3 releases
new 0.1.2 | Apr 15, 2025 |
---|---|
0.1.1 | Apr 15, 2025 |
0.1.0 | Apr 15, 2025 |
#1445 in Encoding
208 downloads per month
Used in 2 crates
(via elb-dl)
28KB
652 lines
base32-fs
This crate implements a variant of BASE32 encoding for hashes that are used as file names: no two encoded hashes can be decoded to the same original hash unless there is a hash collision.
To achieve that the crate
- uses the same characters when encoding and decoding the data
as opposed to the original Crockford's alphabet that permits e.g. both "a" and "A" to be decoded to
10
; - doesn't zero-extend invalid input lengths when decoding.
Besides that the crate
- uses only lowercase letters instead of uppercase as the most common representation of hashes;
- doesn't use padding characters;
- doesn't change the sorting order of the encoded data.
Usage
Encode into PathBuf
use std::path::PathBuf;
use base32_fs::{encode, encoded_len, PathBufOutput};
let input = *b"hello";
let mut output = PathBufOutput::with_capacity(encoded_len(input.len()));
encode(&input, &mut output);
assert_eq!(PathBuf::from("d1jprv3f"), output.into_path_buf());
Decode from PathBuf
use std::path::Path;
use base32_fs::{decode, decoded_len, PathBufInput};
let input = PathBufInput::new(Path::new("d1jprv3f"));
let mut output: Vec<u8> = Vec::new();
decode(input, &mut output);
assert_eq!(b"hello", output.as_slice());
Encode into Vec<u8>
use base32_fs::{encode, encoded_len};
let input = *b"hello";
let mut output: Vec<u8> = Vec::with_capacity(encoded_len(input.len()));
encode(&input, &mut output);
let string = std::str::from_utf8(output.as_slice()).expect("Always a valid UTF-8 byte sequence");
assert_eq!("d1jprv3f", string);
Decode from &[u8]
use std::path::Path;
use base32_fs::{decode, decoded_len, PathBufInput};
let input = b"d1jprv3f";
let mut output: Vec<u8> = Vec::with_capacity(decoded_len(input.len()).unwrap());
decode(input.as_slice(), &mut output);
assert_eq!(b"hello", output.as_slice());