unicode.cc.svn-base

Google浏览器V8内核代码

        result[i] = mapping.chars[i];
      return mapping.length;
    } else {
      // Low bits 2 means a really really special case
      if (allow_caching_ptr) *allow_caching_ptr = false;
      // The cases of this switch are defined in unicode.py in the
      // really_special_cases mapping.
      switch (value >> 2) {
        case 1:
          // Really special case 1: upper case sigma.  This letter
          // converts to two different lower case sigmas depending on
          // whether or not it occurs at the end of a word.
          if (next != 0 && Letter::Is(next)) {
            result[0] = 0x03C3;
          } else {
            result[0] = 0x03C2;
          }
          return 1;
        default:
          return 0;
      }
      return -1;
    }
  } else {
    return 0;
  }
}

uchar Utf8::CalculateValue(const byte* str,
                           unsigned length,
                           unsigned* cursor) {
  static const uchar kMaxOneByteChar = 0x7F;
  static const uchar kMaxTwoByteChar = 0x7FF;
  static const uchar kMaxThreeByteChar = 0xFFFF;
  static const uchar kMaxFourByteChar = 0x1FFFFF;

  // We only get called for non-ascii characters.
  if (length == 1) {
    *cursor += 1;
    return kBadChar;
  }
  int first = str[0];
  int second = str[1] ^ 0x80;
  if (second & 0xC0) {
    *cursor += 1;
    return kBadChar;
  }
  if (first < 0xE0) {
    if (first < 0xC0) {
      *cursor += 1;
      return kBadChar;
    }
    uchar l = ((first << 6) | second) & kMaxTwoByteChar;
    if (l <= kMaxOneByteChar) {
      *cursor += 1;
      return kBadChar;
    }
    *cursor += 2;
    return l;
  }
  if (length == 2) {
    *cursor += 1;
    return kBadChar;
  }
  int third = str[2] ^ 0x80;
  if (third & 0xC0) {
    *cursor += 1;
    return kBadChar;
  }
  if (first < 0xF0) {
    uchar l = ((((first << 6) | second) << 6) | third) & kMaxThreeByteChar;
    if (l <= kMaxTwoByteChar) {
      *cursor += 1;
      return kBadChar;
    }
    *cursor += 3;
    return l;
  }
  if (length == 3) {
    *cursor += 1;
    return kBadChar;
  }
  int fourth = str[3] ^ 0x80;
  if (fourth & 0xC0) {
    *cursor += 1;
    return kBadChar;
  }
  if (first < 0xF8) {
    uchar l = (((((first << 6 | second) << 6) | third) << 6) | fourth) &
              kMaxFourByteChar;
    if (l <= kMaxThreeByteChar) {
      *cursor += 1;
      return kBadChar;
    }
    *cursor += 4;
    return l;
  }
  *cursor += 1;
  return kBadChar;
}

const byte* Utf8::ReadBlock(Buffer<const char*> str, byte* buffer,
    unsigned capacity, unsigned* chars_read_ptr, unsigned* offset_ptr) {
  unsigned offset = *offset_ptr;
  // Bail out early if we've reached the end of the string.
  if (offset == str.length()) {
    *chars_read_ptr = 0;
    return NULL;
  }
  const byte* data = reinterpret_cast<const byte*>(str.data());
  if (data[offset] <= kMaxOneByteChar) {
    // The next character is an ascii char so we scan forward over
    // the following ascii characters and return the next pure ascii
    // substring
    const byte* result = data + offset;
    offset++;
    while ((offset < str.length()) && (data[offset] <= kMaxOneByteChar))
      offset++;
    *chars_read_ptr = offset - *offset_ptr;
    *offset_ptr = offset;
    return result;
  } else {
    // The next character is non-ascii so we just fill the buffer
    unsigned cursor = 0;
    unsigned chars_read = 0;
    while (offset < str.length()) {
      uchar c = data[offset];
      if (c <= kMaxOneByteChar) {
        // Fast case for ascii characters
        if (!CharacterStream::EncodeAsciiCharacter(c,
                                                   buffer,
                                                   capacity,
                                                   cursor))
          break;
        offset += 1;
      } else {
        unsigned chars = 0;
        c = Utf8::ValueOf(data + offset, str.length() - offset, &chars);
        if (!CharacterStream::EncodeNonAsciiCharacter(c,
                                                      buffer,
                                                      capacity,
                                                      cursor))
          break;
        offset += chars;
      }
      chars_read++;
    }
    *offset_ptr = offset;
    *chars_read_ptr = chars_read;
    return buffer;
  }
}