[Htmlparser-cvs] htmlparser/src/org/htmlparser AbstractNode.java,1.16,1.17 Parser.java,1.66,1.67

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

Update of /cvsroot/htmlparser/htmlparser/src/org/htmlparser
In directory sc8-pr-cvs1:/tmp/cvs-serv26197

Modified Files:
	AbstractNode.java Parser.java 
Log Message:
Removed lexer level AbstractNode.
Removed data package from parser level tags.
Separated tag creation from recursion in NodeFactory interface.



Index: AbstractNode.java
===================================================================
RCS file: /cvsroot/htmlparser/htmlparser/src/org/htmlparser/AbstractNode.java,v
retrieving revision 1.16
retrieving revision 1.17
diff -C2 -d -r1.16 -r1.17
*** AbstractNode.java	5 Oct 2003 13:49:40 -0000	1.16
--- AbstractNode.java	20 Oct 2003 01:28:02 -0000	1.17
***************
*** 30,33 ****
--- 30,34 ----
  
  import java.io.Serializable;
+ import org.htmlparser.lexer.Page;
  
  import org.htmlparser.util.NodeList;
***************
*** 36,40 ****
   * AbstractNode, which implements the Node interface, is the base class for all types of nodes, including tags, string elements, etc
   */
! public abstract class AbstractNode implements Node, Serializable {
      /**
       * The beginning position of the tag in the line
--- 37,47 ----
   * AbstractNode, which implements the Node interface, is the base class for all types of nodes, including tags, string elements, etc
   */
! public abstract class AbstractNode implements Node, Serializable
! {
!     /**
!      * The page this node came from.
!      */
!     protected Page mPage;
! 
      /**
       * The beginning position of the tag in the line
***************
*** 59,70 ****
      /**
       * Create an abstract node with the page positions given.
!      * @param begin The starting position of the node.
!      * @param end The ending position of the node.
       */
!     public AbstractNode (int begin, int end)
      {
!         nodeBegin = begin;
!         nodeEnd   = end;
          parent = null;
      }
  
--- 66,81 ----
      /**
       * Create an abstract node with the page positions given.
!      * Remember the page and start & end cursor positions.
!      * @param page The page this tag was read from.
!      * @param start The starting offset of this node within the page.
!      * @param end The ending offset of this node within the page.
       */
!     public AbstractNode (Page page, int start, int end)
      {
!         mPage = page;
!         nodeBegin = start;
!         nodeEnd = end;
          parent = null;
+         children = null;
      }
  
***************
*** 164,177 ****
      /**
       * Returns the beginning position of the tag.
       */
!     public int elementBegin() {
!         return nodeBegin;
      }
  
      /**
       * Returns the ending position fo the tag
       */
!     public int elementEnd() {
!         return nodeEnd;
      }
  
--- 175,210 ----
      /**
       * Returns the beginning position of the tag.
+      * @deprecated Use {@link #getStartPosition}.
       */
!     public int elementBegin()
!     {
!         return (getStartPosition ());
      }
  
      /**
       * Returns the ending position fo the tag
+      * @deprecated Use {@link #getEndPosition}.
       */
!     public int elementEnd()
!     {
!         return (getEndPosition ());
!     }
! 
!     /**
!      * Get the page this node came from.
!      * @return The page that supplied this node.
!      */
!     public Page getPage ()
!     {
!         return (mPage);
!     }
! 
!     /**
!      * Set the page this node came from.
!      * @param page The page that supplied this node.
!      */
!     public void setPage (Page page)
!     {
!         mPage = page;
      }
  

Index: Parser.java
===================================================================
RCS file: /cvsroot/htmlparser/htmlparser/src/org/htmlparser/Parser.java,v
retrieving revision 1.66
retrieving revision 1.67
diff -C2 -d -r1.66 -r1.67
*** Parser.java	13 Oct 2003 21:48:11 -0000	1.66
--- Parser.java	20 Oct 2003 01:28:02 -0000	1.67
***************
*** 1016,1021 ****
      /**
       * Create a new tag node.
!      * This recurses into the created tag by calling the tag's scanner,
!      * if it is in the list of registered scanners.
       * @param lexer The lexer parsing this tag.
       * @param start The beginning position of the tag.
--- 1016,1023 ----
      /**
       * Create a new tag node.
!      * Note that the attributes vector contains at least one element,
!      * which is the tag name (standalone attribute) at position zero.
!      * This can be used to decide which type of node to create, or
!      * gate other processing that may be appropriate.
       * @param lexer The lexer parsing this tag.
       * @param start The beginning position of the tag.
***************
*** 1027,1030 ****
--- 1029,1064 ----
              ParserException
      {
+         return (new Tag (lexer.getPage (), start, end, attributes));
+     }
+ 
+     /**
+      * Scan a new tag node.
+      * Provides composite tags the opportunity to collect their children by
+      * scanning forward using the same lexer that created the composite tag.
+      * On isolating a tag, processing in the lexer is:
+      * <pre><code>
+      * Node node = getNodeFactory ().createTagNode (this, begin, end, attributes);
+      * node = getNodeFactory ().scanTagNode (this, node);
+      * </code></pre>
+      * This two step process, allows a node factory to only handle node
+      * creation if it wishes, and delegate the recursion and scanning of child
+      * nodes to the original factory.
+      * Without giving too much implementation details, the low level lexer node
+      * factory simply returns the same tag, while the higher level parser node
+      * factory checks for a scanner registered for the node type and if there
+      * is one, calls the scanner to create the specific type of node, which
+      * advances the lexer past the children of the node.
+      * @param lexer The lexer that parsed this tag.
+      * @param tag The tag (just) created by createTagNode. Although this is
+      * of type Node, it can safely be cast to the type returned by
+      * {@link #createTagNode createTagNode}.
+      * @return Either the same node or a new node containing children.
+      * In any case the lexer should be positioned to proceed with the isolation
+      * of the next unknown node.
+      */
+     public Node scanTagNode (Lexer lexer, Node tag)
+         throws
+             ParserException
+     {
          String name;
          TagScanner save;
***************
*** 1032,1036 ****
          Tag ret;
          
!         ret = new Tag (lexer.getPage (), start, end, attributes);
          if (!ret.isEndTag ())
          {
--- 1066,1070 ----
          Tag ret;
          
!         ret = (Tag)tag;
          if (!ret.isEndTag ())
          {