wayland_server/display.rs
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
use std::{
os::unix::io::{AsFd, AsRawFd, BorrowedFd},
os::unix::net::UnixStream,
sync::Arc,
};
use wayland_backend::{
protocol::ObjectInfo,
server::{Backend, ClientData, GlobalId, Handle, InitError, InvalidId, ObjectId},
};
use crate::{
global::{GlobalData, GlobalDispatch},
Client, Resource,
};
/// The Wayland display
///
/// This struct is the core of your Wayland compositor. You'll use it in your event loop to drive the
/// protocol processing of all your clients. All other interactions with the protocol itself are done
/// through the [`DisplayHandle`] struct, on which the `State` type parameter is erased for convenience.
///
/// ## Usage
///
/// The main loop of a Wayland compositor generally needs to wait on several sources of events, using
/// tools like `epoll` (on Linux). The Wayland display can be integrated in this mechanism by getting the
/// file descriptor as from [`.backend()`][Self::backend()][`.poll_fd()`][Backend::poll_fd()] and invoking
/// the [`dispatch_clients()`][Self::dispatch_clients()] method whenever it becomes readable.
///
/// To ensure all clients receive the events your compositor sends them, you also need to regularly invoke
/// the [`flush_clients()`][Self::flush_clients()] method, which will write the outgoing buffers into the
/// sockets.
#[derive(Debug)]
pub struct Display<State: 'static> {
backend: Backend<State>,
}
impl<State: 'static> Display<State> {
/// Create a new Wayland display
///
/// Can only fail if both the `server_system` and `dlopen` features of `wayland-backend` were enabled,
/// and the `libwayland-server.so` library could not be found.
pub fn new() -> Result<Display<State>, InitError> {
Ok(Display { backend: Backend::new()? })
}
/// Retrieve a [`DisplayHandle`] for this [`Display`].
///
/// This is the type with which all of your interactions with the Wayland protocol are done.
pub fn handle(&self) -> DisplayHandle {
DisplayHandle { handle: self.backend.handle() }
}
/// Dispatch all requests received from clients to their respective callbacks.
///
/// The `state` argument is the main state of your compositor, which will be accessible from most of your
/// callbacks.
pub fn dispatch_clients(&mut self, state: &mut State) -> std::io::Result<usize> {
self.backend.dispatch_all_clients(state)
}
/// Flush outgoing buffers into their respective sockets.
pub fn flush_clients(&mut self) -> std::io::Result<()> {
self.backend.flush(None)
}
/// Access the underlying [`Backend`] of this [`Display`]
pub fn backend(&mut self) -> &mut Backend<State> {
&mut self.backend
}
}
impl<State> AsFd for Display<State> {
/// Provides fd from [`Backend::poll_fd`] for polling.
fn as_fd(&self) -> BorrowedFd<'_> {
self.backend.poll_fd()
}
}
/// A handle to the Wayland display
///
/// A display handle may be constructed from a [`Handle`] using it's [`From`] implementation.
#[derive(Clone)]
pub struct DisplayHandle {
pub(crate) handle: Handle,
}
impl std::fmt::Debug for DisplayHandle {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
f.debug_struct("DisplayHandle").finish_non_exhaustive()
}
}
impl DisplayHandle {
/// Returns the underlying [`Handle`] from `wayland-backend`.
pub fn backend_handle(&self) -> Handle {
self.handle.clone()
}
/// Insert a new client in your [`Display`]
///
/// This client will be associated with the provided [`ClientData`], that you can then retrieve from
/// it via [`Client::get_data()`], and its requests will be processed by the [`Display`] and your
/// callbacks.
pub fn insert_client(
&mut self,
stream: UnixStream,
data: Arc<dyn ClientData>,
) -> std::io::Result<Client> {
let id = self.handle.insert_client(stream, data.clone())?;
Ok(Client { id, data })
}
/// Retrieve the [`Client`] which owns the object represented by the given ID
pub fn get_client(&self, id: ObjectId) -> Result<Client, InvalidId> {
let client_id = self.handle.get_client(id)?;
Client::from_id(self, client_id)
}
/// Create a new protocol global
///
/// This global will be advertized to clients through the `wl_registry` according to the rules
/// defined by your [`GlobalDispatch`] implementation for the given interface. Whenever a client
/// binds this global, the associated [`GlobalDispatch::bind()`] method will be invoked on your
/// `State`.
pub fn create_global<State, I: Resource + 'static, U: Send + Sync + 'static>(
&self,
version: u32,
data: U,
) -> GlobalId
where
State: GlobalDispatch<I, U> + 'static,
{
self.handle.create_global::<State>(
I::interface(),
version,
Arc::new(GlobalData { data, _types: std::marker::PhantomData }),
)
}
/// Disable this global
///
/// Clients will be notified of the global removal, and it will not be advertized to new clients. However
/// the state associated with this global is not freed, so clients which already know about it can still
/// bind it.
pub fn disable_global<State: 'static>(&self, id: GlobalId) {
self.handle.disable_global::<State>(id)
}
/// Remove this global
///
/// Clients will be notified of the global removal if it was not already disabled. The state associated
/// with this global is freed, meaning clients trying to bind it will receive a protocol error.
///
/// When removing a global, it is recommended to first disable it using
/// [`disable_global()`][Self::disable_global()] to allow some time for clients to register that
/// the global is getting removed, to avoid a race where a client would be killed because it bound a global
/// at the same as the server decided to remove it. After the global has been disabled for some time (like
/// a few seconds) it should be safe to actually remove it.
pub fn remove_global<State: 'static>(&self, id: GlobalId) {
self.handle.remove_global::<State>(id)
}
/// Access the protocol information for a Wayland object
///
/// Returns an error if the object is no longer valid.
pub fn object_info(&self, id: ObjectId) -> Result<ObjectInfo, InvalidId> {
self.handle.object_info(id)
}
/// Send an event to given Wayland object
///
/// This is intended to be a low-level method. You can alternatively use the methods on the
/// type representing your object, or [`Resource::send_event()`], which may be more convenient.
pub fn send_event<I: Resource>(
&self,
resource: &I,
event: I::Event<'_>,
) -> Result<(), InvalidId> {
let msg = resource.write_event(self, event)?;
let msg = msg.map_fd(|fd| fd.as_raw_fd());
self.handle.send_event(msg)
}
/// Trigger a protocol error on this object
///
/// This is intended to be a low-level method. See [`Resource::post_error()`], for a more convenient
/// method.
pub fn post_error<I: Resource>(&self, resource: &I, code: u32, error: String) {
self.handle.post_error(resource.id(), code, std::ffi::CString::new(error).unwrap())
}
/// Access the object data associated with this object
///
/// This is intended to be a low-level method. See [`Resource::object_data()`], for a more convenient
/// method.
pub fn get_object_data(
&self,
id: ObjectId,
) -> Result<Arc<dyn std::any::Any + Send + Sync + 'static>, InvalidId> {
self.handle.get_object_data_any(id)
}
/// Flush outgoing buffers into their respective sockets.
pub fn flush_clients(&mut self) -> std::io::Result<()> {
self.handle.flush(None)
}
}
impl From<Handle> for DisplayHandle {
/// Creates a [`DisplayHandle`] using a [`Handle`] from `wayland-backend`.
fn from(handle: Handle) -> Self {
Self { handle }
}
}