Trait timely::dataflow::operators::binary::Binary [−] [src]

pub trait Binary<G: Scope, D1: Data> {
    fn binary_stream<D2: Data, D3: Data, L: FnMut(&mut InputHandle<G::Timestamp, D1>, &mut InputHandle<G::Timestamp, D2>, &mut OutputHandle<G::Timestamp, D3, Tee<G::Timestamp, D3>>) + 'static, P1: ParallelizationContract<G::Timestamp, D1>, P2: ParallelizationContract<G::Timestamp, D2>>(&self, &Stream<G, D2>, pact1: P1, pact2: P2, name: &str, logic: L) -> Stream<G, D3>;
    fn binary_notify<D2: Data, D3: Data, L: FnMut(&mut InputHandle<G::Timestamp, D1>, &mut InputHandle<G::Timestamp, D2>, &mut OutputHandle<G::Timestamp, D3, Tee<G::Timestamp, D3>>, &mut Notificator<G::Timestamp>) + 'static, P1: ParallelizationContract<G::Timestamp, D1>, P2: ParallelizationContract<G::Timestamp, D2>>(&self, &Stream<G, D2>, pact1: P1, pact2: P2, name: &str, notify: Vec<G::Timestamp>, logic: L) -> Stream<G, D3>;
}

Methods to construct generic streaming and blocking binary operators.

Required Methods

fn binary_stream<D2: Data, D3: Data, L: FnMut(&mut InputHandle<G::Timestamp, D1>, &mut InputHandle<G::Timestamp, D2>, &mut OutputHandle<G::Timestamp, D3, Tee<G::Timestamp, D3>>) + 'static, P1: ParallelizationContract<G::Timestamp, D1>, P2: ParallelizationContract<G::Timestamp, D2>>(&self, &Stream<G, D2>, pact1: P1, pact2: P2, name: &str, logic: L) -> Stream<G, D3>

Creates a new dataflow operator that partitions each of its input stream by a parallelization strategy pact, and repeatedly invokes logic which can read from the input streams and write to the output stream.

Examples

use timely::dataflow::operators::{ToStream, Binary};
use timely::dataflow::channels::pact::Pipeline;

timely::example(|scope| {
    let stream1 = (0..10).to_stream(scope);
    let stream2 = (0..10).to_stream(scope);

    stream1.binary_stream(&stream2, Pipeline, Pipeline, "example", |input1, input2, output| {
        while let Some((time, data)) = input1.next() {
            output.session(&time).give_content(data);
        }
        while let Some((time, data)) = input2.next() {
            output.session(&time).give_content(data);
        }
    });
});

fn binary_notify<D2: Data, D3: Data, L: FnMut(&mut InputHandle<G::Timestamp, D1>, &mut InputHandle<G::Timestamp, D2>, &mut OutputHandle<G::Timestamp, D3, Tee<G::Timestamp, D3>>, &mut Notificator<G::Timestamp>) + 'static, P1: ParallelizationContract<G::Timestamp, D1>, P2: ParallelizationContract<G::Timestamp, D2>>(&self, &Stream<G, D2>, pact1: P1, pact2: P2, name: &str, notify: Vec<G::Timestamp>, logic: L) -> Stream<G, D3>

Creates a new dataflow operator that partitions its input stream by a parallelization strategy pact, and repeatedly invokes logic which can read from the input streams, write to the output stream, and request and receive notifications. The method also requires a vector of the initial notifications the operator requires (commonly none).

Examples

use timely::dataflow::operators::{ToStream, Binary};
use timely::dataflow::channels::pact::Pipeline;

timely::example(|scope| {
    let stream1 = (0..10).to_stream(scope);
    let stream2 = (0..10).to_stream(scope);

    stream1.binary_notify(&stream2, Pipeline, Pipeline, "example", Vec::new(), |input1, input2, output, notificator| {
        while let Some((time, data)) = input1.next() {
            output.session(&time).give_content(data);
            notificator.notify_at(time);
        }
        while let Some((time, data)) = input2.next() {
            output.session(&time).give_content(data);
            notificator.notify_at(time);
        }
        while let Some((time, count)) = notificator.next() {
            println!("done with time: {:?}", time.time());
        }
    });
});

Implementors