htmlparser-cvs

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

Update of /cvsroot/htmlparser/htmlparser/src/org/htmlparser/scanners
In directory sc8-pr-cvs1.sourceforge.net:/tmp/cvs-serv32670/src/org/htmlparser/scanners

Modified Files:
	CompositeTagScanner.java Scanner.java ScriptScanner.java 
	StyleScanner.java TagScanner.java 
Log Message:
Part four of a multiphase refactoring.
Most internals now use the Tag interface.
This interface has been broadened to add set/get scanner and set/get endtag.
Removed the org.htmlparser.tags.Tag class and moved the remaining (minor) functionality
to the TagNode class. So now tags inherit directly from TagNode or CompositeTag.
** NOTE: If you have subclassed org.htmlparser.tags.Tag, use org.htmlparser.nodes.TagNode now.**
Removed deprecated methods getTagBegin/getTagEnd and deleted unused classes:
PeekingIterator and it's Implementation.



Index: CompositeTagScanner.java
===================================================================
RCS file: /cvsroot/htmlparser/htmlparser/src/org/htmlparser/scanners/CompositeTagScanner.java,v
retrieving revision 1.86
retrieving revision 1.87
diff -C2 -d -r1.86 -r1.87
*** CompositeTagScanner.java	24 May 2004 16:18:20 -0000	1.86
--- CompositeTagScanner.java	2 Jul 2004 00:49:28 -0000	1.87
***************
*** 31,39 ****
  import org.htmlparser.Attribute;
  import org.htmlparser.Node;
  import org.htmlparser.lexer.Lexer;
  import org.htmlparser.lexer.Page;
  import org.htmlparser.scanners.Scanner;
- import org.htmlparser.tags.CompositeTag;
- import org.htmlparser.tags.Tag;
  import org.htmlparser.util.NodeList;
  import org.htmlparser.util.ParserException;
--- 31,39 ----
  import org.htmlparser.Attribute;
  import org.htmlparser.Node;
+ import org.htmlparser.Tag;
  import org.htmlparser.lexer.Lexer;
  import org.htmlparser.lexer.Page;
+ import org.htmlparser.nodes.TagNode;
  import org.htmlparser.scanners.Scanner;
  import org.htmlparser.util.NodeList;
  import org.htmlparser.util.ParserException;
***************
*** 101,107 ****
          String name;
          Scanner scanner;
!         CompositeTag ret;
          
!         ret = (CompositeTag)tag;
  
          if (ret.isEmptyXmlTag ())
--- 101,107 ----
          String name;
          Scanner scanner;
!         Tag ret;
          
!         ret = tag;
  
          if (ret.isEmptyXmlTag ())
***************
*** 143,159 ****
                                  {
                                      // fake recursion:
!                                     if ((scanner == this) && (next instanceof CompositeTag))
                                      {
!                                         CompositeTag ondeck = (CompositeTag)next;
!                                         if (ondeck.isEmptyXmlTag ())
                                          {
!                                             ondeck.setEndTag (ondeck);
!                                             finishTag (ondeck, lexer);
!                                             addChild (ret, ondeck);
                                          }
                                          else
                                          {
                                              stack.add (ret);
!                                             ret = ondeck;
                                          }
                                      }
--- 143,158 ----
                                  {
                                      // fake recursion:
!                                     if (scanner == this)
                                      {
!                                         if (next.isEmptyXmlTag ())
                                          {
!                                             next.setEndTag (next);
!                                             finishTag (next, lexer);
!                                             addChild (ret, next);
                                          }
                                          else
                                          {
                                              stack.add (ret);
!                                             ret = next;
                                          }
                                      }
***************
*** 192,196 ****
                                  attributes.addElement (new Attribute (name, null));
                                  Tag opener = (Tag)lexer.getNodeFactory ().createTagNode (
!                                     next.getPage (), next.getStartPosition (), next.getEndPosition (),
                                      attributes);
  
--- 191,195 ----
                                  attributes.addElement (new Attribute (name, null));
                                  Tag opener = (Tag)lexer.getNodeFactory ().createTagNode (
!                                     lexer.getPage (), next.getStartPosition (), next.getEndPosition (),
                                      attributes);
  
***************
*** 202,208 ****
                                      for (int i = stack.size () - 1; (-1 == index) && (i >= 0); i--)
                                      {
!                                         // short circuit here... assume everything on the stack is a CompositeTag and has this as it's scanner
                                          // we'll need to stop if either of those conditions isn't met
!                                         CompositeTag boffo = (CompositeTag)stack.elementAt (i);
                                          if (name.equals (boffo.getTagName ()))
                                              index = i;
--- 201,207 ----
                                      for (int i = stack.size () - 1; (-1 == index) && (i >= 0); i--)
                                      {
!                                         // short circuit here... assume everything on the stack has this as it's scanner
                                          // we'll need to stop if either of those conditions isn't met
!                                         Tag boffo = (Tag)stack.elementAt (i);
                                          if (name.equals (boffo.getTagName ()))
                                              index = i;
***************
*** 214,225 ****
                                          // finish off the current one first
                                          finishTag (ret, lexer);
!                                         addChild ((CompositeTag)stack.elementAt (stack.size () - 1), ret);
                                          for (int i = stack.size () - 1; i > index; i--)
                                          {
!                                             CompositeTag fred = (CompositeTag)stack.remove (i);
                                              finishTag (fred, lexer);
!                                             addChild ((CompositeTag)stack.elementAt (i - 1), fred);
                                          }
!                                         ret = (CompositeTag)stack.remove (index);
                                          node = null;
                                      }
--- 213,224 ----
                                          // finish off the current one first
                                          finishTag (ret, lexer);
!                                         addChild ((Tag)stack.elementAt (stack.size () - 1), ret);
                                          for (int i = stack.size () - 1; i > index; i--)
                                          {
!                                             Tag fred = (Tag)stack.remove (i);
                                              finishTag (fred, lexer);
!                                             addChild ((Tag)stack.elementAt (i - 1), fred);
                                          }
!                                         ret = (Tag)stack.remove (index);
                                          node = null;
                                      }
***************
*** 247,253 ****
                          {
                              node = stack.elementAt (depth - 1);
!                             if (node instanceof CompositeTag)
                              {
!                                 CompositeTag precursor = (CompositeTag)node;
                                  scanner = precursor.getThisScanner ();
                                  if (scanner == this)
--- 246,252 ----
                          {
                              node = stack.elementAt (depth - 1);
!                             if (node instanceof Tag)
                              {
!                                 Tag precursor = (Tag)node;
                                  scanner = precursor.getThisScanner ();
                                  if (scanner == this)
***************
*** 295,311 ****
       * @param lexer A lexer positioned at the end of the tag.
       */
!     protected void finishTag (CompositeTag tag, Lexer lexer)
          throws
              ParserException
      {
          if (null == tag.getEndTag ())
!             tag.setEndTag (createVirtualEndTag (tag, lexer.getPage (), lexer.getCursor ().getPosition ()));
          tag.getEndTag ().setParent (tag);
          tag.doSemanticAction ();
      }
!     
      /**
       * Creates an end tag with the same name as the given tag.
       * @param tag The tag to end.
       * @param page The page the tag is on (virtually).
       * @param position The offset into the page at which the tag is to
--- 294,311 ----
       * @param lexer A lexer positioned at the end of the tag.
       */
!     protected void finishTag (Tag tag, Lexer lexer)
          throws
              ParserException
      {
          if (null == tag.getEndTag ())
!             tag.setEndTag (createVirtualEndTag (tag, lexer, lexer.getPage (), lexer.getCursor ().getPosition ()));
          tag.getEndTag ().setParent (tag);
          tag.doSemanticAction ();
      }
! 
      /**
       * Creates an end tag with the same name as the given tag.
       * @param tag The tag to end.
+      * @param lexer The object containg the node factory.
       * @param page The page the tag is on (virtually).
       * @param position The offset into the page at which the tag is to
***************
*** 315,319 ****
       * equal may be used to distinguish it as a virtual tag later on.
       */
!     protected Tag createVirtualEndTag (Tag tag, Page page, int position)
      {
          Tag ret;
--- 315,321 ----
       * equal may be used to distinguish it as a virtual tag later on.
       */
!     protected Tag createVirtualEndTag (Tag tag, Lexer lexer, Page page, int position)
!         throws
!             ParserException
      {
          Tag ret;
***************
*** 321,328 ****
          Vector attributes;
          
!         name = "/" + tag.getRawTagName();
          attributes = new Vector ();
          attributes.addElement (new Attribute (name, (String)null));
!         ret = new Tag (page, position, position, attributes);
          
          return (ret);
--- 323,331 ----
          Vector attributes;
          
!         name = "/" + tag.getRawTagName ();
          attributes = new Vector ();
          attributes.addElement (new Attribute (name, (String)null));
!         ret = (Tag)lexer.getNodeFactory ().createTagNode (
!                                     page, position, position, attributes);
          
          return (ret);

Index: StyleScanner.java
===================================================================
RCS file: /cvsroot/htmlparser/htmlparser/src/org/htmlparser/scanners/StyleScanner.java,v
retrieving revision 1.35
retrieving revision 1.36
diff -C2 -d -r1.35 -r1.36
*** StyleScanner.java	14 Jun 2004 00:06:52 -0000	1.35
--- StyleScanner.java	2 Jul 2004 00:49:28 -0000	1.36
***************
*** 36,41 ****
  import org.htmlparser.lexer.Cursor;
  import org.htmlparser.lexer.Lexer;
! import org.htmlparser.tags.CompositeTag;
! import org.htmlparser.tags.Tag;
  import org.htmlparser.util.NodeList;
  import org.htmlparser.util.ParserException;
--- 36,40 ----
  import org.htmlparser.lexer.Cursor;
  import org.htmlparser.lexer.Lexer;
! import org.htmlparser.Tag;
  import org.htmlparser.util.NodeList;
  import org.htmlparser.util.ParserException;
***************
*** 75,79 ****
          NodeFactory factory;
          Text content;
!         CompositeTag ret;
  
          done = false;
--- 74,78 ----
          NodeFactory factory;
          Text content;
!         Tag ret;
  
          done = false;
***************
*** 118,123 ****
              // build new end tag if required
              if (null == end)
!                 end = new Tag (lexer.getPage (), endpos, endpos, new Vector ());
!             ret = (CompositeTag)tag;
              ret.setEndTag (end);
              ret.setChildren (new NodeList (content));
--- 117,123 ----
              // build new end tag if required
              if (null == end)
!                 end = (Tag)lexer.getNodeFactory ().createTagNode (
!                     lexer.getPage (), endpos, endpos, new Vector ());
!             ret = tag;
              ret.setEndTag (end);
              ret.setChildren (new NodeList (content));

Index: ScriptScanner.java
===================================================================
RCS file: /cvsroot/htmlparser/htmlparser/src/org/htmlparser/scanners/ScriptScanner.java,v
retrieving revision 1.58
retrieving revision 1.59
diff -C2 -d -r1.58 -r1.59
*** ScriptScanner.java	14 Jun 2004 00:06:52 -0000	1.58
--- ScriptScanner.java	2 Jul 2004 00:49:28 -0000	1.59
***************
*** 33,43 ****
  import org.htmlparser.PrototypicalNodeFactory;
  import org.htmlparser.Remark;
  import org.htmlparser.Text;
  import org.htmlparser.lexer.Cursor;
  import org.htmlparser.lexer.Lexer;
  import org.htmlparser.scanners.ScriptDecoder;
- import org.htmlparser.tags.CompositeTag;
  import org.htmlparser.tags.ScriptTag;
- import org.htmlparser.tags.Tag;
  import org.htmlparser.util.NodeList;
  import org.htmlparser.util.ParserException;
--- 33,42 ----
  import org.htmlparser.PrototypicalNodeFactory;
  import org.htmlparser.Remark;
+ import org.htmlparser.Tag;
  import org.htmlparser.Text;
  import org.htmlparser.lexer.Cursor;
  import org.htmlparser.lexer.Lexer;
  import org.htmlparser.scanners.ScriptDecoder;
  import org.htmlparser.tags.ScriptTag;
  import org.htmlparser.util.NodeList;
  import org.htmlparser.util.ParserException;
***************
*** 80,84 ****
          NodeFactory factory;
          Text content;
!         CompositeTag ret;
  
          done = false;
--- 79,83 ----
          NodeFactory factory;
          Text content;
!         Tag ret;
  
          done = false;
***************
*** 135,140 ****
              // build new end tag if required
              if (null == end)
!                 end = new Tag (lexer.getPage (), endpos, endpos, new Vector ());
!             ret = (CompositeTag)tag;
              ret.setEndTag (end);
              ret.setChildren (new NodeList (content));
--- 134,140 ----
              // build new end tag if required
              if (null == end)
!                 end = (Tag)lexer.getNodeFactory ().createTagNode (
!                     lexer.getPage (), endpos, endpos, new Vector ());
!             ret = tag;
              ret.setEndTag (end);
              ret.setChildren (new NodeList (content));

Index: Scanner.java
===================================================================
RCS file: /cvsroot/htmlparser/htmlparser/src/org/htmlparser/scanners/Scanner.java,v
retrieving revision 1.1
retrieving revision 1.2
diff -C2 -d -r1.1 -r1.2
*** Scanner.java	20 Dec 2003 23:47:55 -0000	1.1
--- Scanner.java	2 Jul 2004 00:49:28 -0000	1.2
***************
*** 27,32 ****
  package org.htmlparser.scanners;
  
  import org.htmlparser.lexer.Lexer;
- import org.htmlparser.tags.Tag;
  import org.htmlparser.util.NodeList;
  import org.htmlparser.util.ParserException;
--- 27,32 ----
  package org.htmlparser.scanners;
  
+ import org.htmlparser.Tag;
  import org.htmlparser.lexer.Lexer;
  import org.htmlparser.util.NodeList;
  import org.htmlparser.util.ParserException;

Index: TagScanner.java
===================================================================
RCS file: /cvsroot/htmlparser/htmlparser/src/org/htmlparser/scanners/TagScanner.java,v
retrieving revision 1.53
retrieving revision 1.54
diff -C2 -d -r1.53 -r1.54
*** TagScanner.java	20 Dec 2003 23:47:55 -0000	1.53
--- TagScanner.java	2 Jul 2004 00:49:28 -0000	1.54
***************
*** 29,34 ****
  import java.io.Serializable;
  
  import org.htmlparser.lexer.Lexer;
- import org.htmlparser.tags.Tag;
  import org.htmlparser.util.NodeList;
  import org.htmlparser.util.ParserException;
--- 29,34 ----
  import java.io.Serializable;
  
+ import org.htmlparser.Tag;
  import org.htmlparser.lexer.Lexer;
  import org.htmlparser.util.NodeList;
  import org.htmlparser.util.ParserException;