Method codePointsFromString

src/TextEncoder.js:83–122

* Creates an array containing the numeric code points of each Unicode * character in the string. While JavaScript uses UCS-2 internally, * this function will convert a pair of surrogate halves (each of which * UCS-2 exposes as separate characters) into a single code point, * matching UTF

(string)

   * @returns {number[]} Array of Unicode code points
   */
  static codePointsFromString (string) {
    // In the worst case every string code unit needs to be translated to
    // a single code point each
    // Create a fixed size array that gets sliced at the end
    const length = string.length
    const codePoints = new Array(length)

    let codeUnit, nextCodeUnit
    let j = 0
    let i = 0

    while (i < length) {
      codeUnit = string.charCodeAt(i++)

      if (codeUnit >= 0xD800 && codeUnit <= 0xDBFF && i < length) {
        // Identified a high surrogate
        nextCodeUnit = string.charCodeAt(i++)

        // There is a next character
        if ((nextCodeUnit & 0xFC00) === 0xDC00) {
          // Low surrogate
          codePoints[j++] =
            ((codeUnit & 0x3FF) << 10) +
            (nextCodeUnit & 0x3FF) +
            0x10000
        } else {
          // Unmatched surrogate; Only append this code unit, in case
          // the next code unit is the high surrogate of a surrogate pair
          codePoints[j++] = codeUnit
          i--
        }
      } else {
        // Identified BMP character
        codePoints[j++] = codeUnit
      }
    }

    // Slice the fixed size array to the portion actually in use
    return codePoints.slice(0, j)
  }

  /**
   * Encodes Unicode code points to bytes using given encoding.