Struct re_smart_channel::receiver::Receiver

pub struct Receiver<T: Send> {
    pub(crate) rx: Receiver<SmartMessage<T>>,
    stats: Arc<SharedStats>,
    pub(crate) source: Arc<SmartChannelSource>,
    connected: AtomicBool,
}

Fields

rx: Receiver<SmartMessage<T>>

stats: Arc<SharedStats>

source: Arc<SmartChannelSource>

connected: AtomicBool

Implementations

impl<T: Send> Receiver<T>

pub(crate) fn new( rx: Receiver<SmartMessage<T>>, stats: Arc<SharedStats>, source: Arc<SmartChannelSource> ) -> Self

pub fn is_connected(&self) -> bool

Are we still connected?

Once false, we will never be connected again: the source has run dry.

This is only updated once one of the receive methods fails.

pub fn recv(&self) -> Result<SmartMessage<T>, RecvError>

pub fn try_recv(&self) -> Result<SmartMessage<T>, TryRecvError>

pub fn recv_timeout( &self, timeout: Duration ) -> Result<SmartMessage<T>, RecvTimeoutError>

pub fn recv_with_send_time(&self) -> Result<SmartMessage<T>, RecvError>

Receives without registering the latency.

This is for use with crate::Sender::send_at when chaining to another channel created with Self::chained_channel.

pub fn source(&self) -> &SmartChannelSource

Where is the data coming from?

pub fn is_empty(&self) -> bool

Is the channel currently empty of messages?

pub fn len(&self) -> usize

Number of messages in the channel right now.

pub fn latency_ns(&self) -> u64

Latest known latency from sending a message to receiving it, it nanoseconds.

pub fn latency_sec(&self) -> f32

Latest known latency from sending a message to receiving it, in seconds

pub fn chained_channel(&self) -> (Sender<T>, Self)

Create a new channel that use the same stats as this one.

This means both channels will see the same latency numbers.

Care must be taken to use Self::recv_with_send_time and crate::Sender::send_at. This is a very leaky abstraction, and it would be nice with a refactor.