Base51

ASCII-only bigint encodings, suitable for C-like identifiers, CSS classnames, JSON serialization, etc.

Usage

yarn add @anireact/base51

npm install @anireact/base51

import { e51, d51, e93, d93 } from '@anireact/base51';

// Encode an integer with Base51:
assert(e51(0xe81b73abdbca50c433839f9255368d1an) === 'ISuPzhbXXDxQyswSJLwjnxv');

// Decode Base51 string:
assert(d51('ISuPzhbXXDxQyswSJLwjnxv') === 0xe81b73abdbca50c433839f9255368d1an);

// Encode the same integer using Base93:
assert(e93(0xe81b73abdbca50c433839f9255368d1an) === '-82Fga7uV LTYF3lNGw1');

// And then decode the string:
assert(d93('-82Fga7uV LTYF3lNGw1') === 0xe81b73abdbca50c433839f9255368d1an);

Base51

Base51 is designed to encode integers into ASCII-only strings, suitable for usage as C-like identifiers, CSS classnames, UNIX filenames, URL fragments, $\TeX$ commands, etc. The encoding uses ASCII alphabetic characters except l.

In addition, the decoder recognizes confusable characters as similar-looking characters from the main character set, which allows the encoding to be used to create user-visible identifiers.

NB: When Base51 strings are used to create identifiers in generated code, they should be prefixed to avoid conflicts with the target language keywords.

Base93

Base93 is designed to encode integers into ASCII-only strings, suitable for JSON serialization. The encoding uses all printable ASCII characters except " and \. Unlike Base51, the decoder doesn’t care about confusable characters.

Efficiency

Encoding	Compared to binary	Compared to Base64
Base51	$\frac{\ln{51}}{\ln{256}}≈70.91%$	$\frac{\ln{51}}{\ln{64}}≈94.54%$
Base93	$\frac{\ln{93}}{\ln{256}}≈81.74%$	$\frac{\ln{93}}{\ln{64}}≈109.0%$

Performance

The algorithm has quadratic time complexity, and therefore should not be used to process large numbers (more than approx. 100 bytes).

Support

If you have any issues, feel free to fork this plugin, or contact me on Telegram https://t.me/miyaokamarina.

License

MIT © 2023 Yuri Zemskov