RE: [Htmlparser-developer] Language Support

[Claude D. <CD...@ar...>]

I've been reading further on the use of InputSource in XML and recall
your interest in using a similar mechanism. In practice it seems easy
enough to do this and I've provided some same code to illustrate. The
InputSource provides either a Reader, InputStream of System ID (usually
a local file name) and they can be checked for existence, in that order.
=20
I raise this issue (in this context) because one of the reasons the XML
community adopted the InputSource was because it could contain
additional information about he character set encoding (which is not
used here but could be). This becomes much more important if you start
considering internationalization.
=20
Here's some sample code that is compilable (but untested - though it
should work):

import java.io.*;
import org.xml.sax.*;
import com.kizna.html.util.HTMLParserException;
 =20
public class InputSourceReader
  extends BufferedReader
{
  public InputSourceReader(InputSource source)
    throws HTMLParserException
  {
    super(getReaderFromInputSource(source));
  }
 =20
  protected static Reader getReaderFromInputSource(InputSource source)
    throws HTMLParserException
  {
    Reader reader =3D source.getCharacterStream();
    if (reader !=3D null)
    {
      return reader;
    }
=20
    InputStream input =3D source.getByteStream();
    if (input !=3D null)
    {
      return new InputStreamReader(input);
    }
   =20
    String systemId =3D source.getSystemId();
    if (systemId !=3D null)
    {
      try
      {
        return new FileReader(systemId);
      }
      catch (FileNotFoundException e)
      {
        throw new HTMLParserException("Invalid InputSource", e);
      }
    }
    throw new HTMLParserException("Invalid InputSource");
  }
}


-----Original Message-----
From: Somik Raha [mailto:so...@ya...]=20
Sent: Monday, August 05, 2002 7:07 PM
To: htm...@li...
Subject: [Htmlparser-developer] Language Support


Hi Folks,
    Amit Rana is a new developer on HTMLParser. He has considerable
experience in internationalization - and he is currently working to
enable language support and switching. Two languages high on my list are
- French and Finnish, considering we've had French and Finnish
developers on this project. We also want to do Japanese support.
    The architecture that Amit is trying is nice - it will simply
require publishing of a standard English properties file - and for any
language support, a corresponding translated properties file will be
loaded up.
    Amit --> you can probably give a more detailed explanation here.
=20
Regards,
Somik