Struct TcpStream 

pub struct TcpStream(/* private fields */);

A TCP stream between a local and a remote socket.

After creating a TcpStream by either connecting to a remote host or accepting a connection on a TcpListener, data can be transmitted by reading and writing to it.

The connection will be closed when the value is dropped. The reading and writing portions of the connection can also be shut down individually with the shutdown method.

The Transmission Control Protocol is specified in IETF RFC 793.

Examples

use std::io::prelude::*;
use std::net::TcpStream;

fn main() -> std::io::Result<()> {
    let mut stream = TcpStream::connect("127.0.0.1:34254")?;

    stream.write(&[1])?;
    stream.read(&mut [0; 128])?;
    Ok(())
} // the stream is closed here

Platform-specific Behavior

On Unix, writes to the underlying socket in SOCK_STREAM mode are made with MSG_NOSIGNAL flag. This suppresses the emission of the SIGPIPE signal when writing to disconnected socket. In some cases, getting a SIGPIPE would trigger process termination.

Implementations

impl TcpStream

pub fn connect<A: ToSocketAddrs>(addr: A) -> Result<TcpStream>

Opens a TCP connection to a remote host.

addr is an address of the remote host. Anything which implements ToSocketAddrs trait can be supplied for the address; see this trait documentation for concrete examples.

If addr yields multiple addresses, connect will be attempted with each of the addresses until a connection is successful. If none of the addresses result in a successful connection, the error returned from the last connection attempt (the last address) is returned.

Examples

Open a TCP connection to 127.0.0.1:8080:

use std::net::TcpStream;

if let Ok(stream) = TcpStream::connect("127.0.0.1:8080") {
    println!("Connected to the server!");
} else {
    println!("Couldn't connect to server...");
}

Open a TCP connection to 127.0.0.1:8080. If the connection fails, open a TCP connection to 127.0.0.1:8081:

use std::net::{SocketAddr, TcpStream};

let addrs = [
    SocketAddr::from(([127, 0, 0, 1], 8080)),
    SocketAddr::from(([127, 0, 0, 1], 8081)),
];
if let Ok(stream) = TcpStream::connect(&addrs[..]) {
    println!("Connected to the server!");
} else {
    println!("Couldn't connect to server...");
}

pub fn connect_timeout( addr: &SocketAddr, timeout: Duration, ) -> Result<TcpStream>

Opens a TCP connection to a remote host with a timeout.

Unlike connect, connect_timeout takes a single SocketAddr since timeout must be applied to individual addresses.

It is an error to pass a zero Duration to this function.

Unlike other methods on TcpStream, this does not correspond to a single system call. It instead calls connect in nonblocking mode and then uses an OS-specific mechanism to await the completion of the connection request.

pub fn peer_addr(&self) -> Result<SocketAddr>

Returns the socket address of the remote peer of this TCP connection.

Examples

use std::net::{Ipv4Addr, SocketAddr, SocketAddrV4, TcpStream};

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
assert_eq!(stream.peer_addr().unwrap(),
           SocketAddr::V4(SocketAddrV4::new(Ipv4Addr::new(127, 0, 0, 1), 8080)));

pub fn local_addr(&self) -> Result<SocketAddr>

Returns the socket address of the local half of this TCP connection.

Examples

use std::net::{IpAddr, Ipv4Addr, TcpStream};

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
assert_eq!(stream.local_addr().unwrap().ip(),
           IpAddr::V4(Ipv4Addr::new(127, 0, 0, 1)));

pub fn shutdown(&self, how: Shutdown) -> Result<()>

Shuts down the read, write, or both halves of this connection.

This function will cause all pending and future I/O on the specified portions to return immediately with an appropriate value (see the documentation of Shutdown).

Platform-specific behavior

Calling this function multiple times may result in different behavior, depending on the operating system. On Linux, the second call will return Ok(()), but on macOS, it will return ErrorKind::NotConnected. This may change in the future.

Examples

use std::net::{Shutdown, TcpStream};

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.shutdown(Shutdown::Both).expect("shutdown should succeed");

pub fn try_clone(&self) -> Result<TcpStream>

Creates a new independently owned handle to the underlying socket.

The returned TcpStream is a reference to the same stream that this object references. Both handles will read and write the same stream of data, and options set on one stream will be propagated to the other stream.

Examples

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
let stream_clone = stream.try_clone().expect("clone should succeed");

pub fn set_read_timeout(&self, dur: Option<Duration>) -> Result<()>

Sets the read timeout to the timeout specified.

If the value specified is None, then read calls will block indefinitely. An Err is returned if the zero Duration is passed to this method.

Platform-specific behavior

Platforms may return a different error code whenever a read times out as a result of setting this option. For example Unix typically returns an error of the kind WouldBlock, but Windows may return TimedOut.

Examples

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_read_timeout(None).expect("set_read_timeout should succeed");

An Err is returned if the zero Duration is passed to this method:

use std::io;
use std::net::TcpStream;
use std::time::Duration;

let stream = TcpStream::connect("127.0.0.1:8080").unwrap();
let result = stream.set_read_timeout(Some(Duration::new(0, 0)));
let err = result.unwrap_err();
assert_eq!(err.kind(), io::ErrorKind::InvalidInput)

pub fn set_write_timeout(&self, dur: Option<Duration>) -> Result<()>

Sets the write timeout to the timeout specified.

If the value specified is None, then write calls will block indefinitely. An Err is returned if the zero Duration is passed to this method.

Platform-specific behavior

Platforms may return a different error code whenever a write times out as a result of setting this option. For example Unix typically returns an error of the kind WouldBlock, but Windows may return TimedOut.

Examples

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_write_timeout(None).expect("set_write_timeout should succeed");

An Err is returned if the zero Duration is passed to this method:

use std::io;
use std::net::TcpStream;
use std::time::Duration;

let stream = TcpStream::connect("127.0.0.1:8080").unwrap();
let result = stream.set_write_timeout(Some(Duration::new(0, 0)));
let err = result.unwrap_err();
assert_eq!(err.kind(), io::ErrorKind::InvalidInput)

pub fn read_timeout(&self) -> Result<Option<Duration>>

Returns the read timeout of this socket.

If the timeout is None, then read calls will block indefinitely.

Platform-specific behavior

Some platforms do not provide access to the current timeout.

Examples

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_read_timeout(None).expect("set_read_timeout should succeed");
assert_eq!(stream.read_timeout().unwrap(), None);

pub fn write_timeout(&self) -> Result<Option<Duration>>

Returns the write timeout of this socket.

If the timeout is None, then write calls will block indefinitely.

Platform-specific behavior

Some platforms do not provide access to the current timeout.

Examples

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_write_timeout(None).expect("set_write_timeout should succeed");
assert_eq!(stream.write_timeout().unwrap(), None);

pub fn peek(&self, buf: &mut [u8]) -> Result<usize>

Receives data on the socket from the remote address to which it is connected, without removing that data from the queue. On success, returns the number of bytes peeked.

Successive calls return the same data. This is accomplished by passing MSG_PEEK as a flag to the underlying recv system call.

Examples

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8000")
                       .expect("Couldn't connect to the server...");
let mut buf = [0; 10];
let len = stream.peek(&mut buf).expect("peek should succeed");

pub fn set_linger(&self, linger: Option<Duration>) -> Result<()>

🔬This is a nightly-only experimental API. (tcp_linger #88494)

Sets the value of the SO_LINGER option on this socket.

This value controls how the socket is closed when data remains to be sent. If SO_LINGER is set, the socket will remain open for the specified duration as the system attempts to send pending data. Otherwise, the system may close the socket immediately, or wait for a default timeout.

Examples

#![feature(tcp_linger)]

use std::net::TcpStream;
use std::time::Duration;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_linger(Some(Duration::from_secs(0))).expect("set_linger should succeed");

pub fn linger(&self) -> Result<Option<Duration>>

🔬This is a nightly-only experimental API. (tcp_linger #88494)

Gets the value of the SO_LINGER option on this socket.

For more information about this option, see TcpStream::set_linger.

Examples

#![feature(tcp_linger)]

use std::net::TcpStream;
use std::time::Duration;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_linger(Some(Duration::from_secs(0))).expect("set_linger should succeed");
assert_eq!(stream.linger().unwrap(), Some(Duration::from_secs(0)));

pub fn set_keepalive(&self, keepalive: bool) -> Result<()>

🔬This is a nightly-only experimental API. (tcp_keepalive #155889)

Sets the value of the SO_KEEPALIVE option on this socket.

If set to true, the operating system will periodically send keepalive probes on an idle connection to verify that the remote peer is still reachable. If the peer fails to respond after a system-determined number of probes, the connection is considered broken and subsequent I/O calls will return an error.

This is useful for detecting dead peers on long-lived connections where no application-level traffic is exchanged, such as database or SSH connections.

The timing and frequency of keepalive probes are controlled by system-level settings and are not configured by this method alone.

Examples

#![feature(tcp_keepalive)]

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_keepalive(true).expect("set_keepalive should succeed");

pub fn keepalive(&self) -> Result<bool>

🔬This is a nightly-only experimental API. (tcp_keepalive #155889)

Gets the value of the SO_KEEPALIVE option on this socket.

For more information about this option, see TcpStream::set_keepalive.

Examples

#![feature(tcp_keepalive)]

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_keepalive(true).expect("set_keepalive should succeed");
assert_eq!(stream.keepalive().unwrap_or(false), true);

pub fn set_nodelay(&self, nodelay: bool) -> Result<()>

Sets the value of the TCP_NODELAY option on this socket.

If set, this option disables the Nagle algorithm. This means that segments are always sent as soon as possible, even if there is only a small amount of data. When not set, data is buffered until there is a sufficient amount to send out, thereby avoiding the frequent sending of small packets.

Examples

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_nodelay(true).expect("set_nodelay should succeed");

pub fn nodelay(&self) -> Result<bool>

Gets the value of the TCP_NODELAY option on this socket.

For more information about this option, see TcpStream::set_nodelay.

Examples

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_nodelay(true).expect("set_nodelay should succeed");
assert_eq!(stream.nodelay().unwrap_or(false), true);

pub fn set_ttl(&self, ttl: u32) -> Result<()>

Sets the value for the IP_TTL option on this socket.

This value sets the time-to-live field that is used in every packet sent from this socket.

Examples

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_ttl(100).expect("set_ttl should succeed");

pub fn ttl(&self) -> Result<u32>

Gets the value of the IP_TTL option for this socket.

For more information about this option, see TcpStream::set_ttl.

Examples

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.set_ttl(100).expect("set_ttl should succeed");
assert_eq!(stream.ttl().unwrap_or(0), 100);

pub fn take_error(&self) -> Result<Option<Error>>

Gets the value of the SO_ERROR option on this socket.

This will retrieve the stored error in the underlying socket, clearing the field in the process. This can be useful for checking errors between calls.

Examples

use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:8080")
                       .expect("Couldn't connect to the server...");
stream.take_error().expect("No error was expected...");

pub fn set_nonblocking(&self, nonblocking: bool) -> Result<()>

Moves this TCP stream into or out of nonblocking mode.

This will result in read, write, recv and send system operations becoming nonblocking, i.e., immediately returning from their calls. If the IO operation is successful, Ok is returned and no further action is required. If the IO operation could not be completed and needs to be retried, an error with kind io::ErrorKind::WouldBlock is returned.

On most Unix platforms, calling this method corresponds to calling ioctl FIONBIO. On Windows, calling this method corresponds to calling ioctlsocket FIONBIO.

Examples

Reading bytes from a TCP stream in non-blocking mode:

use std::io::{self, Read};
use std::net::TcpStream;

let mut stream = TcpStream::connect("127.0.0.1:7878")
    .expect("Couldn't connect to the server...");
stream.set_nonblocking(true).expect("set_nonblocking should succeed");

let mut buf = vec![];
loop {
    match stream.read_to_end(&mut buf) {
        Ok(_) => break,
        Err(ref e) if e.kind() == io::ErrorKind::WouldBlock => {
            // wait until network socket is ready, typically implemented
            // via platform-specific APIs such as epoll or IOCP
            wait_for_fd();
        }
        Err(e) => panic!("encountered IO error: {e}"),
    };
};
println!("bytes: {buf:?}");

Trait Implementations

impl AsFd for TcpStream

Available on (Hermit or Motor OS or Trusty or Unix or WASI) and non-Trusty only.

fn as_fd(&self) -> BorrowedFd<'_>

Borrows the file descriptor.

impl AsRawFd for TcpStream

Available on (Hermit or Motor OS or Trusty or Unix or WASI) and non-Trusty only.

fn as_raw_fd(&self) -> RawFd

Extracts the raw file descriptor.

impl AsRawSocket for TcpStream

Available on Windows only.

fn as_raw_socket(&self) -> RawSocket

Extracts the raw socket.

impl AsSocket for TcpStream

Available on Windows only.

fn as_socket(&self) -> BorrowedSocket<'_>

Borrows the socket.

impl Debug for TcpStream

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl From<OwnedFd> for TcpStream

Available on (Hermit or Motor OS or Trusty or Unix or WASI) and non-Trusty only.

fn from(owned_fd: OwnedFd) -> Self

Converts to this type from the input type.

impl From<OwnedSocket> for TcpStream

Available on Windows only.

fn from(owned: OwnedSocket) -> Self

Converts to this type from the input type.

impl From<TcpStream> for OwnedSocket

Available on Windows only.

fn from(tcp_stream: TcpStream) -> OwnedSocket

Takes ownership of a TcpStream’s socket.

impl From<TcpStream> for OwnedFd

Available on (Hermit or Motor OS or Trusty or Unix or WASI) and non-Trusty only.

fn from(tcp_stream: TcpStream) -> OwnedFd

Takes ownership of a TcpStream’s socket file descriptor.

impl FromRawFd for TcpStream

Available on (Hermit or Motor OS or Trusty or Unix or WASI) and non-Trusty only.

unsafe fn from_raw_fd(fd: RawFd) -> TcpStream

Constructs a new instance of Self from the given raw file descriptor.

impl FromRawSocket for TcpStream

Available on Windows only.

unsafe fn from_raw_socket(sock: RawSocket) -> TcpStream

Constructs a new I/O object from the specified raw socket.

impl IntoRawFd for TcpStream

Available on (Hermit or Motor OS or Trusty or Unix or WASI) and non-Trusty only.

fn into_raw_fd(self) -> RawFd

Consumes this object, returning the raw underlying file descriptor.

impl IntoRawSocket for TcpStream

Available on Windows only.

fn into_raw_socket(self) -> RawSocket

Consumes this object, returning the raw underlying socket.

impl Read for TcpStream

fn read(&mut self, buf: &mut [u8]) -> Result<usize>

Pull some bytes from this source into the specified buffer, returning how many bytes were read.

fn read_buf(&mut self, buf: BorrowedCursor<'_, u8>) -> Result<()>

🔬This is a nightly-only experimental API. (read_buf #78485)

Pull some bytes from this source into the specified buffer.

fn read_vectored(&mut self, bufs: &mut [IoSliceMut<'_>]) -> Result<usize>

Like read, except that it reads into a slice of buffers.

fn is_read_vectored(&self) -> bool

🔬This is a nightly-only experimental API. (can_vector #69941)

Determines if this Reader has an efficient read_vectored implementation.

fn read_to_end(&mut self, buf: &mut Vec<u8>) -> Result<usize, Error>

Reads all bytes until EOF in this source, placing them into buf.

fn read_to_string(&mut self, buf: &mut String) -> Result<usize, Error>

Reads all bytes until EOF in this source, appending them to buf.

fn read_exact(&mut self, buf: &mut [u8]) -> Result<(), Error>

Reads the exact number of bytes required to fill buf.

fn read_buf_exact( &mut self, cursor: BorrowedCursor<'_, u8>, ) -> Result<(), Error>

🔬This is a nightly-only experimental API. (read_buf #78485)

Reads the exact number of bytes required to fill cursor.

fn by_ref(&mut self) -> &mut Self

where Self: Sized,

Creates a “by reference” adapter for this instance of Read.

fn bytes(self) -> Bytes<Self>

where Self: Sized,

Transforms this Read instance to an Iterator over its bytes.

fn chain<R>(self, next: R) -> Chain<Self, R>

where R: Read, Self: Sized,

Creates an adapter which will chain this stream with another.

fn take(self, limit: u64) -> Take<Self>

where Self: Sized,

Creates an adapter which will read at most limit bytes from it.

fn read_array<const N: usize>(&mut self) -> Result<[u8; N], Error>

where Self: Sized,

🔬This is a nightly-only experimental API. (read_array #148848)

Read and return a fixed array of bytes from this source.

fn read_le<T>(&mut self) -> Result<T, Error>

where T: FromEndianBytes, Self: Sized,

🔬This is a nightly-only experimental API. (read_le #156984)

Read and return a type (e.g. an integer) in little-endian order.

fn read_be<T>(&mut self) -> Result<T, Error>

where T: FromEndianBytes, Self: Sized,

🔬This is a nightly-only experimental API. (read_le #156984)

Read and return a type (e.g. an integer) in big-endian order.

impl Read for &TcpStream

fn read(&mut self, buf: &mut [u8]) -> Result<usize>

Pull some bytes from this source into the specified buffer, returning how many bytes were read.

fn read_buf(&mut self, buf: BorrowedCursor<'_, u8>) -> Result<()>

🔬This is a nightly-only experimental API. (read_buf #78485)

Pull some bytes from this source into the specified buffer.

fn read_vectored(&mut self, bufs: &mut [IoSliceMut<'_>]) -> Result<usize>

Like read, except that it reads into a slice of buffers.

fn is_read_vectored(&self) -> bool

🔬This is a nightly-only experimental API. (can_vector #69941)

Determines if this Reader has an efficient read_vectored implementation.

fn read_to_end(&mut self, buf: &mut Vec<u8>) -> Result<usize, Error>

Reads all bytes until EOF in this source, placing them into buf.

fn read_to_string(&mut self, buf: &mut String) -> Result<usize, Error>

Reads all bytes until EOF in this source, appending them to buf.

fn read_exact(&mut self, buf: &mut [u8]) -> Result<(), Error>

Reads the exact number of bytes required to fill buf.

fn read_buf_exact( &mut self, cursor: BorrowedCursor<'_, u8>, ) -> Result<(), Error>

🔬This is a nightly-only experimental API. (read_buf #78485)

Reads the exact number of bytes required to fill cursor.

fn by_ref(&mut self) -> &mut Self

where Self: Sized,

Creates a “by reference” adapter for this instance of Read.

fn bytes(self) -> Bytes<Self>

where Self: Sized,

Transforms this Read instance to an Iterator over its bytes.

fn chain<R>(self, next: R) -> Chain<Self, R>

where R: Read, Self: Sized,

Creates an adapter which will chain this stream with another.

fn take(self, limit: u64) -> Take<Self>

where Self: Sized,

Creates an adapter which will read at most limit bytes from it.

fn read_array<const N: usize>(&mut self) -> Result<[u8; N], Error>

where Self: Sized,

🔬This is a nightly-only experimental API. (read_array #148848)

Read and return a fixed array of bytes from this source.

fn read_le<T>(&mut self) -> Result<T, Error>

where T: FromEndianBytes, Self: Sized,

🔬This is a nightly-only experimental API. (read_le #156984)

Read and return a type (e.g. an integer) in little-endian order.

fn read_be<T>(&mut self) -> Result<T, Error>

where T: FromEndianBytes, Self: Sized,

🔬This is a nightly-only experimental API. (read_le #156984)

Read and return a type (e.g. an integer) in big-endian order.

impl TcpStreamExt for TcpStream

Available on Android or Cygwin or Linux only.

fn set_quickack(&self, quickack: bool) -> Result<()>

Enable or disable TCP_QUICKACK.

fn quickack(&self) -> Result<bool>

Gets the value of the TCP_QUICKACK option on this socket.

fn set_deferaccept(&self, accept: Duration) -> Result<()>

🔬This is a nightly-only experimental API. (tcp_deferaccept #119639)

Available on Linux only.

A socket listener will be awakened solely when data arrives.

fn deferaccept(&self) -> Result<Duration>

🔬This is a nightly-only experimental API. (tcp_deferaccept #119639)

Available on Linux only.

Gets the accept delay value of the TCP_DEFER_ACCEPT option.

impl Write for TcpStream

fn write(&mut self, buf: &[u8]) -> Result<usize>

Writes a buffer into this writer, returning how many bytes were written.

fn write_vectored(&mut self, bufs: &[IoSlice<'_>]) -> Result<usize>

Like write, except that it writes from a slice of buffers.

fn is_write_vectored(&self) -> bool

🔬This is a nightly-only experimental API. (can_vector #69941)

Determines if this Writer has an efficient write_vectored implementation.

fn flush(&mut self) -> Result<()>

Flushes this output stream, ensuring that all intermediately buffered contents reach their destination.

fn write_all(&mut self, buf: &[u8]) -> Result<(), Error>

Attempts to write an entire buffer into this writer.

fn write_all_vectored(&mut self, bufs: &mut [IoSlice<'_>]) -> Result<(), Error>

🔬This is a nightly-only experimental API. (write_all_vectored #70436)

Attempts to write multiple buffers into this writer.

fn write_fmt(&mut self, args: Arguments<'_>) -> Result<(), Error>

Writes a formatted string into this writer, returning any error encountered.

fn by_ref(&mut self) -> &mut Self

where Self: Sized,

Creates a “by reference” adapter for this instance of Write.

impl Write for &TcpStream

fn write(&mut self, buf: &[u8]) -> Result<usize>

Writes a buffer into this writer, returning how many bytes were written.

fn write_vectored(&mut self, bufs: &[IoSlice<'_>]) -> Result<usize>

Like write, except that it writes from a slice of buffers.

fn is_write_vectored(&self) -> bool

🔬This is a nightly-only experimental API. (can_vector #69941)

Determines if this Writer has an efficient write_vectored implementation.

fn flush(&mut self) -> Result<()>

Flushes this output stream, ensuring that all intermediately buffered contents reach their destination.

fn write_all(&mut self, buf: &[u8]) -> Result<(), Error>

Attempts to write an entire buffer into this writer.

fn write_all_vectored(&mut self, bufs: &mut [IoSlice<'_>]) -> Result<(), Error>

🔬This is a nightly-only experimental API. (write_all_vectored #70436)

Attempts to write multiple buffers into this writer.

fn write_fmt(&mut self, args: Arguments<'_>) -> Result<(), Error>

Writes a formatted string into this writer, returning any error encountered.

fn by_ref(&mut self) -> &mut Self

where Self: Sized,

Creates a “by reference” adapter for this instance of Write.