Struct BufWriter

pub struct BufWriter<W: ?Sized + Write> {
    buf: Vec<u8>,
    panicked: bool,
    inner: W,
}

Wraps a writer and buffers its output.

It can be excessively inefficient to work directly with something that implements Write. For example, every call to write on TcpStream results in a system call. A BufWriter<W> keeps an in-memory buffer of data and writes it to an underlying writer in large, infrequent batches.

BufWriter<W> can improve the speed of programs that make small and repeated write calls to the same file or network socket. It does not help when writing very large amounts at once, or writing just one or a few times. It also provides no advantage when writing to a destination that is in memory, like a Vec<u8>.

It is critical to call flush before BufWriter<W> is dropped. Though dropping will attempt to flush the contents of the buffer, any errors that happen in the process of dropping will be ignored. Calling flush ensures that the buffer is empty and thus dropping will not even attempt file operations.

Examples

Let’s write the numbers one through ten to a TcpStream:

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

let mut stream = TcpStream::connect("127.0.0.1:34254").unwrap();

for i in 0..10 {
    stream.write(&[i+1]).unwrap();
}

Because we’re not buffering, we write each one in turn, incurring the overhead of a system call per byte written. We can fix this with a BufWriter<W>:

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

let mut stream = BufWriter::new(TcpStream::connect("127.0.0.1:34254").unwrap());

for i in 0..10 {
    stream.write(&[i+1]).unwrap();
}
stream.flush().unwrap();

By wrapping the stream with a BufWriter<W>, these ten writes are all grouped together by the buffer and will all be written out in one system call when the stream is flushed.

Fields

buf: Vec<u8>

panicked: bool

inner: W

Implementations

impl<W: Write> BufWriter<W>

pub fn new(inner: W) -> BufWriter<W>

Creates a new BufWriter<W> with a default buffer capacity. The default is currently 8 KiB, but may change in the future.

Examples

use std::io::BufWriter;
use std::net::TcpStream;

let mut buffer = BufWriter::new(TcpStream::connect("127.0.0.1:34254").unwrap());

pub(crate) fn try_new_buffer() -> Result<Vec<u8>>

pub(crate) fn with_buffer(inner: W, buf: Vec<u8>) -> Self

pub fn with_capacity(capacity: usize, inner: W) -> BufWriter<W>

Creates a new BufWriter<W> with at least the specified buffer capacity.

Examples

Creating a buffer with a buffer of at least a hundred bytes.

use std::io::BufWriter;
use std::net::TcpStream;

let stream = TcpStream::connect("127.0.0.1:34254").unwrap();
let mut buffer = BufWriter::with_capacity(100, stream);

pub fn into_inner(self) -> Result<W, IntoInnerError<BufWriter<W>>>

Unwraps this BufWriter<W>, returning the underlying writer.

The buffer is written out before returning the writer.

Errors

An Err will be returned if an error occurs while flushing the buffer.

Examples

use std::io::BufWriter;
use std::net::TcpStream;

let mut buffer = BufWriter::new(TcpStream::connect("127.0.0.1:34254").unwrap());

// unwrap the TcpStream and flush the buffer
let stream = buffer.into_inner().unwrap();

pub fn into_parts(self) -> (W, Result<Vec<u8>, WriterPanicked>)

Disassembles this BufWriter<W>, returning the underlying writer, and any buffered but unwritten data.

If the underlying writer panicked, it is not known what portion of the data was written. In this case, we return WriterPanicked for the buffered data (from which the buffer contents can still be recovered).

into_parts makes no attempt to flush data and cannot fail.

Examples

use std::io::{BufWriter, Write};

let mut buffer = [0u8; 10];
let mut stream = BufWriter::new(buffer.as_mut());
write!(stream, "too much data").unwrap();
stream.flush().expect_err("it doesn't fit");
let (recovered_writer, buffered_data) = stream.into_parts();
assert_eq!(recovered_writer.len(), 0);
assert_eq!(&buffered_data.unwrap(), b"ata");

impl<W: ?Sized + Write> BufWriter<W>

pub(in io) fn flush_buf( &mut self, ) -> Result<()>

Send data in our local buffer into the inner writer, looping as necessary until either it’s all been sent or an error occurs.

Because all the data in the buffer has been reported to our owner as “successfully written” (by returning nonzero success values from write), any 0-length writes from inner must be reported as i/o errors from this method.

pub(super) fn write_to_buf(&mut self, buf: &[u8]) -> usize

Buffer some data without flushing it, regardless of the size of the data. Writes as much as possible without exceeding capacity. Returns the number of bytes written.

pub fn get_ref(&self) -> &W

Gets a reference to the underlying writer.

Examples

use std::io::BufWriter;
use std::net::TcpStream;

let mut buffer = BufWriter::new(TcpStream::connect("127.0.0.1:34254").unwrap());

// we can use reference just like buffer
let reference = buffer.get_ref();

pub fn get_mut(&mut self) -> &mut W

Gets a mutable reference to the underlying writer.

It is inadvisable to directly write to the underlying writer.

Examples

use std::io::BufWriter;
use std::net::TcpStream;

let mut buffer = BufWriter::new(TcpStream::connect("127.0.0.1:34254").unwrap());

// we can use reference just like buffer
let reference = buffer.get_mut();

pub fn buffer(&self) -> &[u8]

Returns a reference to the internally buffered data.

Examples

use std::io::BufWriter;
use std::net::TcpStream;

let buf_writer = BufWriter::new(TcpStream::connect("127.0.0.1:34254").unwrap());

// See how many bytes are currently buffered
let bytes_buffered = buf_writer.buffer().len();

pub(in io) fn buffer_mut( &mut self, ) -> &mut Vec<u8>

Returns a mutable reference to the internal buffer.

This can be used to write data directly into the buffer without triggering writers to the underlying writer.

That the buffer is a Vec is an implementation detail. Callers should not modify the capacity as there currently is no public API to do so and thus any capacity changes would be unexpected by the user.

pub fn capacity(&self) -> usize

Returns the number of bytes the internal buffer can hold without flushing.

Examples

use std::io::BufWriter;
use std::net::TcpStream;

let buf_writer = BufWriter::new(TcpStream::connect("127.0.0.1:34254").unwrap());

// Check the capacity of the inner buffer
let capacity = buf_writer.capacity();
// Calculate how many bytes can be written without flushing
let without_flush = capacity - buf_writer.buffer().len();

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

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

unsafe fn write_to_buffer_unchecked(&mut self, buf: &[u8])

fn spare_capacity(&self) -> usize

Trait Implementations

impl<I: Write + ?Sized> BufferedWriterSpec for BufWriter<I>

fn buffer_size(&self) -> usize

fn copy_from<R: Read + ?Sized>(&mut self, reader: &mut R) -> Result<u64>

impl<T: ?Sized + CopyWrite> CopyWrite for BufWriter<T>

fn properties(&self) -> CopyParams

Extracts the file descriptor and hints/metadata, delegating through wrappers if necessary.

impl<W> Debug for BufWriter<W>

where W: Debug + ?Sized + Write,

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

Formats the value using the given formatter.

impl<W: ?Sized + Write> Drop for BufWriter<W>

fn drop(&mut self)

Executes the destructor for this type.

impl<W: ?Sized + Write + Seek> Seek for BufWriter<W>

fn seek(&mut self, pos: SeekFrom) -> Result<u64>

Seek to the offset, in bytes, in the underlying writer.

Seeking always writes out the internal buffer before seeking.

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

Rewind to the beginning of a stream.

fn stream_len(&mut self) -> Result<u64>

🔬This is a nightly-only experimental API. (seek_stream_len #59359)

Returns the length of this stream (in bytes).

fn stream_position(&mut self) -> Result<u64>

Returns the current seek position from the start of the stream.

fn seek_relative(&mut self, offset: i64) -> Result<()>

Seeks relative to the current position.

impl<W: ?Sized + Write> Write for BufWriter<W>

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

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

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

Attempts to write an entire buffer into this writer.

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

🔬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<()>

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.