org.jsoup.nodes.Node#childNodeSize ( )源码实例Demo

private static void readNodes(List<Node> nodeList, List<String> textList)
{
	String tempText;
	for(Node node : nodeList)
	{
		if(node.childNodeSize()>0)
		{
			readNodes(node.childNodes(), textList);
		}
		else
		{
			if(node.nodeName().equals("#text"))
			{
				tempText=((TextNode) node).getWholeText();
				tempText=newline.matcher(tempText).replaceAll("");
				if(!tempText.isEmpty())
					textList.add(tempText);
			}
		}
	}
}
 

private static void readNodesWithTags(List<Node> nodeList, List<Map.Entry<String,String>> textListMap, String tag)
{
	for(Node node : nodeList)
	{
		if(node.childNodeSize()>0)
		{
			readNodesWithTags(node.childNodes(), textListMap, node.nodeName());
		}
		else
		{
			if(node.nodeName().equals("#text"))
			{
				if(tag.equalsIgnoreCase("body"))
					tag="p";
				textListMap.add(new AbstractMap.SimpleEntry<String,String>(tag, ((TextNode) node).getWholeText() ));
			}
		}
	}
}
 

/**
 * Start a depth-first traverse of the root and all of its descendants.
 * @param root the root node point to traverse.
 */
public void traverse(Node root) {
    Node node = root;
    int depth = 0;
    
    while (node != null) {
        visitor.head(node, depth);
        if (node.childNodeSize() > 0) {
            node = node.childNode(0);
            depth++;
        } else {
            while (node.nextSibling() == null && depth > 0) {
                visitor.tail(node, depth);
                node = node.parentNode();
                depth--;
            }
            visitor.tail(node, depth);
            if (node == root)
                break;
            node = node.nextSibling();
        }
    }
}
 

/**
 * Start a depth-first traverse of the root and all of its descendants.
 * @param visitor Node visitor.
 * @param root the root node point to traverse.
 */
public static void traverse(NodeVisitor visitor, Node root) {
    Node node = root;
    int depth = 0;
    
    while (node != null) {
        visitor.head(node, depth);
        if (node.childNodeSize() > 0) {
            node = node.childNode(0);
            depth++;
        } else {
            while (node.nextSibling() == null && depth > 0) {
                visitor.tail(node, depth);
                node = node.parentNode();
                depth--;
            }
            visitor.tail(node, depth);
            if (node == root)
                break;
            node = node.nextSibling();
        }
    }
}
 

/**
 * Start a depth-first traverse of the root and all of its descendants.
 * @param visitor Node visitor.
 * @param root the root node point to traverse.
 */
public static void traverse(NodeVisitor visitor, Node root) {
    Node node = root;
    int depth = 0;
    
    while (node != null) {
        visitor.head(node, depth);
        if (node.childNodeSize() > 0) {
            node = node.childNode(0);
            depth++;
        } else {
            while (node.nextSibling() == null && depth > 0) {
                visitor.tail(node, depth);
                node = node.parentNode();
                depth--;
            }
            visitor.tail(node, depth);
            if (node == root)
                break;
            node = node.nextSibling();
        }
    }
}
 

/**
 * Start a depth-first traverse of the root and all of its descendants.
 * @param root the root node point to traverse.
 */
public void traverse(Node root) {
    Node node = root;
    int depth = 0;
    
    while (node != null) {
        visitor.head(node, depth);
        if (node.childNodeSize() > 0) {
            node = node.childNode(0);
            depth++;
        } else {
            while (node.nextSibling() == null && depth > 0) {
                visitor.tail(node, depth);
                node = node.parent();
                depth--;
            }
            visitor.tail(node, depth);
            if (node == root)
                break;
            node = node.nextSibling();
        }
    }
}
 

@Override
public void head(Node node, int depth)
{
    if (node.childNodeSize() == 0) {
        if (node instanceof TextNode && StringUtil.isBlank(node.outerHtml())) {
            return;
        }
        mergeToResult(node);
        nodes.add(node);
    }
}
 

/**
 * Walk the HTML document node by node, creating annotations and text.
 *
 * @param builder the builder
 * @param root the root
 * @param depth the depth
 */
private void walk(
    final JCasBuilder builder, final Node root, final int depth, final boolean captureText) {
  if (root == null) {
    return;
  }

  final int begin = builder.getCurrentOffset();
  if (captureText) {
    // Generate the text and the annotations
    final String text = mapToText(root);
    if (!Strings.isNullOrEmpty(text)) {
      builder.addText(text);
    }
  }

  List<Annotation> annotations = null;
  if (root instanceof Element) {
    annotations = mapElementToAnnotations(builder.getJCas(), (Element) root);
  }

  // BUG: With multiple mappers depth here is wrong! It puts all mappers at the same depth...
  // (though in fairness they are all the same begin-end and same element too)

  // Walk the children
  if (root.childNodeSize() > 0) {
    for (final Node node : root.childNodes()) {
      walk(builder, node, depth + 1, captureText);
    }
  }

  // Add annotations to the JCas
  final int end = builder.getCurrentOffset();
  if (annotations != null && !annotations.isEmpty()) {
    builder.addAnnotations(annotations, begin, end, depth);
  }
}
 

/**
 * Removes all comments from the {@code node} tree.
 *
 * @param node
 *            a Jsoup node
 */
static void removeCommentsRecursively(Node node) {
    int i = 0;
    while (i < node.childNodeSize()) {
        Node child = node.childNode(i);
        if (child instanceof Comment) {
            child.remove();
        } else {
            removeCommentsRecursively(child);
            i++;
        }
    }
}