Method _decodeCodePointsFromUTF8Bytes

src/TextEncoder.js:194–257

(bytes)

  }

  static _decodeCodePointsFromUTF8Bytes (bytes) {
    // In the worst case byte needs to be represented by one code point
    // Create a fixed size array that gets sliced at the end
    const size = bytes.length
    const codePoints = new Array(size)

    let remainingBytes = 0
    let i = -1
    let j = 0
    let byte, codePoint

    while (++i < size) {
      byte = bytes[i]

      if (byte > 0b01111111 && byte <= 0b10111111) {
        // Continuation byte identified
        if (--remainingBytes < 0) {
          throw new TextEncodingError(
            'Invalid UTF-8 encoded text: ' +
            `Unexpected continuation byte at 0x${i.toString(16)}`, i)
        }

        // Append bits to current code point
        codePoint = (codePoint << 6) | (byte & 0x3F)

        if (remainingBytes === 0) {
          // Completed a code point
          codePoints[j++] = codePoint
        }
      } else if (remainingBytes > 0) {
        // this must be a continuation byte
        throw new TextEncodingError(
          'Invalid UTF-8 encoded text: ' +
          `Continuation byte expected at 0x${i.toString(16)}`, i)
      } else if (byte <= 0b01111111) {
        // 1 byte code point
        codePoints[j++] = byte
      } else if (byte <= 0b11011111) {
        // 2 byte code point
        codePoint = byte & 0b00011111
        remainingBytes = 1
      } else if (byte <= 0b11101111) {
        // 3 byte code point
        codePoint = byte & 0b00001111
        remainingBytes = 2
      } else if (byte <= 0b11110111) {
        // 4 byte code point
        codePoint = byte & 0b00000111
        remainingBytes = 3
      } else {
        throw new TextEncodingError(
          'Invalid UTF-8 encoded text: ' +
          `Invalid byte ${byte} at 0x${i.toString(16)}`, i)
      }
    }

    if (remainingBytes !== 0) {
      throw new TextEncodingError(