return flags;
}


/*
** Checks for proper UTF-8. It uses the method described in:
**   http://en.wikipedia.org/wiki/UTF-8#Invalid_byte_sequences
** except for the "overlong form" of \u0000 which is not considered invalid
** here: Some languages like Java and Tcl use it. For UTF-8 characters
** > 7f, the variable 'c2' not necessary means the previous character.
** It's number of higher 1-bits indicate the number of continuation bytes
** that are expected to be followed. E.g. when 'c2' has a value in the range
** 0xc0..0xdf it means that 'c' is expected to contain the last continuation
** byte of a UTF-8 character. A value 0xe0..0xef means that after 'c' one
** more continuation byte is expected.
*/

int invalid_utf8(const Blob *pContent){
  const unsigned char *z = (unsigned char *) blob_buffer(pContent);
  unsigned int n = blob_size(pContent);
  unsigned char c, c2;

  if( n==0 ) return 0;  /* Empty file -> OK */
  c = *z;
  while( --n>0 ){
    c2 = c;
    c = *++z;
    if( c2>=0x80 ){
      if( (c2!=0xc0) || (c!=0x80) ){
        if( ((c2==0xf4) && (c>=0x90)) ||
            (c2<0xc2) || (c2>0xf4) || ((c&0xc0)!=0x80) ){
   	      return 1; /* Invalid UTF-8 */
        }
      }
      c = (c2 >= 0xe0) ? (c2<<1)+1 : ' ';
    }
  }
  return c>=0x80; /* Last byte must be ASCII. */
}


/*

#define UTF16_LENGTH_MASK_SZ   (LENGTH_MASK_SZ-(sizeof(WCHAR_T)-sizeof(char)))
#define UTF16_LENGTH_MASK      ((1<<UTF16_LENGTH_MASK_SZ)-1)

/*
** This macro is used to swap the byte order of a UTF-16 character in the
** looks_like_utf16() function.
*/
#define UTF16_SWAP(ch)         ((((ch) << 8) & 0xff00) | (((ch) >> 8) & 0xff))
#define UTF16_SWAP_IF(expr,ch) ((expr) ? UTF16_SWAP((ch)) : (ch))

/*
** This function attempts to scan each logical line within the blob to
** determine the type of content it appears to contain.  The return value
** is a combination of one or more of the LOOK_XXX flags (see above):
**

/*
** This function returns an array of bytes representing the byte-order-mark
** for UTF-8.
*/
const unsigned char *get_utf8_bom(int *pnByte){
  static const unsigned char bom[] = {
    0xef, 0xbb, 0xbf, 0x00, 0x00, 0x00
  };
  if( pnByte ) *pnByte = 3;
  return bom;
}

/*
** This function returns non-zero if the blob starts with a UTF-8