htmlparser-cvs

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

Update of /cvsroot/htmlparser/htmlparser/src/org/htmlparser
In directory sc8-pr-cvs1.sourceforge.net:/tmp/cvs-serv14744

Modified Files:
	Parser.java 
Log Message:
Further fix to bug #973137 Double-bytes characters are messed after parsing.
Created a proper String based source with the encoding only optionally specified.
A string is no longer converted to a byte array and then back to characters.



Index: Parser.java
===================================================================
RCS file: /cvsroot/htmlparser/htmlparser/src/org/htmlparser/Parser.java,v
retrieving revision 1.94
retrieving revision 1.95
diff -C2 -d -r1.94 -r1.95
*** Parser.java	16 Jun 2004 02:17:25 -0000	1.94
--- Parser.java	3 Jul 2004 13:56:07 -0000	1.95
***************
*** 27,31 ****
  package org.htmlparser;
  
- import java.io.ByteArrayInputStream;
  import java.io.File;
  import java.io.IOException;
--- 27,30 ----
***************
*** 794,802 ****
      /**
       * 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.
--- 793,799 ----
      /**
       * Creates the parser on an input string.
       * @param html The string containing HTML.
!      * @param charset <em>Optional</em>. The character set encoding that will
!      * be reported by {@link #getEncoding}. If charset is <code>null</code>
       * the default character set is used.
       * @return A parser with the <code>html</code> string as input.
***************
*** 804,828 ****
      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);
--- 801,809 ----
      public static Parser createParser (String html, String charset)
      {
          Parser ret;
  
          if (null == html)
              throw new IllegalArgumentException ("html cannot be null");
!         ret = new Parser (new Lexer (new Page (html, charset)));
  
          return (ret);