[−][src]Trait futures::future::Future

pub trait Future {
    type Item;
    type Error;
    fn poll(&mut self) -> Poll<Self::Item, Self::Error>;

    fn wait(self) -> Result<Self::Item, Self::Error>
    where
        Self: Sized,
    { ... }

    fn map<F, U>(self, f: F) -> Map<Self, F>
    where
        F: FnOnce(Self::Item) -> U,
        Self: Sized,
    { ... }

    fn map_err<F, E>(self, f: F) -> MapErr<Self, F>
    where
        F: FnOnce(Self::Error) -> E,
        Self: Sized,
    { ... }

    fn from_err<E: From<Self::Error>>(self) -> FromErr<Self, E>
    where
        Self: Sized,
    { ... }

    fn then<F, B>(self, f: F) -> Then<Self, B, F>
    where
        F: FnOnce(Result<Self::Item, Self::Error>) -> B,
        B: IntoFuture,
        Self: Sized,
    { ... }

    fn and_then<F, B>(self, f: F) -> AndThen<Self, B, F>
    where
        F: FnOnce(Self::Item) -> B,
        B: IntoFuture<Error = Self::Error>,
        Self: Sized,
    { ... }

    fn or_else<F, B>(self, f: F) -> OrElse<Self, B, F>
    where
        F: FnOnce(Self::Error) -> B,
        B: IntoFuture<Item = Self::Item>,
        Self: Sized,
    { ... }

    fn select<B>(self, other: B) -> Select<Self, B::Future>
    where
        B: IntoFuture<Item = Self::Item, Error = Self::Error>,
        Self: Sized,
    { ... }

    fn select2<B>(self, other: B) -> Select2<Self, B::Future>
    where
        B: IntoFuture,
        Self: Sized,
    { ... }

    fn join<B>(self, other: B) -> Join<Self, B::Future>
    where
        B: IntoFuture<Error = Self::Error>,
        Self: Sized,
    { ... }

    fn join3<B, C>(self, b: B, c: C) -> Join3<Self, B::Future, C::Future>
    where
        B: IntoFuture<Error = Self::Error>,
        C: IntoFuture<Error = Self::Error>,
        Self: Sized,
    { ... }

    fn join4<B, C, D>(
        self, 
        b: B, 
        c: C, 
        d: D
    ) -> Join4<Self, B::Future, C::Future, D::Future>
    where
        B: IntoFuture<Error = Self::Error>,
        C: IntoFuture<Error = Self::Error>,
        D: IntoFuture<Error = Self::Error>,
        Self: Sized,
    { ... }

    fn join5<B, C, D, E>(
        self, 
        b: B, 
        c: C, 
        d: D, 
        e: E
    ) -> Join5<Self, B::Future, C::Future, D::Future, E::Future>
    where
        B: IntoFuture<Error = Self::Error>,
        C: IntoFuture<Error = Self::Error>,
        D: IntoFuture<Error = Self::Error>,
        E: IntoFuture<Error = Self::Error>,
        Self: Sized,
    { ... }

    fn into_stream(self) -> IntoStream<Self>
    where
        Self: Sized,
    { ... }

    fn flatten(self) -> Flatten<Self>
    where
        Self::Item: IntoFuture,
        <Self::Item as IntoFuture>::Error: From<Self::Error>,
        Self: Sized,
    { ... }

    fn flatten_stream(self) -> FlattenStream<Self>
    where
        Self::Item: Stream<Error = Self::Error>,
        Self: Sized,
    { ... }

    fn fuse(self) -> Fuse<Self>
    where
        Self: Sized,
    { ... }

    fn inspect<F>(self, f: F) -> Inspect<Self, F>
    where
        F: FnOnce(&Self::Item),
        Self: Sized,
    { ... }

    fn catch_unwind(self) -> CatchUnwind<Self>
    where
        Self: Sized + UnwindSafe,
    { ... }

    fn shared(self) -> Shared<Self>
    where
        Self: Sized,
    { ... }
}

Trait for types which are a placeholder of a value that may become available at some later point in time.

In addition to the documentation here you can also find more information about futures online at https://tokio.rs

Futures are used to provide a sentinel through which a value can be referenced. They crucially allow chaining and composing operations through consumption which allows expressing entire trees of computation as one sentinel value.

The ergonomics and implementation of the Future trait are very similar to the Iterator trait in that there is just one methods you need to implement, but you get a whole lot of others for free as a result.

The poll method

The core method of future, poll, is used to attempt to generate the value of a Future. This method does not block but is allowed to inform the caller that the value is not ready yet. Implementations of poll may themselves do work to generate the value, but it's guaranteed that this will never block the calling thread.

A key aspect of this method is that if the value is not yet available the current task is scheduled to receive a notification when it's later ready to be made available. This follows what's typically known as a "readiness" or "pull" model where values are pulled out of futures on demand, and otherwise a task is notified when a value might be ready to get pulled out.

The poll method is not intended to be called in general, but rather is typically called in the context of a "task" which drives a future to completion. For more information on this see the task module.

More information about the details of poll and the nitty-gritty of tasks can be found online at tokio.rs.

Combinators

Like iterators, futures provide a large number of combinators to work with futures to express computations in a much more natural method than scheduling a number of callbacks. For example the map method can change a Future<Item=T> to a Future<Item=U> or an and_then combinator could create a future after the first one is done and only be resolved when the second is done.

Combinators act very similarly to the methods on the Iterator trait itself or those on Option and Result. Like with iterators, the combinators are zero-cost and don't impose any extra layers of indirection you wouldn't otherwise have to write down.

More information about combinators can be found on tokio.rs.

Associated Types

type Item

The type of value that this future will resolved with if it is successful.

type Error

The type of error that this future will resolve with if it fails in a normal fashion.

Loading content...

Required methods

fn poll(&mut self) -> Poll<Self::Item, Self::Error>

Query this future to see if its value has become available, registering interest if it is not.

This function will check the internal state of the future and assess whether the value is ready to be produced. Implementers of this function should ensure that a call to this never blocks as event loops may not work properly otherwise.

When a future is not ready yet, the Async::NotReady value will be returned. In this situation the future will also register interest of the current task in the value being produced. This is done by calling task::park to retrieve a handle to the current Task. When the future is then ready to make progress (e.g. it should be polled again) the unpark method is called on the Task.

More information about the details of poll and the nitty-gritty of tasks can be found online at tokio.rs.

Runtime characteristics

This function, poll, is the primary method for 'making progress' within a tree of futures. For example this method will be called repeatedly as the internal state machine makes its various transitions. Executors are responsible for ensuring that this function is called in the right location (e.g. always on an I/O thread or not). Unless it is otherwise arranged to be so, it should be ensured that implementations of this function finish very quickly.

Returning quickly prevents unnecessarily clogging up threads and/or event loops while a poll function call, for example, takes up compute resources to perform some expensive computation. If it is known ahead of time that a call to poll may end up taking awhile, the work should be offloaded to a thread pool (or something similar) to ensure that poll can return quickly.

Note that the poll function is not called repeatedly in a loop for futures typically, but only whenever the future itself is ready. If you're familiar with the poll(2) or select(2) syscalls on Unix it's worth noting that futures typically do not suffer the same problems of "all wakeups must poll all events". Futures have enough support for only polling futures which cause a wakeup.

Return value

This function returns Async::NotReady if the future is not ready yet, Err if the future is finished but resolved to an error, or Async::Ready with the result of this future if it's finished successfully. Once a future has finished it is considered a contract error to continue polling the future.

If NotReady is returned, then the future will internally register interest in the value being produced for the current task (through task::park). In other words, the current task will receive a notification (through the unpark method) once the value is ready to be produced or the future can make progress.

Note that if NotReady is returned it only means that this task will receive a notification. Historical calls to poll with different tasks will not receive notifications. In other words, implementers of the Future trait need not store a queue of tasks to notify, but only the last task that called this method. Alternatively callers of this method can only rely on the most recent task which call poll being notified when a future is ready.

Panics

Once a future has completed (returned Ready or Err from poll), then any future calls to poll may panic, block forever, or otherwise cause wrong behavior. The Future trait itself provides no guarantees about the behavior of poll after a future has completed.

Callers who may call poll too many times may want to consider using the fuse adaptor which defines the behavior of poll, but comes with a little bit of extra cost.

Additionally, calls to poll must always be made from within the context of a task. If a current task is not set then this method will likely panic.

Errors

This future may have failed to finish the computation, in which case the Err variant will be returned with an appropriate payload of an error.

Loading content...

Provided methods

fn wait(self) -> Result<Self::Item, Self::Error> where
    Self: Sized, 

Block the current thread until this future is resolved.

This method will consume ownership of this future, driving it to completion via poll and blocking the current thread while it's waiting for the value to become available. Once the future is resolved the result of this future is returned.

Note: This method is not appropriate to call on event loops or similar I/O situations because it will prevent the event loop from making progress (this blocks the thread). This method should only be called when it's guaranteed that the blocking work associated with this future will be completed by another thread.

This method is only available when the use_std feature of this library is activated, and it is activated by default.

Panics

This function does not attempt to catch panics. If the poll function of this future panics, panics will be propagated to the caller.

fn map<F, U>(self, f: F) -> Map<Self, F> where
    F: FnOnce(Self::Item) -> U,
    Self: Sized, 

Map this future's result to a different type, returning a new future of the resulting type.

This function is similar to the Option::map or Iterator::map where it will change the type of the underlying future. This is useful to chain along a computation once a future has been resolved.

The closure provided will only be called if this future is resolved successfully. If this future returns an error, panics, or is dropped, then the closure provided will never be invoked.

Note that this function consumes the receiving future and returns a wrapped version of it, similar to the existing map methods in the standard library.

Examples

use futures::prelude::*;
use futures::future;

let future = future::ok::<u32, u32>(1);
let new_future = future.map(|x| x + 3);
assert_eq!(new_future.wait(), Ok(4));

Calling map on an errored Future has no effect:

use futures::prelude::*;
use futures::future;

let future = future::err::<u32, u32>(1);
let new_future = future.map(|x| x + 3);
assert_eq!(new_future.wait(), Err(1));

fn map_err<F, E>(self, f: F) -> MapErr<Self, F> where
    F: FnOnce(Self::Error) -> E,
    Self: Sized, 

Map this future's error to a different error, returning a new future.

This function is similar to the Result::map_err where it will change the error type of the underlying future. This is useful for example to ensure that futures have the same error type when used with combinators like select and join.

The closure provided will only be called if this future is resolved with an error. If this future returns a success, panics, or is dropped, then the closure provided will never be invoked.

Note that this function consumes the receiving future and returns a wrapped version of it.

Examples

use futures::future::*;

let future = err::<u32, u32>(1);
let new_future = future.map_err(|x| x + 3);
assert_eq!(new_future.wait(), Err(4));

Calling map_err on a successful Future has no effect:

use futures::future::*;

let future = ok::<u32, u32>(1);
let new_future = future.map_err(|x| x + 3);
assert_eq!(new_future.wait(), Ok(1));

fn from_err<E: From<Self::Error>>(self) -> FromErr<Self, E> where
    Self: Sized, 

Map this future's error to any error implementing From for this future's Error, returning a new future.

This function does for futures what try! does for Result, by letting the compiler infer the type of the resulting error. Just as map_err above, this is useful for example to ensure that futures have the same error type when used with combinators like select and join.

Note that this function consumes the receiving future and returns a wrapped version of it.

Examples

use futures::prelude::*;
use futures::future;

let future_with_err_u8 = future::err::<(), u8>(1);
let future_with_err_u32 = future_with_err_u8.from_err::<u32>();

fn then<F, B>(self, f: F) -> Then<Self, B, F> where
    F: FnOnce(Result<Self::Item, Self::Error>) -> B,
    B: IntoFuture,
    Self: Sized, 

Chain on a computation for when a future finished, passing the result of the future to the provided closure f.

This function can be used to ensure a computation runs regardless of the conclusion of the future. The closure provided will be yielded a Result once the future is complete.

The returned value of the closure must implement the IntoFuture trait and can represent some more work to be done before the composed future is finished. Note that the Result type implements the IntoFuture trait so it is possible to simply alter the Result yielded to the closure and return it.

If this future is dropped or panics then the closure f will not be run.

Note that this function consumes the receiving future and returns a wrapped version of it.

Examples

use futures::prelude::*;
use futures::future;

let future_of_1 = future::ok::<u32, u32>(1);
let future_of_4 = future_of_1.then(|x| {
    x.map(|y| y + 3)
});

let future_of_err_1 = future::err::<u32, u32>(1);
let future_of_4 = future_of_err_1.then(|x| {
    match x {
        Ok(_) => panic!("expected an error"),
        Err(y) => future::ok::<u32, u32>(y + 3),
    }
});

fn and_then<F, B>(self, f: F) -> AndThen<Self, B, F> where
    F: FnOnce(Self::Item) -> B,
    B: IntoFuture<Error = Self::Error>,
    Self: Sized, 

Execute another future after this one has resolved successfully.

This function can be used to chain two futures together and ensure that the final future isn't resolved until both have finished. The closure provided is yielded the successful result of this future and returns another value which can be converted into a future.

Note that because Result implements the IntoFuture trait this method can also be useful for chaining fallible and serial computations onto the end of one future.

If this future is dropped, panics, or completes with an error then the provided closure f is never called.

Note that this function consumes the receiving future and returns a wrapped version of it.

Examples

use futures::prelude::*;
use futures::future::{self, FutureResult};

let future_of_1 = future::ok::<u32, u32>(1);
let future_of_4 = future_of_1.and_then(|x| {
    Ok(x + 3)
});

let future_of_err_1 = future::err::<u32, u32>(1);
future_of_err_1.and_then(|_| -> FutureResult<u32, u32> {
    panic!("should not be called in case of an error");
});

fn or_else<F, B>(self, f: F) -> OrElse<Self, B, F> where
    F: FnOnce(Self::Error) -> B,
    B: IntoFuture<Item = Self::Item>,
    Self: Sized, 

Execute another future if this one resolves with an error.

Return a future that passes along this future's value if it succeeds, and otherwise passes the error to the closure f and waits for the future it returns. The closure may also simply return a value that can be converted into a future.

Note that because Result implements the IntoFuture trait this method can also be useful for chaining together fallback computations, where when one fails, the next is attempted.

If this future is dropped, panics, or completes successfully then the provided closure f is never called.

Note that this function consumes the receiving future and returns a wrapped version of it.

Examples

use futures::prelude::*;
use futures::future::{self, FutureResult};

let future_of_err_1 = future::err::<u32, u32>(1);
let future_of_4 = future_of_err_1.or_else(|x| -> Result<u32, u32> {
    Ok(x + 3)
});

let future_of_1 = future::ok::<u32, u32>(1);
future_of_1.or_else(|_| -> FutureResult<u32, u32> {
    panic!("should not be called in case of success");
});

fn select<B>(self, other: B) -> Select<Self, B::Future> where
    B: IntoFuture<Item = Self::Item, Error = Self::Error>,
    Self: Sized, 

Waits for either one of two futures to complete.

This function will return a new future which awaits for either this or the other future to complete. The returned future will finish with both the value resolved and a future representing the completion of the other work. Both futures must have the same item and error type.

Note that this function consumes the receiving futures and returns a wrapped version of them.

Examples

use futures::prelude::*;
use futures::future;
use std::thread;
use std::time;

let future1 = future::lazy(|| {
    thread::sleep(time::Duration::from_secs(5));
    future::ok::<char, ()>('a')
});

let future2 = future::lazy(|| {
    thread::sleep(time::Duration::from_secs(3));
    future::ok::<char, ()>('b')
});

let (value, last_future) = future1.select(future2).wait().ok().unwrap();
assert_eq!(value, 'a');
assert_eq!(last_future.wait().unwrap(), 'b');

A poor-man's join implemented on top of select:

use futures::prelude::*;
use futures::future;

fn join<A>(a: A, b: A) -> Box<Future<Item=(u32, u32), Error=u32>>
    where A: Future<Item = u32, Error = u32> + 'static,
{
    Box::new(a.select(b).then(|res| -> Box<Future<Item=_, Error=_>> {
        match res {
            Ok((a, b)) => Box::new(b.map(move |b| (a, b))),
            Err((a, _)) => Box::new(future::err(a)),
        }
    }))
}

fn select2<B>(self, other: B) -> Select2<Self, B::Future> where
    B: IntoFuture,
    Self: Sized, 

Waits for either one of two differently-typed futures to complete.

This function will return a new future which awaits for either this or the other future to complete. The returned future will finish with both the value resolved and a future representing the completion of the other work.

Note that this function consumes the receiving futures and returns a wrapped version of them.

Also note that if both this and the second future have the same success/error type you can use the Either::split method to conveniently extract out the value at the end.

Examples

use futures::prelude::*;
use futures::future::{self, Either};

// A poor-man's join implemented on top of select2

fn join<A, B, E>(a: A, b: B) -> Box<Future<Item=(A::Item, B::Item), Error=E>>
    where A: Future<Error = E> + 'static,
          B: Future<Error = E> + 'static,
          E: 'static,
{
    Box::new(a.select2(b).then(|res| -> Box<Future<Item=_, Error=_>> {
        match res {
            Ok(Either::A((x, b))) => Box::new(b.map(move |y| (x, y))),
            Ok(Either::B((y, a))) => Box::new(a.map(move |x| (x, y))),
            Err(Either::A((e, _))) => Box::new(future::err(e)),
            Err(Either::B((e, _))) => Box::new(future::err(e)),
        }
    }))
}

fn join<B>(self, other: B) -> Join<Self, B::Future> where
    B: IntoFuture<Error = Self::Error>,
    Self: Sized, 

Joins the result of two futures, waiting for them both to complete.

This function will return a new future which awaits both this and the other future to complete. The returned future will finish with a tuple of both results.

Both futures must have the same error type, and if either finishes with an error then the other will be dropped and that error will be returned.

Note that this function consumes the receiving future and returns a wrapped version of it.

Examples

use futures::prelude::*;
use futures::future;

let a = future::ok::<u32, u32>(1);
let b = future::ok::<u32, u32>(2);
let pair = a.join(b);

assert_eq!(pair.wait(), Ok((1, 2)));

If one or both of the joined Futures is errored, the resulting Future will be errored:

use futures::prelude::*;
use futures::future;

let a = future::ok::<u32, u32>(1);
let b = future::err::<u32, u32>(2);
let pair = a.join(b);

assert_eq!(pair.wait(), Err(2));

fn join3<B, C>(self, b: B, c: C) -> Join3<Self, B::Future, C::Future> where
    B: IntoFuture<Error = Self::Error>,
    C: IntoFuture<Error = Self::Error>,
    Self: Sized, 

Same as join, but with more futures.

fn join4<B, C, D>(
    self,
    b: B,
    c: C,
    d: D
) -> Join4<Self, B::Future, C::Future, D::Future> where
    B: IntoFuture<Error = Self::Error>,
    C: IntoFuture<Error = Self::Error>,
    D: IntoFuture<Error = Self::Error>,
    Self: Sized, 

Same as join, but with more futures.

fn join5<B, C, D, E>(
    self,
    b: B,
    c: C,
    d: D,
    e: E
) -> Join5<Self, B::Future, C::Future, D::Future, E::Future> where
    B: IntoFuture<Error = Self::Error>,
    C: IntoFuture<Error = Self::Error>,
    D: IntoFuture<Error = Self::Error>,
    E: IntoFuture<Error = Self::Error>,
    Self: Sized, 

Same as join, but with more futures.

fn into_stream(self) -> IntoStream<Self> where
    Self: Sized, 

Convert this future into a single element stream.

The returned stream contains single success if this future resolves to success or single error if this future resolves into error.

Examples

use futures::prelude::*;
use futures::future;

let future = future::ok::<_, bool>(17);
let mut stream = future.into_stream();
assert_eq!(Ok(Async::Ready(Some(17))), stream.poll());
assert_eq!(Ok(Async::Ready(None)), stream.poll());

let future = future::err::<bool, _>(19);
let mut stream = future.into_stream();
assert_eq!(Err(19), stream.poll());
assert_eq!(Ok(Async::Ready(None)), stream.poll());

fn flatten(self) -> Flatten<Self> where
    Self::Item: IntoFuture,
    <Self::Item as IntoFuture>::Error: From<Self::Error>,
    Self: Sized, 

Flatten the execution of this future when the successful result of this future is itself another future.

This can be useful when combining futures together to flatten the computation out the final result. This method can only be called when the successful result of this future itself implements the IntoFuture trait and the error can be created from this future's error type.

This method is roughly equivalent to self.and_then(|x| x).

Note that this function consumes the receiving future and returns a wrapped version of it.

Examples

use futures::prelude::*;
use futures::future;

let nested_future = future::ok::<_, u32>(future::ok::<u32, u32>(1));
let future = nested_future.flatten();
assert_eq!(future.wait(), Ok(1));

Calling flatten on an errored Future, or if the inner Future is errored, will result in an errored Future:

use futures::prelude::*;
use futures::future;

let nested_future = future::ok::<_, u32>(future::err::<u32, u32>(1));
let future = nested_future.flatten();
assert_eq!(future.wait(), Err(1));

fn flatten_stream(self) -> FlattenStream<Self> where
    Self::Item: Stream<Error = Self::Error>,
    Self: Sized, 

Flatten the execution of this future when the successful result of this future is a stream.

This can be useful when stream initialization is deferred, and it is convenient to work with that stream as if stream was available at the call site.

Note that this function consumes this future and returns a wrapped version of it.

Examples

use futures::prelude::*;
use futures::future;
use futures::stream;

let stream_items = vec![17, 18, 19];
let future_of_a_stream = future::ok::<_, bool>(stream::iter_ok(stream_items));

let stream = future_of_a_stream.flatten_stream();

let mut iter = stream.wait();
assert_eq!(Ok(17), iter.next().unwrap());
assert_eq!(Ok(18), iter.next().unwrap());
assert_eq!(Ok(19), iter.next().unwrap());
assert_eq!(None, iter.next());

fn fuse(self) -> Fuse<Self> where
    Self: Sized, 

Fuse a future such that poll will never again be called once it has completed.

Currently once a future has returned Ready or Err from poll any further calls could exhibit bad behavior such as blocking forever, panicking, never returning, etc. If it is known that poll may be called too often then this method can be used to ensure that it has defined semantics.

Once a future has been fused and it returns a completion from poll, then it will forever return NotReady from poll again (never resolve). This, unlike the trait's poll method, is guaranteed.

This combinator will drop this future as soon as it's been completed to ensure resources are reclaimed as soon as possible.

Examples

use futures::prelude::*;
use futures::future;

let mut future = future::ok::<i32, u32>(2);
assert_eq!(future.poll(), Ok(Async::Ready(2)));

// Normally, a call such as this would panic:
//future.poll();

// This, however, is guaranteed to not panic
let mut future = future::ok::<i32, u32>(2).fuse();
assert_eq!(future.poll(), Ok(Async::Ready(2)));
assert_eq!(future.poll(), Ok(Async::NotReady));

fn inspect<F>(self, f: F) -> Inspect<Self, F> where
    F: FnOnce(&Self::Item),
    Self: Sized, 

Do something with the item of a future, passing it on.

When using futures, you'll often chain several of them together. While working on such code, you might want to check out what's happening at various parts in the pipeline. To do that, insert a call to inspect().

Examples

use futures::prelude::*;
use futures::future;

let future = future::ok::<u32, u32>(1);
let new_future = future.inspect(|&x| println!("about to resolve: {}", x));
assert_eq!(new_future.wait(), Ok(1));

fn catch_unwind(self) -> CatchUnwind<Self> where
    Self: Sized + UnwindSafe, 

Catches unwinding panics while polling the future.

In general, panics within a future can propagate all the way out to the task level. This combinator makes it possible to halt unwinding within the future itself. It's most commonly used within task executors. It's not recommended to use this for error handling.

Note that this method requires the UnwindSafe bound from the standard library. This isn't always applied automatically, and the standard library provides an AssertUnwindSafe wrapper type to apply it after-the fact. To assist using this method, the Future trait is also implemented for AssertUnwindSafe<F> where F implements Future.

This method is only available when the use_std feature of this library is activated, and it is activated by default.

Examples

use futures::prelude::*;
use futures::future::{self, FutureResult};

let mut future = future::ok::<i32, u32>(2);
assert!(future.catch_unwind().wait().is_ok());

let mut future = future::lazy(|| -> FutureResult<i32, u32> {
    panic!();
    future::ok::<i32, u32>(2)
});
assert!(future.catch_unwind().wait().is_err());

fn shared(self) -> Shared<Self> where
    Self: Sized, 

Create a cloneable handle to this future where all handles will resolve to the same result.

The shared() method provides a method to convert any future into a cloneable future. It enables a future to be polled by multiple threads.

The returned Shared future resolves successfully with SharedItem<Self::Item> or erroneously with SharedError<Self::Error>. Both SharedItem and SharedError implements Deref to allow shared access to the underlying result. Ownership of Self::Item and Self::Error cannot currently be reclaimed.

This method is only available when the use_std feature of this library is activated, and it is activated by default.

Examples

use futures::prelude::*;
use futures::future;

let future = future::ok::<_, bool>(6);
let shared1 = future.shared();
let shared2 = shared1.clone();
assert_eq!(6, *shared1.wait().unwrap());
assert_eq!(6, *shared2.wait().unwrap());
use std::thread;
use futures::prelude::*;
use futures::future;

let future = future::ok::<_, bool>(6);
let shared1 = future.shared();
let shared2 = shared1.clone();
let join_handle = thread::spawn(move || {
    assert_eq!(6, *shared2.wait().unwrap());
});
assert_eq!(6, *shared1.wait().unwrap());
join_handle.join().unwrap();
Loading content...

Implementations on Foreign Types

impl<F, T, E> Future for Option<F> where
    F: Future<Item = T, Error = E>, [src]

type Item = Option<T>

type Error = E

impl<F: Future> Future for AssertUnwindSafe<F>[src]

type Item = F::Item

type Error = F::Error

impl<F: ?Sized + Future> Future for Box<F>[src]

type Item = F::Item

type Error = F::Error

impl<'a, F: ?Sized + Future> Future for &'a mut F[src]

type Item = F::Item

type Error = F::Error

Loading content...

Implementors

impl<A> Future for Flatten<A> where
    A: Future,
    A::Item: IntoFuture,
    <<A as Future>::Item as IntoFuture>::Error: From<<A as Future>::Error>, [src]

type Item = <<A as Future>::Item as IntoFuture>::Item

type Error = <<A as Future>::Item as IntoFuture>::Error

impl<A> Future for SelectAll<A> where
    A: Future, [src]

type Item = (A::Item, usize, Vec<A>)

type Error = (A::Error, usize, Vec<A>)

impl<A> Future for SelectOk<A> where
    A: Future, [src]

type Item = (A::Item, Vec<A>)

type Error = A::Error

impl<A, B> Future for Either<A, B> where
    A: Future,
    B: Future<Item = A::Item, Error = A::Error>, [src]

type Item = A::Item

type Error = A::Error

impl<A, B> Future for Join<A, B> where
    A: Future,
    B: Future<Error = A::Error>, [src]

type Item = (A::Item, B::Item)

type Error = A::Error

impl<A, B> Future for Select<A, B> where
    A: Future,
    B: Future<Item = A::Item, Error = A::Error>, [src]

type Item = (A::Item, SelectNext<A, B>)

type Error = (A::Error, SelectNext<A, B>)

impl<A, B> Future for Select2<A, B> where
    A: Future,
    B: Future, [src]

type Item = Either<(A::Item, B), (B::Item, A)>

type Error = Either<(A::Error, B), (B::Error, A)>

impl<A, B> Future for SelectNext<A, B> where
    A: Future,
    B: Future<Item = A::Item, Error = A::Error>, [src]

type Item = A::Item

type Error = A::Error

impl<A, B, C> Future for Join3<A, B, C> where
    A: Future,
    B: Future<Error = A::Error>,
    C: Future<Error = A::Error>, [src]

type Item = (A::Item, B::Item, C::Item)

type Error = A::Error

impl<A, B, C, D> Future for Join4<A, B, C, D> where
    A: Future,
    B: Future<Error = A::Error>,
    C: Future<Error = A::Error>,
    D: Future<Error = A::Error>, [src]

type Item = (A::Item, B::Item, C::Item, D::Item)

type Error = A::Error

impl<A, B, C, D, E> Future for Join5<A, B, C, D, E> where
    A: Future,
    B: Future<Error = A::Error>,
    C: Future<Error = A::Error>,
    D: Future<Error = A::Error>,
    E: Future<Error = A::Error>, [src]

type Item = (A::Item, B::Item, C::Item, D::Item, E::Item)

type Error = A::Error

impl<A, B, F> Future for AndThen<A, B, F> where
    A: Future,
    B: IntoFuture<Error = A::Error>,
    F: FnOnce(A::Item) -> B, [src]

type Item = B::Item

type Error = B::Error

impl<A, B, F> Future for OrElse<A, B, F> where
    A: Future,
    B: IntoFuture<Item = A::Item>,
    F: FnOnce(A::Error) -> B, [src]

type Item = B::Item

type Error = B::Error

impl<A, B, F> Future for Then<A, B, F> where
    A: Future,
    B: IntoFuture,
    F: FnOnce(Result<A::Item, A::Error>) -> B, [src]

type Item = B::Item

type Error = B::Error

impl<A, F> Future for Inspect<A, F> where
    A: Future,
    F: FnOnce(&A::Item), [src]

type Item = A::Item

type Error = A::Error

impl<A: Future> Future for Fuse<A>[src]

type Item = A::Item

type Error = A::Error

impl<A: Future, E: From<A::Error>> Future for FromErr<A, E>[src]

type Item = A::Item

type Error = E

impl<F> Future for CatchUnwind<F> where
    F: Future + UnwindSafe, [src]

type Item = Result<F::Item, F::Error>

type Error = Box<dyn Any + Send>

impl<F> Future for Shared<F> where
    F: Future, [src]

type Item = SharedItem<F::Item>

type Error = SharedError<F::Error>

impl<F, R> Future for Lazy<F, R> where
    F: FnOnce() -> R,
    R: IntoFuture, [src]

type Item = R::Item

type Error = R::Error

impl<F: Future> Future for futures::sync::oneshot::Execute<F>[src]

type Item = ()

type Error = ()

impl<F: Future> Future for futures::unsync::oneshot::Execute<F>[src]

type Item = ()

type Error = ()

impl<I> Future for JoinAll<I> where
    I: IntoIterator,
    I::Item: IntoFuture, [src]

type Item = Vec<<I::Item as IntoFuture>::Item>

type Error = <I::Item as IntoFuture>::Error

impl<S> Future for Collect<S> where
    S: Stream, [src]

type Item = Vec<S::Item>

type Error = S::Error

impl<S> Future for Concat<S> where
    S: Stream,
    S::Item: Extend<<<S as Stream>::Item as IntoIterator>::Item> + IntoIterator, [src]

type Item = S::Item

type Error = S::Error

impl<S> Future for Concat2<S> where
    S: Stream,
    S::Item: Extend<<<S as Stream>::Item as IntoIterator>::Item> + IntoIterator + Default, [src]

type Item = S::Item

type Error = S::Error

impl<S, F, Fut, T> Future for Fold<S, F, Fut, T> where
    S: Stream,
    F: FnMut(T, S::Item) -> Fut,
    Fut: IntoFuture<Item = T>,
    S::Error: From<Fut::Error>, [src]

type Item = T

type Error = S::Error

impl<S, F, U> Future for ForEach<S, F, U> where
    S: Stream,
    F: FnMut(S::Item) -> U,
    U: IntoFuture<Item = (), Error = S::Error>, [src]

type Item = ()

type Error = S::Error

impl<S, T, A, F> Future for LoopFn<A, F> where
    F: FnMut(S) -> A,
    A: IntoFuture<Item = Loop<T, S>>, [src]

type Item = T

type Error = A::Error

impl<S: Sink> Future for Flush<S>[src]

type Item = S

type Error = S::SinkError

impl<S: Sink> Future for Send<S>[src]

type Item = S

type Error = S::SinkError

impl<S: Stream> Future for StreamFuture<S>[src]

type Item = (Option<S::Item>, S)

type Error = (S::Error, S)

impl<S: Stream> Future for futures::sync::mpsc::Execute<S>[src]

type Item = ()

type Error = ()

impl<S: Stream> Future for futures::unsync::mpsc::Execute<S>[src]

type Item = ()

type Error = ()

impl<T> Future for futures::sync::oneshot::Receiver<T>[src]

type Item = T

type Error = Canceled

impl<T> Future for BiLockAcquire<T>[src]

type Item = BiLockAcquired<T>

type Error = ()

impl<T> Future for futures::unsync::oneshot::Receiver<T>[src]

type Item = T

type Error = Canceled

impl<T, E> Future for Empty<T, E>[src]

type Item = T

type Error = E

impl<T, E> Future for FutureResult<T, E>[src]

type Item = T

type Error = E

impl<T, E> Future for futures::sync::oneshot::SpawnHandle<T, E>[src]

type Item = T

type Error = E

impl<T, E> Future for futures::unsync::oneshot::SpawnHandle<T, E>[src]

type Item = T

type Error = E

impl<T, E, F> Future for PollFn<F> where
    F: FnMut() -> Poll<T, E>, [src]

type Item = T

type Error = E

impl<T, U> Future for SendAll<T, U> where
    T: Sink,
    U: Stream<Item = T::SinkItem>,
    T::SinkError: From<U::Error>, [src]

type Item = (T, U)

type Error = T::SinkError

impl<T, U> Future for Forward<T, U> where
    U: Sink<SinkItem = T::Item>,
    T: Stream,
    T::Error: From<U::SinkError>, [src]

type Item = (T, U)

type Error = T::Error

impl<U, A, F> Future for Map<A, F> where
    A: Future,
    F: FnOnce(A::Item) -> U, [src]

type Item = U

type Error = A::Error

impl<U, A, F> Future for MapErr<A, F> where
    A: Future,
    F: FnOnce(A::Error) -> U, [src]

type Item = A::Item

type Error = U

Loading content...