Trait bstr::io::BufReadExt

source · 
pub trait BufReadExt: BufRead {
    // Provided methods
    fn byte_lines(self) -> ByteLines<Self> ⓘ
       where Self: Sized { ... }
    fn byte_records(self, terminator: u8) -> ByteRecords<Self> ⓘ
       where Self: Sized { ... }
    fn for_byte_line<F>(&mut self, for_each_line: F) -> Result<()>
       where Self: Sized,
             F: FnMut(&[u8]) -> Result<bool> { ... }
    fn for_byte_record<F>(
        &mut self,
        terminator: u8,
        for_each_record: F
    ) -> Result<()>
       where Self: Sized,
             F: FnMut(&[u8]) -> Result<bool> { ... }
    fn for_byte_line_with_terminator<F>(
        &mut self,
        for_each_line: F
    ) -> Result<()>
       where Self: Sized,
             F: FnMut(&[u8]) -> Result<bool> { ... }
    fn for_byte_record_with_terminator<F>(
        &mut self,
        terminator: u8,
        for_each_record: F
    ) -> Result<()>
       where Self: Sized,
             F: FnMut(&[u8]) -> Result<bool> { ... }
}
Available on crate feature std only.
Expand description

An extension trait for std::io::BufRead which provides convenience APIs for dealing with byte strings.

Provided Methods§

source

fn byte_lines(self) -> ByteLines<Self> ⓘ
where Self: Sized,

Returns an iterator over the lines of this reader, where each line is represented as a byte string.

Each item yielded by this iterator is a io::Result<Vec<u8>>, where an error is yielded if there was a problem reading from the underlying reader.

On success, the next line in the iterator is returned. The line does not contain a trailing \n or \r\n.

§Examples

Basic usage:

use std::io;

use bstr::io::BufReadExt;

let mut cursor = io::Cursor::new(b"lorem\nipsum\r\ndolor");

let mut lines = vec![];
for result in cursor.byte_lines() {
    let line = result?;
    lines.push(line);
}
assert_eq!(lines.len(), 3);
assert_eq!(lines[0], "lorem".as_bytes());
assert_eq!(lines[1], "ipsum".as_bytes());
assert_eq!(lines[2], "dolor".as_bytes());
source

fn byte_records(self, terminator: u8) -> ByteRecords<Self> ⓘ
where Self: Sized,

Returns an iterator over byte-terminated records of this reader, where each record is represented as a byte string.

Each item yielded by this iterator is a io::Result<Vec<u8>>, where an error is yielded if there was a problem reading from the underlying reader.

On success, the next record in the iterator is returned. The record does not contain its trailing terminator.

Note that calling byte_records(b'\n') differs from byte_lines() in that it has no special handling for \r.

§Examples

Basic usage:

use std::io;

use bstr::io::BufReadExt;

let mut cursor = io::Cursor::new(b"lorem\x00ipsum\x00dolor");

let mut records = vec![];
for result in cursor.byte_records(b'\x00') {
    let record = result?;
    records.push(record);
}
assert_eq!(records.len(), 3);
assert_eq!(records[0], "lorem".as_bytes());
assert_eq!(records[1], "ipsum".as_bytes());
assert_eq!(records[2], "dolor".as_bytes());
source

fn for_byte_line<F>(&mut self, for_each_line: F) -> Result<()>
where Self: Sized, F: FnMut(&[u8]) -> Result<bool>,

Executes the given closure on each line in the underlying reader.

If the closure returns an error (or if the underlying reader returns an error), then iteration is stopped and the error is returned. If false is returned, then iteration is stopped and no error is returned.

The closure given is called on exactly the same values as yielded by the byte_lines iterator. Namely, lines do not contain trailing \n or \r\n bytes.

This routine is useful for iterating over lines as quickly as possible. Namely, a single allocation is reused for each line.

§Examples

Basic usage:

use std::io;

use bstr::io::BufReadExt;

let mut cursor = io::Cursor::new(b"lorem\nipsum\r\ndolor");

let mut lines = vec![];
cursor.for_byte_line(|line| {
    lines.push(line.to_vec());
    Ok(true)
})?;
assert_eq!(lines.len(), 3);
assert_eq!(lines[0], "lorem".as_bytes());
assert_eq!(lines[1], "ipsum".as_bytes());
assert_eq!(lines[2], "dolor".as_bytes());
source

fn for_byte_record<F>( &mut self, terminator: u8, for_each_record: F ) -> Result<()>
where Self: Sized, F: FnMut(&[u8]) -> Result<bool>,

Executes the given closure on each byte-terminated record in the underlying reader.

If the closure returns an error (or if the underlying reader returns an error), then iteration is stopped and the error is returned. If false is returned, then iteration is stopped and no error is returned.

The closure given is called on exactly the same values as yielded by the byte_records iterator. Namely, records do not contain a trailing terminator byte.

This routine is useful for iterating over records as quickly as possible. Namely, a single allocation is reused for each record.

§Examples

Basic usage:

use std::io;

use bstr::io::BufReadExt;

let mut cursor = io::Cursor::new(b"lorem\x00ipsum\x00dolor");

let mut records = vec![];
cursor.for_byte_record(b'\x00', |record| {
    records.push(record.to_vec());
    Ok(true)
})?;
assert_eq!(records.len(), 3);
assert_eq!(records[0], "lorem".as_bytes());
assert_eq!(records[1], "ipsum".as_bytes());
assert_eq!(records[2], "dolor".as_bytes());
source

fn for_byte_line_with_terminator<F>(&mut self, for_each_line: F) -> Result<()>
where Self: Sized, F: FnMut(&[u8]) -> Result<bool>,

Executes the given closure on each line in the underlying reader.

If the closure returns an error (or if the underlying reader returns an error), then iteration is stopped and the error is returned. If false is returned, then iteration is stopped and no error is returned.

Unlike for_byte_line, the lines given to the closure do include the line terminator, if one exists.

This routine is useful for iterating over lines as quickly as possible. Namely, a single allocation is reused for each line.

This is identical to for_byte_record_with_terminator with a terminator of \n.

§Examples

Basic usage:

use std::io;

use bstr::io::BufReadExt;

let mut cursor = io::Cursor::new(b"lorem\nipsum\r\ndolor");

let mut lines = vec![];
cursor.for_byte_line_with_terminator(|line| {
    lines.push(line.to_vec());
    Ok(true)
})?;
assert_eq!(lines.len(), 3);
assert_eq!(lines[0], "lorem\n".as_bytes());
assert_eq!(lines[1], "ipsum\r\n".as_bytes());
assert_eq!(lines[2], "dolor".as_bytes());
source

fn for_byte_record_with_terminator<F>( &mut self, terminator: u8, for_each_record: F ) -> Result<()>
where Self: Sized, F: FnMut(&[u8]) -> Result<bool>,

Executes the given closure on each byte-terminated record in the underlying reader.

If the closure returns an error (or if the underlying reader returns an error), then iteration is stopped and the error is returned. If false is returned, then iteration is stopped and no error is returned.

Unlike for_byte_record, the lines given to the closure do include the record terminator, if one exists.

This routine is useful for iterating over records as quickly as possible. Namely, a single allocation is reused for each record.

§Examples

Basic usage:

use std::io;

use bstr::{io::BufReadExt, B};

let mut cursor = io::Cursor::new(b"lorem\x00ipsum\x00dolor");

let mut records = vec![];
cursor.for_byte_record_with_terminator(b'\x00', |record| {
    records.push(record.to_vec());
    Ok(true)
})?;
assert_eq!(records.len(), 3);
assert_eq!(records[0], B(b"lorem\x00"));
assert_eq!(records[1], B("ipsum\x00"));
assert_eq!(records[2], B("dolor"));

Implementors§