js实现unicode码字符串与utf8字节数据互转详解

js的string变量存储字符串使用的是unicode编码，要保存时必须选择其他编码后进行传输，比如转成utf-8,utf-32等。存储到数据库中为utf-8编码，读取出来如何转换成正确的字符串就成了问题。现在给出解决方案，可以正确支持中文、emoji表情、英文混合的字符串编码互转。

/**
 * created by hdwang on 2019/1/28.
 */
var convertutf8 = (function() {

  /**
   * unicode string to utf-8
   * @param text 字符串
   * @returns {*} utf-8编码
   */
  function tobytes(text) {
    var result = [], i = 0;
    text = encodeuri(text);
    while (i < text.length) {
      var c = text.charcodeat(i++);

      // if it is a % sign, encode the following 2 bytes as a hex value
      if (c === 37) {
        result.push(parseint(text.substr(i, 2), 16))
        i += 2;

        // otherwise, just the actual byte
      } else {
        result.push(c)
      }
    }

    return coercearray(result);
  }


  /**
   * utf8 byte to unicode string
   * @param utf8bytes
   * @returns {string}
   */
  function utf8bytetounicodestr(utf8bytes){
    var unicodestr ="";
    for (var pos = 0; pos < utf8bytes.length;){
      var flag= utf8bytes[pos];
      var unicode = 0 ;
      if ((flag >>>7) === 0 ) {
        unicodestr+= string.fromcharcode(utf8bytes[pos]);
        pos += 1;

      } else if ((flag &0xfc) === 0xfc ){
        unicode = (utf8bytes[pos] & 0x3) << 30;
        unicode |= (utf8bytes[pos+1] & 0x3f) << 24;
        unicode |= (utf8bytes[pos+2] & 0x3f) << 18;
        unicode |= (utf8bytes[pos+3] & 0x3f) << 12;
        unicode |= (utf8bytes[pos+4] & 0x3f) << 6;
        unicode |= (utf8bytes[pos+5] & 0x3f);
        unicodestr+= string.fromcodepoint(unicode) ;
        pos += 6;

      }else if ((flag &0xf8) === 0xf8 ){
        unicode = (utf8bytes[pos] & 0x7) << 24;
        unicode |= (utf8bytes[pos+1] & 0x3f) << 18;
        unicode |= (utf8bytes[pos+2] & 0x3f) << 12;
        unicode |= (utf8bytes[pos+3] & 0x3f) << 6;
        unicode |= (utf8bytes[pos+4] & 0x3f);
        unicodestr+= string.fromcodepoint(unicode) ;
        pos += 5;

      } else if ((flag &0xf0) === 0xf0 ){
        unicode = (utf8bytes[pos] & 0xf) << 18;
        unicode |= (utf8bytes[pos+1] & 0x3f) << 12;
        unicode |= (utf8bytes[pos+2] & 0x3f) << 6;
        unicode |= (utf8bytes[pos+3] & 0x3f);
        unicodestr+= string.fromcodepoint(unicode) ;
        pos += 4;

      } else if ((flag &0xe0) === 0xe0 ){
        unicode = (utf8bytes[pos] & 0x1f) << 12;;
        unicode |= (utf8bytes[pos+1] & 0x3f) << 6;
        unicode |= (utf8bytes[pos+2] & 0x3f);
        unicodestr+= string.fromcharcode(unicode) ;
        pos += 3;

      } else if ((flag &0xc0) === 0xc0 ){ //110
        unicode = (utf8bytes[pos] & 0x3f) << 6;
        unicode |= (utf8bytes[pos+1] & 0x3f);
        unicodestr+= string.fromcharcode(unicode) ;
        pos += 2;

      } else{
        unicodestr+= string.fromcharcode(utf8bytes[pos]);
        pos += 1;
      }
    }
    return unicodestr;
  }



  function checkint(value) {
    return (parseint(value) === value);
  }

  function checkints(arrayish) {
    if (!checkint(arrayish.length)) { return false; }

    for (var i = 0; i < arrayish.length; i++) {
      if (!checkint(arrayish[i]) || arrayish[i] < 0 || arrayish[i] > 255) {
        return false;
      }
    }

    return true;
  }

  function coercearray(arg, copy) {

    // arraybuffer view
    if (arg.buffer && arg.name === 'uint8array') {

      if (copy) {
        if (arg.slice) {
          arg = arg.slice();
        } else {
          arg = array.prototype.slice.call(arg);
        }
      }

      return arg;
    }

    // it's an array; check it is a valid representation of a byte
    if (array.isarray(arg)) {
      if (!checkints(arg)) {
        throw new error('array contains invalid value: ' + arg);
      }

      return new uint8array(arg);
    }

    // something else, but behaves like an array (maybe a buffer? arguments?)
    if (checkint(arg.length) && checkints(arg)) {
      return new uint8array(arg);
    }

    throw new error('unsupported array-like object');
  }

  return {
    tobytes: tobytes,
    frombytes: utf8bytetounicodestr
  }
})()

针对emoji的字节字符，占两个unicode字符。使用string.fromcharcode也可以实现，需要进行两次fromcharcode，没有frompointcode方便。下面展示了utf-8的4字节转换为unicode(utf-16)的过程。

//高char10位[一个unicode字符] (2+6+2=10)
unicode =  ((utf8bytes[pos] & 0x3)) << 8 |((utf8bytes[pos+1] & 0x3f) << 2) |((utf8bytes[pos+2] >> 4) & 0x03);

//减去‭1f600‬中的1，这里减去6个0即可，低位char已经占据10位
unicode = unicode - parseint('1000000',2)

//加上utf-16高char的标识符
unicode = 0xd800 + unicode;
console.log(unicode);
unicodestr += string.fromcharcode(unicode);

//低char10位[一个unicode字符](4+6)
unicode = ((utf8bytes[pos+2] & 0x0f) << 6) | (utf8bytes[pos+3] & 0x3f);
//加上utf-16低char的标识符
unicode = 0xdc00 + unicode;
console.log(unicode);
unicodestr+= string.fromcharcode(unicode);
pos += 4;

以上所述是小编给大家介绍的js实现unicode码字符串与utf8字节数据互转详解整合，希望对大家有所帮助