Function hex_decode 

pub fn hex_decode<'a>(
    src: &[u8],
    dst: &'a mut [u8],
) -> Result<&'a mut [u8], Error>

Decodes all of src into dst without allocation, accepting either letter case.

src must contain an even number of ASCII hex digits, without a prefix, whitespace or separators. Uppercase and lowercase digits may be mixed. Leading zeroes are preserved as bytes; this converts a byte sequence, not an integer.

dst must have at least src.len() / 2 bytes. The returned mutable slice covers exactly the written prefix and borrows only dst. Spare destination bytes remain unchanged. Empty input returns an empty slice and changes nothing.

Use hex_decode_with_case to restrict letter case, or hex_decode_array when the decoded size must match a fixed array exactly.

Errors

Errors are checked in this order:

1. Error::OddLength if the input has an odd number of bytes.
2. Error::OutputTooSmall if the destination is too small. required counts the full decoded output size in bytes.
3. Error::InvalidChar for the first invalid input byte. index is a zero-based byte offset in src, not a Unicode character position.

Every error leaves the entire destination unchanged, including when an invalid byte occurs after a long valid prefix. A short destination never causes silent prefix decoding; slice src explicitly if that is intended.

Examples

use faster_hex::hex_decode;

let mut destination = [0xa5; 5];
let bytes = {
    let source = *b"00aBcD";
    hex_decode(&source, &mut destination)?
}; // The source is no longer needed.
assert_eq!(bytes, &[0, 0xab, 0xcd]);
bytes[0] = 0xff;
assert_eq!(destination, [0xff, 0xab, 0xcd, 0xa5, 0xa5]);

Invalid input does not commit a partial result:

use faster_hex::{hex_decode, Error};

let mut destination = [0xa5; 3];
assert!(matches!(hex_decode(b"00ff0g", &mut destination),
    Err(Error::InvalidChar { index: 5, byte: b'g', .. })));
assert_eq!(destination, [0xa5; 3]);