vtd-xml解析多次出现的元素

Question

我正在尝试解析以下XML并创建与主要元素相对应的Java对象：

<bookCase>
   <material>wood</material>
   <shelves>12</shelves>
   ...
   <listOfBooks>
      <book name="book1">
         <author>someone</author>
         <pages>200</pages>
         ...
      </book>
      <book name="book2">
         <author>someone else</author>
         <pages>500</pages>
         ...
      </book>
</bookCase>

所以我想创建BookCase对象，它包含一个Book对象列表。

我正在尝试将AutoPilot与XPath一起使用，但是尽管我可以获得主值(material=“material=”，shelves="12")，但我不知道如何遍历listOfBooks并创建两个Book对象。有什么想法吗？

使用下面这个简单的方法，我可以获取两个Book元素的索引，但是我该如何处理它们呢？

private List<Integer> getBooksNodes() throws VTDException {
    final String xpath = "/bookCase/listOfBooks/book";
    ap.selectXPath(xpath);
    final List<Integer> nodeList = new ArrayList<>();
    int node;
    while ((node = ap.evalXPath()) != -1) nodeList.add(node);
    ap.resetXPath();
    return nodeList;
}

如何告诉AutoPilot分别为每本书探索XPath "author“和"pages”的值，以便每次AutoPilot浏览完该节后都能创建一个book对象？

Answer 1

好的，下面是浏览author和pages节点的代码……我并没有假设图书节点必须有author或pages节点……

private List<Integer> getBooksNodes(VTDNav vn) throws VTDException {
    final String xpath = "/bookCase/listOfBooks/book";
    ap.selectXPath(xpath);
    final List<Integer> nodeList = new ArrayList<>();
    int node;
    while ((node = ap.evalXPath()) != -1) {
        nodeList.add(node);
        // the logic that browses the pages and author nodes
        // just print em out...
        // remember vn is automatically moved to the xpath output by  
        // autoPilot
        // we are gonna move the cursor manually now
        if (vn.toElement(FIRST_CHILD,"author")){
           int i = vn.getText();
           if (i!=-1)
              System.out.println(" author is ====>" + vn.toString(i));
           vn.toElement(PARENT);
        }

        if (vn.toElement(FIRST_CHILD,"pages")){
           int i = vn.getText();
           if (i!=-1)
              System.out.println(" author is ====>" + vn.toString(i));
           vn.toElement(PARENT);
        }

        // also remember that in a xpath eval loop, if you are gonna 
        // move the cursor manually
        // the node position going into the cursor logic must be identical 
        // to the node position existing the cursor logic
        // an easy way to accomplish this is 
        // vn.push() ;
        // your code that does manual node traversal
        // vn.pop();
        // but the logic above didn't use them
        // because move cursor to child, then moving back, essentially move 
        // the cursor to the original position
    }
    ap.resetXPath();
    return nodeList;
}