Struct LossyDecoder 

pub struct LossyDecoder<F: FnMut(&str)> { /* private fields */ }

A push-based, lossy decoder for UTF-8. Errors are replaced with the U+FFFD replacement character.

Users “push” bytes into the decoder, which in turn “pushes” &str slices into a callback.

Note: Dropping the decoder signals the end of the input: If the last input chunk ended with an incomplete byte sequence for a code point, this is an error and a replacement character is emitted. Use std::mem::forget to inhibit this behavior.

Examples

Single-shot lossy decoding (like String::from_utf8_lossy but returning String):

use utf8_zero::LossyDecoder;

fn string_from_utf8_lossy(input: &[u8]) -> String {
    let mut string = String::new();
    LossyDecoder::new(|s| string.push_str(s)).feed(input);
    string
}

assert_eq!(string_from_utf8_lossy(b"Hello\xC0World"), "Hello\u{FFFD}World");

Streaming chunks – a multi-byte code point split across two feeds:

use utf8_zero::LossyDecoder;

let mut output = String::new();
{
    let mut decoder = LossyDecoder::new(|s| output.push_str(s));
    decoder.feed(b"Hello \xC3");  // first byte of U+00E9
    decoder.feed(b"\xA9!");       // second byte + excl mark
}
assert_eq!(output, "Hello \u{00E9}!");

Implementations

impl<F: FnMut(&str)> LossyDecoder<F>

pub fn new(push_str: F) -> Self

Create a new decoder from a callback.

pub fn feed(&mut self, input: &[u8])

Feed one chunk of input into the decoder.

The input is decoded lossily and the callback called once or more with &str string slices.

If the UTF-8 byte sequence for one code point was split into this bytes chunk and previous bytes chunks, it will be correctly pieced back together.

Trait Implementations

impl<F: FnMut(&str)> Drop for LossyDecoder<F>

fn drop(&mut self)

Executes the destructor for this type.