Function cm_utf8_decode_character

Source/cm_utf8.c:42–94

};

char const* cm_utf8_decode_character(char const* first, char const* last,
                                     unsigned int* pc)
{
  /* We need at least one byte.  */
  if (first == last) {
    return 0;
  }

  /* Count leading ones in the first byte.  */
  unsigned char c = (unsigned char)*first++;
  unsigned char const ones = cm_utf8_ones[c];
  switch (ones) {
    case 0:
      *pc = c;
      return first; /* One-byte character.  */
    case 1:
    case 7:
    case 8:
      return 0; /* Invalid leading byte.  */
    default:
      break;
  }

  /* Extract bits from this multi-byte character.  */
  {
    unsigned int uc = c & cm_utf8_mask[ones];
    int left;
    for (left = ones - 1; left && first != last; --left) {
      c = (unsigned char)*first++;
      if (cm_utf8_ones[c] != 1) {
        return 0;
      }
      uc = (uc << 6) | (c & cm_utf8_mask[1]);
    }

    if (left > 0 || uc < cm_utf8_min[ones]) {
      return 0;
    }

    /* UTF-16 surrogate halves. */
    if (0xD800 <= uc && uc <= 0xDFFF) {
      return 0;
    }

    /* Invalid codepoints. */
    if (0x10FFFF < uc) {
      return 0;
    }

    *pc = uc;
    return first;
  }
}

int cm_utf8_is_valid(char const* s)
{