set_ports_format() call is never
completed because the node ports have not changed, stalling the event stack.
This fixes the issue by checking the Props param after configuring the ports,
and completes the task if there is one pending and the node already has ports
available.
Fixes#527
Syslog calls this level "notice" and I prefer it because we use it
to display significant messages that are not warnings, but they
are not really "standard", as GLib wants them to be. There is nothing
"standard" about log messages in general.
Also, make these notice messages be enabled at debug level 2, together
with warnings. The default log.level is 2 and it is a good idea to show
notices by default too.
Finally, show them in the log with "N" and also change criticals to be
shown with "E", meaning "error"... Then promote G_LOG_LEVEL_ERROR
messages to be shown with "F", meaning "fatal", because in fact these
messages are always fatal and always call abort(). Still, keep the term
"critical" in the functions to make sure that whoever uses them is aware
that this level is only for critical conditions and not suitable to
display any kind of error.
The intention is to make checks for enabled log topics faster.
Every topic has its own structure that is statically defined in the file
where the logs are printed from. The structure is initialized transparently
when it is first used and it contains all the log level flags for the levels
that this topic should print messages. It is then checked on the wp_log()
macro before printing the message.
Topics from SPA/PipeWire are also handled natively, so messages are printed
directly without checking if the topic is enabled, since the PipeWire and SPA
macros do the checking themselves.
Messages coming from GLib are checked inside the handler.
An internal WpLogFields object is used to manage the state of each log
message, populating all the fields appropriately from the place they
are coming from (wp_log, spa_log, glib log), formatting the message and
then printing it. For printing to the journald, we still use the glib
message handler, converting all the needed fields to GLogField on demand.
That message handler does not do any checks for the topic or the level, so
we can just call it to send the message.
This change completely refactors the way components are loaded in wireplumber:
- The module_init() function must return a GObject now. This object is either
a WpPlugin or a WpSiFactory in the current modules.
- When the component loader initializes a module, it automatically registers
the WpPlugin or WpSiFactory with their respective methods. There is no need
to register the WpPlugin or WpSiFactory in the module now.
- The wp_core_load_component() API has been refactored to be asynchronows. This
allows the component loader to automatically activate WpPlugin objects, and
therefore allows the application to directly get the WpPlugin without having
to find it. This simplifies a lot of things.
- The 'ifexists' and 'nofail' component flags now work even if the respective
WpPlugin could not be activated.
- The code that loads components in main.c has also been simplified a lot,
and the option to load dangling components has also been removed.
Some nodes can omit the format/rate/channels to indicate that they can
deal with all possibilities and adapt to what they are linked to.
See pipewire#876
"unpositioned" nodes don't have a specific channel layout
(ex the Pro Audio profile); in that case, we should not try to copy this
undefined layout to the peer node, we need to configure both in their
preferred layout and do a best-effort linking of ports.
We can use spa_format_audio_raw_parse() because we need to parse
the samplerate here optionally.
Many virtual sinks will not set a rate, which makes parsing fail and
the node will end up with the default channels/positions, which is
most likely wrong.
Fixes channel positions for virtual sinks.
* also respect node.exclusive
* also send error to clients if a target node was not found
or if it is not possible to link to it
* also allow dont-reconnect nodes to be handled normally
the first time they appear, until they are linked once
* find out if the node supports iec958 or dsd
* use spa_pod_fixate() to get the default format of the node, like p-m-s does
* use _enum_formats_sync() to utilize the param cache, assuming that the node
always has all features enabled (which is always the case when coming from
an ObjectManager that was defined in Lua)
* simplify the activation / deactivation procedure
* configure nodes with unpositioned channels early, as if they were devices
* populate most session item properties from create-item.lua to keep
things more compact and readable
* use a standard naming scheme for the session item properties
* use session item properties instead of node properties in policy-node.lua
* improve policy-node's performance by converting the properties dictionary
less times for each session item
* refactor some policy logic and make things slighly more readable
* change the accepted values for 'context' in wp_si_linkable_get_ports();
use "input" and "output" to keep things clear, because the previous use
of NULL and "reverse" were implying that a node has only one "standard"
direction, but this is complicated for sinks w/ monitors and duplex nodes
* allow using monitors (which are Audio/Sink nodes in fact) as sources
* treat Audio/Duplex nodes as sinks, like p-m-s does
* respect the "stream.capture.sink" property of streams
Fixes#66
If 'disable.dsp' is set to TRUE, the adapter will always be configured in
passthrough mode. Then, if 2 adapters want to be linked with different formats,
the si-standard-link session item will automatically reconfigure one of the
adapters in convert mode so that the format matches.
This is needed so that the async callback passed to _set_ports_format is called
when changing the adapter's port format and the port has not been removed or
added. Eg: when configuring the adapter in S16 convert mode to S32 convert mode.
Calling sync() in loops may end up looping forever if the operating
system's scheduling allows pipewire to reply to this sync() before
wireplumber's event loop has a chance to become idle.
In order for the new ports to appear, the object manager that monitors
them needs to emit "objects-changed", which only ever happens in an
idle callback. If the reply to sync() arrives before the idle callback,
it gets prioritized and processed, which causes another sync(), and so on...
Since setting PortFormat in the adapter always changes the ports,
watching for "ports-changed" feels like a better solution. Still,
there is more room for improvement.
When the autoconnect property is not available or false, we must
simply configure the stream in its default channel map and leave it
unconnected.
Fixes pw-cat -p --target=0 ..
Don't-remix streams don't change their channel mapping based on
the device they connect to but are always configured with their
default channel mapping, just like devices.
This fixes the gnome-control-center channel test.
This patch also removes the 1 channel -> many patch because it is
not the right thing to do in this case and should be implemented
with some proper channel mapping patch later.
