[dom4j-user] Hack for determining the encoding of a document

[Alain Schneble (Realize IT GmbH)]

Hello

We had the same problem of determining the encoding of an xml-document
in our company a year ago. Because we couldn't find any code on the
net that served this purpose we decided to implement a "hack" by ourselves.
Because it seems that the time hasn't changed, I decided to post this code
here. I hope it will be useful (at least as a temporary solution).

Bye
Alain Schneble

Realize IT GmbH
CH - 8640 Rapperswil

so here is the hack:
(it could be that a part of the getEncodingName method comes from
a version of the xerces sources, but I'm not 100% sure anymore...)

public final class CharsetFetcher 
{
 public static Charset getEncoding(URL urlXML) 
 {
  /*
   * let's get the (physical) encoding
   */
  BufferedInputStream bis = new BufferedInputStream(urlXML.openStream());

  byte[] b = new byte[4];
  bis.read(b);
  bis.close();
      
  Object[] encoding = getEncodingName(b, 4);
      
  char[] cbuf = new char[128];
      

  InputStreamReader isr;
      
  /* 
   * this is a hack
   * we need to support only UTF-8 or UTF-16 files
   */
  if (encoding[0] == "UTF-8") 
  {
   isr = new InputStreamReader(urlXML.openStream(), "UTF-8");         
  } 
  else 
  {
   isr = new InputStreamReader(urlXML.openStream(), "UTF-16");
  }
      
  isr.read(cbuf, 0, 128);
  isr.close();
      
  /*
   * ok, now read what's really inside the encoding attribute
   * defined in the xml file
   */
  String s = new String(cbuf);
  s = s.toUpperCase();
  int nEnc = s.indexOf("ENCODING") + "ENCODING".length();
      
  int nQuotBegin = s.indexOf("\"", nEnc) + 1;
  int nQuotEnd = s.indexOf("\"", nQuotBegin);
      
  s = s.substring(nQuotBegin, nQuotEnd);   
      
  /*
   * the encoding name will now become our "Working-Charset"
   * used mainly for saving the file back with the same encoding type
   * as the original file
   */
  return Charset.forName(s);
 }

 /* 
  * Returns the (physical) Encoding name that could be determined based
  * on the byte representations of the characters in the first four bytes
  * of a text file
  */
 private static Object[] getEncodingName(byte[] b4, int count) 
 {
  if (count < 2) 
  {
   return new Object[]{"UTF-8", null};
  }
      
  // UTF-16, with BOM
  int b0 = b4[0] & 0xFF;
  int b1 = b4[1] & 0xFF;
  if (b0 == 0xFE && b1 == 0xFF) 
  {
   // UTF-16, big-endian
   return new Object [] {"UTF-16BE", new Boolean(true)};
  }
  if (b0 == 0xFF && b1 == 0xFE) 
  {
   // UTF-16, little-endian
   return new Object [] {"UTF-16LE", new Boolean(false)};
  }
      
  // default to UTF-8 if we don't have enough bytes to make a
  // good determination of the encoding
  if (count < 3) 
  {
   return new Object [] {"UTF-8", null};
  }
      
  // UTF-8 with a BOM
  int b2 = b4[2] & 0xFF;
  if (b0 == 0xEF && b1 == 0xBB && b2 == 0xBF) 
  {
   return new Object [] {"UTF-8", null};
  }
      
  // default to UTF-8 if we don't have enough bytes to make a
  // good determination of the encoding
  if (count < 4) 
  {
   return new Object [] {"UTF-8", null};
  }
      
  // other encodings
  int b3 = b4[3] & 0xFF;
  if (b0 == 0x00 && b1 == 0x00 && b2 == 0x00 && b3 == 0x3C) 
  {
   // UCS-4, big endian (1234)
   return new Object [] {"ISO-10646-UCS-4", new Boolean(true)};
  }
  if (b0 == 0x3C && b1 == 0x00 && b2 == 0x00 && b3 == 0x00) 
  {
   // UCS-4, little endian (4321)
   return new Object [] {"ISO-10646-UCS-4", new Boolean(false)};
  }
  if (b0 == 0x00 && b1 == 0x00 && b2 == 0x3C && b3 == 0x00) 
  {
   // UCS-4, unusual octet order (2143)
   // REVISIT: What should this be?
   return new Object [] {"ISO-10646-UCS-4", null};
  }
  if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x00) 
  {
   // UCS-4, unusual octect order (3412)
   // REVISIT: What should this be?
   return new Object [] {"ISO-10646-UCS-4", null};
  }
  if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x3F) 
  {
   // UTF-16, big-endian, no BOM
   // (or could turn out to be UCS-2...
   // REVISIT: What should this be?
   return new Object [] {"UTF-16BE", new Boolean(true)};
  }
  if (b0 == 0x3C && b1 == 0x00 && b2 == 0x3F && b3 == 0x00) 
  {
   // UTF-16, little-endian, no BOM
   // (or could turn out to be UCS-2...
   return new Object [] {"UTF-16LE", new Boolean(false)};
  }
  if (b0 == 0x4C && b1 == 0x6F && b2 == 0xA7 && b3 == 0x94) 
  {
   // EBCDIC
   // return CP037 instead of EBCDIC here
   return new Object [] {"CP037", null};
  }
      
  // default encoding
  return new Object [] {"UTF-8", null};
      
 }
}


†+w­zf¢–+,¦‰ìo$…ëFŠ÷›iÉ®ˆ+j`èzĺ·Šàzw­…éiÊ›zj+zšèvç-Š÷ƒ¡ë"¶¥§*.r·šµæÞ×r‡^Hp1Ä,åDjwazZn²¥¥ƒ”
Xœw«zm§ÿì¢êÜyú+éÞ÷h«^ýÚ&â;¬z¹šŠX§‚X¬µÚ&â;¬z¹b²Û,¢êÜyú+éÞm¦Ïÿ–+-²Ê.­Ç¢¸ë–+-³ùb²Ø~Ýn#ºÇ