htmlparser-cvs

[Derrick O. <der...@us...>]

Update of /cvsroot/htmlparser/htmlparser/src/org/htmlparser
In directory sc8-pr-cvs1.sourceforge.net:/tmp/cvs-serv9637

Modified Files:
	Parser.java 
Log Message:
Fix bug #973137 Double-bytes characters are messed after parsing.
Add an encoding parameter to the static createParser() method.



Index: Parser.java
===================================================================
RCS file: /cvsroot/htmlparser/htmlparser/src/org/htmlparser/Parser.java,v
retrieving revision 1.93
retrieving revision 1.94
diff -C2 -d -r1.93 -r1.94
*** Parser.java	14 Jun 2004 01:26:51 -0000	1.93
--- Parser.java	16 Jun 2004 02:17:25 -0000	1.94
***************
*** 27,33 ****
--- 27,35 ----
  package org.htmlparser;
  
+ import java.io.ByteArrayInputStream;
  import java.io.File;
  import java.io.IOException;
  import java.io.Serializable;
+ import java.io.UnsupportedEncodingException;
  import java.net.MalformedURLException;
  import java.net.URL;
***************
*** 792,807 ****
      /**
       * Creates the parser on an input string.
!      * @param inputHTML
!      * @return Parser
       */
!     public static Parser createParser(String inputHTML)
      {
!         Lexer lexer;
          Parser ret;
  
!         if (null == inputHTML)
              throw new IllegalArgumentException ("html cannot be null");
!         lexer = new Lexer (new Page (inputHTML));
!         ret = new Parser (lexer);
  
          return (ret);
--- 794,828 ----
      /**
       * Creates the parser on an input string.
!      * Uses the character set encoding to create a stream of bytes that is
!      * fed into the parser as if it had come off the wire.
!      * @param html The string containing HTML.
!      * @param charset Character set encoding to use when converting the
!      * <code>html</code> to a stream of bytes. If charset is <code>null</code>
!      * the default character set is used.
!      * @return A parser with the <code>html</code> string as input.
       */
!     public static Parser createParser (String html, String charset)
      {
!         ByteArrayInputStream stream;
          Parser ret;
  
!         if (null == html)
              throw new IllegalArgumentException ("html cannot be null");
!         if (null == charset)
!             charset = Page.DEFAULT_CHARSET;
!         try
!         {
!             stream = new ByteArrayInputStream (html.getBytes (charset));
!             ret = new Parser (new Lexer (new Page (stream, charset)));
!         }
!         catch (UnsupportedEncodingException uee)
!         {
!             String msg;
!             
!             msg = uee.getMessage ();
!             if (null == msg)
!                 msg = "unsupported encoding (" + charset + ") exception";
!             ret = new Parser (new Lexer (new Page (msg)));
!         }
  
          return (ret);