Never free a proxy without the application doing a pw_proxy_destroy.
It's hard to use when proxies are freed randomly when the server
removes the ids. You have to add destroy notify to all proxies and
deal with the arbirary order in which proxies can be freed.
Instead notify the client of the remove and let it destroy the
proxies itself in the right order. This is in line with how wayland
handles proxies.
A pw_proxy_destroy() will now send a destroy to the server and mark
the proxy as a zombie, waiting for the remove_id confirmation and
then destroy the proxy.
A server remove_id will mark the proxy as removed and emits the
removed event. The app should then pw_proxy_destroy the proxy
to free it.
Leaks all proxies in the session manager because cleanup now needs
to be handled by the app correctly.
This is more in line with wayland and it allows us to create new
interfaces in modules without having to add anything to the type
enum. It also removes some lookups to map type_id to readable
name in debug.
Add method to set and get typed objects from the context. This can
be used to pass around context objects between modules without having
to register them.
Don't pass pw_properties around when we simply need to pass around
config info, only use pw_properties when used to construct an object
that keeps the properties.
The proxy API is the one that we would like to expose for applications
and the other API is used internally when implementing modules or
factories.
The current pw_core object is really a context for all objects so
name it that way. It also makes it possible to rename pw_core_proxy
to pw_proxy later.
The pw_remote object is really a wrapper around the pw_core_proxy.
The events it emits are also available in the core proxy and are
generally awkward to use.
With some clever new pw_core_proxy_* methods and a pw_core_connect
to create the core_proxy, we can convert all code away from pw_remote.
This is a first step in this conversion, using the pw_remote behind
the scenes. It leaks into some places because it really needs to become
its own struct in a next step.
Let the core load a set of default modules.
Add a key to control what default set to load, falling back to a
reasonable set. Make the daemon not load any set and rely on the
config script to load modules.
media-session: Wait for devices to appear and be bound before setting
the profile on the device. Then wait for all the nodes to appear
before attempting the create endpoints on the device.
Let the session manager monitor device objects as well.
Make the alsa-endpoint monitor a separate service instead of letting
the alsa-monitor call it directly. This means that it listens for
device objects and then tries to configure the endpoints when the
device profile is set to active. This does not work yet because we
can't link the nodes to the device yet because there is no way to know
what the global id is of the device we created.
Make sure implementations of objects run in a separate remote connection
because the main remote connection might block while waiting for a
return value from the implementation.
Trigger an object update after all object info is collected. We do this
by triggering a roundtrip after receiving the info event. When we get
the reply, we can assume all info is flushed. This includes the
parameters that we received.
Add a method to create a node and track the proxy. We can then
use this to track the params on a node.
Don't listen end enumerate the params in the monitors bbut let the
session do that. Use the collected params from the session to create
endpoints and streams.
Always activate the default stream on alsa nodes to create the
ports and have something we can present to apps.
Create an endpoint link object when linking endpoints. Keep track
of the links in the endpoint_link and cleanup when they are all
gone.
Improve properties on session objects.
Create a helper in the session manager to link all ports between
2 nodes.
Make the output endpoint call the input endpoint create_link, passing
the node or ports to link to.
Make the input endpoint complete the link by calling the session
helper to link all ports.
Remove the node policy, we're only using a policy for the endpoints
now.
Move all the media-session object monitoring into one place and
provide an API to get to the session objects.
Make API to add module specific info to objects.
Add methods to export and create objects in the session. This should
make it possible to link proxy to implementation and avoid a server
roundtrip in some cases.
pulseaudio card is mapped to device
pulseaudio sink/source is mapped to an endpoint
prepare to map streams to card profiles
Add Route param to implement the endpoint routing later (ports)
Create an alsa endpoint for each device
Create one stream for each endpoint (Playback/Capture)
Implement create_link on the endpoint. The idea is to call
create link on the peer endpoint to complete the link. Remove
create_link on the session.
Add stream-monitor to turn pw_stream nodes into endpoints
Add a policy manager that tries to link endpoints
Use enum pw_direction for the endpoint direction. We can use the
media_class to determine if this is a pw_stream or not but it should
not really matter, you can link any output to any input.
Add autoconnect property for endpoints to make the policy connect.
Separate the session manager in a monitor and policy part.
The monitor manages the devices and endpoints.
The policy watches the nodes/ports/clients and applies the policy
of linking them.
Because both now have a separate connection, we can remove some
hacks in the protocol. When a remote was both the implementer and
user of an object we could get in a deadlock when the user was
blocked waiting and the implementator was blocked sending a reply.
We used to un-busy a client when it was expecting a reply from a
ping or sync for this reason.
Add and use some more keys for the endpoints and streams.
Metadata allows apps to attach properties to objects that can be
read by other apps.
Not complete yet, properties should be removed when the object is
removed.
