otp/utils/base32.js

/**
 * Enhanced Base32 implementation compatible with RFC 4648
 * Optimized for WeChat Mini Program environment
 */

// RFC 4648 standard Base32 alphabet and padding
const DEFAULT_ALPHABET = 'ABCDEFGHIJKLMNOPQRSTUVWXYZ234567';
const PADDING = '=';

// Pre-computed lookup tables
let ENCODE_TABLE = [...DEFAULT_ALPHABET];
let DECODE_TABLE = new Map(ENCODE_TABLE.map((char, index) => [char, index]));

// Valid padding lengths for Base32
const VALID_PADDING_LENGTHS = new Set([0, 1, 3, 4, 6]);

/**
 * Custom error class for Base32 operations
 */
class Base32Error extends Error {
  constructor(message, type = 'GENERAL_ERROR') {
    super(message);
    this.name = 'Base32Error';
    this.type = type;
  }
}

/**
 * String to UTF-8 bytes conversion (WeChat Mini Program compatible)
 * @private
 */
function stringToUint8Array(str) {
  const arr = [];
  for (let i = 0; i < str.length; i++) {
    let code = str.charCodeAt(i);
    
    if (code < 0x80) {
      arr.push(code);
    } else if (code < 0x800) {
      arr.push(0xc0 | (code >> 6));
      arr.push(0x80 | (code & 0x3f));
    } else if (code < 0xd800 || code >= 0xe000) {
      arr.push(0xe0 | (code >> 12));
      arr.push(0x80 | ((code >> 6) & 0x3f));
      arr.push(0x80 | (code & 0x3f));
    } else {
      // Handle surrogate pairs
      i++;
      code = 0x10000 + (((code & 0x3ff) << 10) | (str.charCodeAt(i) & 0x3ff));
      arr.push(0xf0 | (code >> 18));
      arr.push(0x80 | ((code >> 12) & 0x3f));
      arr.push(0x80 | ((code >> 6) & 0x3f));
      arr.push(0x80 | (code & 0x3f));
    }
  }
  return new Uint8Array(arr);
}

/**
 * UTF-8 bytes to string conversion (WeChat Mini Program compatible)
 * @private
 */
function uint8ArrayToString(bytes) {
  let result = '';
  for (let i = 0; i < bytes.length; i++) {
    const byte = bytes[i];
    if (byte < 0x80) {
      result += String.fromCharCode(byte);
    } else if (byte < 0xe0) {
      result += String.fromCharCode(
        ((byte & 0x1f) << 6) |
        (bytes[++i] & 0x3f)
      );
    } else if (byte < 0xf0) {
      result += String.fromCharCode(
        ((byte & 0x0f) << 12) |
        ((bytes[++i] & 0x3f) << 6) |
        (bytes[++i] & 0x3f)
      );
    } else {
      const codePoint = (
        ((byte & 0x07) << 18) |
        ((bytes[++i] & 0x3f) << 12) |
        ((bytes[++i] & 0x3f) << 6) |
        (bytes[++i] & 0x3f)
      );
      result += String.fromCodePoint(codePoint);
    }
  }
  return result;
}

/**
 * Sets a custom alphabet for Base32 encoding/decoding
 * @param {string} alphabet - The custom Base32 alphabet (32 unique characters)
 * @throws {Base32Error} If the alphabet is invalid
 */
function setCustomAlphabet(alphabet) {
  if (typeof alphabet !== 'string' || alphabet.length !== 32 || new Set(alphabet).size !== 32) {
    throw new Base32Error(
      'Invalid custom alphabet: must be 32 unique characters',
      'INVALID_ALPHABET'
    );
  }
  
  ENCODE_TABLE = [...alphabet];
  DECODE_TABLE = new Map(ENCODE_TABLE.map((char, index) => [char, index]));
}

/**
 * Resets to the default Base32 alphabet
 */
function resetToDefaultAlphabet() {
  ENCODE_TABLE = [...DEFAULT_ALPHABET];
  DECODE_TABLE = new Map(ENCODE_TABLE.map((char, index) => [char, index]));
}

/**
 * Validates padding length and position
 * @private
 */
function validatePadding(input) {
  const paddingMatch = input.match(/=+$/);
  if (!paddingMatch) return true;
  
  const paddingLength = paddingMatch[0].length;
  if (!VALID_PADDING_LENGTHS.has(paddingLength)) {
    throw new Base32Error(
      `Invalid padding length: ${paddingLength}`,
      'INVALID_PADDING'
    );
  }

  if (input.indexOf('=') !== input.length - paddingLength) {
    throw new Base32Error(
      'Padding character in invalid position',
      'INVALID_PADDING_POSITION'
    );
  }

  return true;
}

/**
 * Encodes a string or Uint8Array to Base32
 * @param {string|Uint8Array} input - The input to encode
 * @param {Object} [options] - Encoding options
 * @param {boolean} [options.noPadding=false] - Whether to omit padding
 * @param {boolean} [options.strict=false] - Whether to enable strict mode
 * @param {string} [options.alphabet] - Optional custom Base32 alphabet
 * @returns {string} The Base32 encoded string
 * @throws {Base32Error} If the input is invalid
 */
function encode(input, options = {}) {
  try {
    if (options.alphabet) {
      setCustomAlphabet(options.alphabet);
    }

    const data = typeof input === 'string' 
      ? stringToUint8Array(input)
      : input;

    if (!(data instanceof Uint8Array)) {
      throw new Base32Error(
        'Input must be a string or Uint8Array',
        'INVALID_INPUT_TYPE'
      );
    }

    let bits = 0;
    let value = 0;
    let output = '';

    for (let i = 0; i < data.length; i++) {
      value = (value << 8) | data[i];
      bits += 8;

      while (bits >= 5) {
        output += ENCODE_TABLE[(value >>> (bits - 5)) & 31];
        bits -= 5;
      }
    }

    if (bits > 0) {
      output += ENCODE_TABLE[(value << (5 - bits)) & 31];
    }

    if (!options.noPadding) {
      const pad = 8 - (output.length % 8);
      if (pad !== 8) {
        output += PADDING.repeat(pad);
      }
    }

    return output;
  } catch (error) {
    if (error instanceof Base32Error) throw error;
    throw new Base32Error('Encoding failed: ' + error.message, 'ENCODING_ERROR');
  } finally {
    if (options.alphabet) {
      resetToDefaultAlphabet();
    }
  }
}

/**
 * Decodes a Base32 string to Uint8Array
 * @param {string} input - The Base32 string to decode
 * @param {Object} [options] - Decoding options
 * @param {boolean} [options.strict=false] - Whether to enable strict mode
 * @param {boolean} [options.returnString=false] - Whether to return a string instead of Uint8Array
 * @param {string} [options.alphabet] - Optional custom Base32 alphabet
 * @returns {Uint8Array|string} The decoded data
 * @throws {Base32Error} If the input is invalid
 */
function decode(input, options = {}) {
  try {
    if (options.alphabet) {
      setCustomAlphabet(options.alphabet);
    }

    if (typeof input !== 'string') {
      throw new Base32Error('Input must be a string', 'INVALID_INPUT_TYPE');
    }

    const upperInput = input.toUpperCase();
    
    if (options.strict) {
      validatePadding(upperInput);
    }

    const cleanInput = upperInput.replace(/=+$/, '');
    
    if (!/^[A-Z2-7]*$/.test(cleanInput)) {
      throw new Base32Error('Invalid Base32 character', 'INVALID_CHARACTER');
    }

    let bits = 0;
    let value = 0;
    const output = [];

    for (let i = 0; i < cleanInput.length; i++) {
      const charValue = DECODE_TABLE.get(cleanInput[i]);
      value = (value << 5) | charValue;
      bits += 5;

      if (bits >= 8) {
        output.push((value >>> (bits - 8)) & 255);
        bits -= 8;
      }
    }

    const result = new Uint8Array(output);
    return options.returnString ? uint8ArrayToString(result) : result;
  } catch (error) {
    if (error instanceof Base32Error) throw error;
    throw new Base32Error('Decoding failed: ' + error.message, 'DECODING_ERROR');
  } finally {
    if (options.alphabet) {
      resetToDefaultAlphabet();
    }
  }
}

/**
 * Validates if a string is a valid Base32 encoding
 * @param {string} input - The string to validate
 * @param {Object} [options] - Validation options
 * @param {boolean} [options.strict=false] - Whether to enable strict mode
 * @returns {boolean} True if the string is valid Base32
 */
function isValid(input, options = {}) {
  try {
    if (typeof input !== 'string') return false;
    const upperInput = input.toUpperCase();
    
    if (options.strict) {
      validatePadding(upperInput);
    }

    const cleanInput = upperInput.replace(/=+$/, '');
    return /^[A-Z2-7]*$/.test(cleanInput);
  } catch {
    return false;
  }
}

/**
 * Encodes a string as Base32 without padding
 * @param {string|Uint8Array} input - The input to encode
 * @returns {string} The Base32 encoded string without padding
 */
function encodeNoPadding(input) {
  return encode(input, { noPadding: true });
}

/**
 * Normalizes a Base32 string
 * @param {string} input - The Base32 string to normalize
 * @returns {string} The normalized Base32 string
 */
function normalize(input) {
  if (!isValid(input)) {
    throw new Base32Error('Invalid Base32 string', 'INVALID_INPUT');
  }

  const base = input.toUpperCase().replace(/=+$/, '');
  const padding = 8 - (base.length % 8);
  return padding === 8 ? base : base + PADDING.repeat(padding);
}

/**
 * Calculates the length of encoded Base32 string
 * @param {number} inputLength - Length of input in bytes
 * @param {Object} [options] - Options
 * @param {boolean} [options.noPadding=false] - Whether padding will be omitted
 * @returns {number} Length of Base32 encoded string
 */
function encodedLength(inputLength, options = {}) {
  const baseLength = Math.ceil(inputLength * 8 / 5);
  if (options.noPadding) return baseLength;
  return Math.ceil(baseLength / 8) * 8;
}

module.exports = {
  encode,
  decode,
  encodeNoPadding,
  isValid,
  normalize,
  encodedLength,
  Base32Error,
  setCustomAlphabet,
  resetToDefaultAlphabet
};