Expand description
§Base122 Encoding Library
A high-performance Base122 encoding/decoding library for Rust, based on the original kevinAlbs Base122 algorithm.
Base122 is a binary-to-text encoding that is approximately 14% more space-efficient than Base64, making it ideal for data URIs and other space-constrained applications.
§Algorithm Overview
Base122 uses a bitwise approach:
- Extracts 7-bit chunks from input data
- Maps safe characters directly to single bytes
- Encodes “dangerous characters” using UTF-8 multi-byte sequences
- Achieves ~87% compression efficiency (7 bits input per 8 bits output)
§Dangerous Characters
Six characters are considered “dangerous” for transmission and are specially encoded:
\0(null) - can truncate strings\n(newline) - breaks single-line formats\r(carriage return) - breaks single-line formats"(double quote) - conflicts with JSON/HTML attributes&(ampersand) - conflicts with HTML entities\(backslash) - conflicts with escape sequences
§Performance
- Theoretical efficiency: 87.5% (7 bits input / 8 bits output)
- Actual efficiency: ~87% for large data
- vs Base64: ~14% smaller output
- Dependencies: Zero - pure Rust implementation
§Examples
use base122_rs::{encode, decode};
// Basic encoding/decoding
let data = b"Hello, World!";
let encoded = encode(data);
let decoded = decode(&encoded).unwrap();
assert_eq!(data, &decoded[..]);
// Binary data with dangerous characters
let binary_data = vec![0, 10, 13, 34, 38, 92, 65, 66, 67];
let encoded = encode(&binary_data);
let decoded = decode(&encoded).unwrap();
assert_eq!(binary_data, decoded);