Wrong Document Error when adding deep copy node to new document (Resolved)

[Al Johnston]

I am getting a wrong document error that I have tried a number of ways to solve. I'd very much appreciate any help you can give.

    
73           DocumentBuilderFactory domFactory = DocumentBuilderFactory.newInstance();
74           domFactory.setNamespaceAware(true);
75           DocumentBuilder builder = domFactory.newDocumentBuilder();
76           Document doc = builder.parse(new InputSource(new StringReader(xmlString)));
77           XPathFactory factory = XPathFactory.newInstance();
78           XPath xpath = factory.newXPath();
79           XPathExpression expr = xpath.compile("//" + parentNode + "[" + childNodeLabel + "="
80                   + "'" + childNodeValue + "']");
81           NodeList foundNodes = (NodeList) expr.evaluate(doc, XPathConstants.NODESET);
82           System.out.println("Numbers of nodes: " + foundNodes.getLength());
83         //system prints "3" which is correct for the test in question.
84          DOMSource domSource = new DOMSource();
85          doc = builder.newDocument();
86          Element e = doc.createElement("root");
87           for(int i = 0; i < foundNodes.getLength(); i++) {
88               Element el = (Element) foundNodes.item(i);
89              Node n = el.cloneNode(true);               
90               e.appendChild(n);
               
           }
           doc.appendChild(e);
           domSource.setNode((Node) doc.getDocumentElement());
           StringWriter writer = new StringWriter();
           StreamResult results = new StreamResult(writer);
           TransformerFactory tf = TransformerFactory.newInstance();
           Transformer transformer = tf.newTransformer();
           transformer.transform(domSource, results);

The error is thrown at line 90. This code has gone through many iterations before getting to here as I've tried to solve the problem many different ways. The reason for the System.out.println was so that I could see how many records xpath found. I discovered it found 3. Thinking I was saving time, the first time around, I tried:

    
           Node nodes = (Node) expr.evaluate(doc, XPathConstants.NODE);
           domSource.setNode(nodes);

Then I skipped all the Element creation and NodeList iteration. Of course, there are three nodes and I got back only in my String representation when I tried that. Since then I'm trying to add each node (and all its children) to an empty Document. If the code does not make it clear, my purpose here is to Transfomer a Document into a well-formed String representation of the paired down XML Document created from the starting XML Document. Document A has n numbers of parent Nodes and Document B will have n - (Nodes where some child Node != some node_name and some child node != some node_value). Maybe there's an easier way to go about all this?

Appreciate any help.

Regards,
Al

[Paul Clapham]

I would just use the Document's importNode() method to get nodes which can be appended to the document.

[Al Johnston]

Thanks Paul.

I had already tried that as one of the many iterations I went through to get this to work. I would get either a Wrong Document error or a Hierarchical Error depending on what I did after the document.import(node, true) method. That's IF I then directly added the node as a child. Otherwise, I'd get no errors but the document would be empty. So, to fix it all, I had to do this (for anyone else who may have a problem with this in the future):

   
           DocumentBuilderFactory domFactory = DocumentBuilderFactory.newInstance();
           domFactory.setNamespaceAware(true);
           DocumentBuilder builder = domFactory.newDocumentBuilder();
           Document doc = builder.parse(new InputSource(new StringReader(xmlString)));
           XPathFactory factory = XPathFactory.newInstance();
           XPath xpath = factory.newXPath();
           XPathExpression expr = xpath.compile("//" + parentNode + "[" + childNodeLabel + "="
                   + "'" + childNodeValue + "']");

           NodeList foundNodes = (NodeList) expr.evaluate(doc, XPathConstants.NODESET);

           System.out.println("Numbers of nodes: " + foundNodes.getLength());
           //system prints "3" which is correct for the test in question.
           DOMSource domSource = new DOMSource();
           doc = builder.newDocument();
           Element e = doc.createElement("videos");//will change to root variable
           for(int i = 0; i < foundNodes.getLength(); i++) {
               Element el = doc.createElement("video");//will change to parent tag variable
               Node node = foundNodes.item(i);
               Node n = doc.importNode(node, true);
               el.appendChild(n);
               e.appendChild(el);
           }
           doc.appendChild(e);           
           domSource.setNode(doc);
           StringWriter writer = new StringWriter();
           StreamResult results = new StreamResult(writer);
           TransformerFactory tf = TransformerFactory.newInstance();
           Transformer transformer = tf.newTransformer();
           transformer.transform(domSource, results);
           StringBuilder s = new StringBuilder(writer.toString());
           return s.toString();
 

Not to complain or anything, but the API and what it indicates it will do in this case is not very precise. There seem to be many ways to import and add Nodes to Documents but it's not as simple as just saying "import" and having the Node be where the API says the node will be. First of all, if you don't have a root node in the Document, you'll get an error (I did not want a root node but now I have one). Then, if you have a root node and try to import it and add it to the document by saying:

   
Node node_you_deep_copied = another_node.clone(true);
doc.appendChild(node_you_deep_copied);
 

Throws same error as:

   
Node node_you_deep_coped = doc.importNode(another_node, true);
doc.appendChild(node_you_deep_copied);
 

You'll get an error stating that the node is being used in a document other than the one it was created in. Huh? I deep copied it!

The only way I could get it to work is to add the nodes to Elements as children of the Elements and then add the Element (with at least one root value) to the document. There is already a lot of noise in this code. It would be great if the API were simpler and also worked in this case.

Thanks again for your help.

Best,
Al

[Paul Clapham]

Al Johnston wrote:but it's not as simple as just saying "import" and having the Node be where the API says the node will be.

Here's all the API has to say about where the imported node will be:

The returned node has no parent

All that "importNode" does is to make the node belong to the Document. It doesn't insert it anywhere in the document. You have to do that after you import it.

First of all, if you don't have a root node in the Document, you'll get an error (I did not want a root node but now I have one).

Every XML document must have a root element. So if you want an XML document, you have to start by creating a root element. You don't have a choice about that.

[Al Johnston]

Thanks Paul. That makes sense and it has been a long time since I worked with XML in any material way as almost everything here is JSON. But, the gist of my concern was around the following: Do you know why (doc.importNode OR node.clone) in conjunction with doc.appendChild throws a wrong document error? Why must I create an Element to do all this and then append the Element to the doc?

[Paul Clapham]

No, sorry, I don't really know. I do as little work with the W3C DOM as possible; if there's another way to do what I want, I'm going to choose that other way.