Java Code Examples for org.apache.xerces.parsers.DOMParser#parse()

/**
 * Parse an entire HTML document or a document fragment. Use lowercase translation for names of tags and attributes.
 * @param document a HTML code to parse.
 * @param encoding an encoding to use for a parser.
 * @return a parsed document representation.
 */
public static Document parseDocument(String document, String encoding) throws IOException, SAXException {
    DOMParser parser = new DOMParser(new HTMLConfiguration());

    try {
        // These URLs are predefined parameters' names (check org.cyberneko.html.HTMLConfiguration for more information)
        parser.setProperty("http://cyberneko.org/html/properties/names/elems", "lower");
        parser.setProperty("http://cyberneko.org/html/properties/default-encoding", encoding);
    } catch (SAXNotRecognizedException | SAXNotSupportedException e) {
        logger.error("Unexpected parser configuration error occurred: " + e.getMessage());
        throw new RuntimeException(e);
    }

    StringReader reader = new StringReader(document);
    InputSource source = new InputSource(reader);
    parser.parse(source);

    return parser.getDocument();
}
 

private List<String> parseXMLPage(Page page) {
    DOMParser parser = new DOMParser();
    try {
        parser.parse(new InputSource(new ByteArrayInputStream(page.getContent())));
    } catch (SAXException | IOException e) {
        throw new RuntimeException("Failed to parse search results.", e);
    }
    Document doc = parser.getDocument();
    NodeList list = doc.getElementsByTagName("d:Url");
    List<String> urls = new ArrayList<String>();
    for (int j = 0; j < list.getLength(); j++) {
        Node node = list.item(j);
        NodeList children = node.getChildNodes();
        Node child = children.item(0);
        urls.add(child.getTextContent());
    }
    return urls;
}
 

@Override
public Document parse() throws SAXException, IOException
{
    //temporay NekoHTML fix until nekohtml gets fixed
    if (!neko_fixed)
    {
        HTMLElements.Element li = HTMLElements.getElement(HTMLElements.LI);
        HTMLElements.Element[] oldparents = li.parent;
        li.parent = new HTMLElements.Element[oldparents.length + 1];
        for (int i = 0; i < oldparents.length; i++)
            li.parent[i] = oldparents[i];
        li.parent[oldparents.length] = HTMLElements.getElement(HTMLElements.MENU);
        neko_fixed = true;
    }
    
    DOMParser parser = new DOMParser(new HTMLConfiguration());
    parser.setProperty("http://cyberneko.org/html/properties/names/elems", "lower");
    if (charset != null)
        parser.setProperty("http://cyberneko.org/html/properties/default-encoding", charset);
    parser.parse(new org.xml.sax.InputSource(getDocumentSource().getInputStream()));
    return parser.getDocument();
}
 

public Document parse() throws SAXException, IOException
{
    DOMParser parser = new DOMParser(new HTMLConfiguration());
    parser.setProperty("http://cyberneko.org/html/properties/names/elems", "lower");
    parser.setProperty("http://cyberneko.org/html/properties/names/attrs", "lower");
    if (charset != null)
        parser.setProperty("http://cyberneko.org/html/properties/default-encoding", charset);
    
    //preparation for filters, not used now
    /*XMLDocumentFilter attributeFilter = new DOMAttributeFilter();
    XMLDocumentFilter[] filters = { attributeFilter };
    parser.setProperty("http://cyberneko.org/html/properties/filters", filters);*/        
    
    parser.parse(new org.xml.sax.InputSource(is));
    doc = parser.getDocument();
    return doc;
}
 

private String getProcessKey(InputStream workflowDefinition) throws Exception
{
    try 
    {
        InputSource inputSource = new InputSource(workflowDefinition);
        DOMParser parser = new DOMParser();
        parser.parse(inputSource);
        Document document = parser.getDocument();
        NodeList elemnts = document.getElementsByTagName("process");
        if (elemnts.getLength() < 1)
        {
            throw new IllegalArgumentException("The input stream does not contain a process definition!");
        }
        NamedNodeMap attributes = elemnts.item(0).getAttributes();
        Node idAttrib = attributes.getNamedItem("id");
        if (idAttrib == null)
        {
            throw new IllegalAccessError("The process definition does not have an id!");
        }
        
        if(activitiUtil.isMultiTenantWorkflowDeploymentEnabled())
        {
            // Workflow-definition is deployed tenant-aware, key should be altered
            return factory.getDomainProcessKey(idAttrib.getNodeValue());
        }
        else
        {
            return idAttrib.getNodeValue();
        }
    }
    finally
    {
        workflowDefinition.close();
    }
}
 

/**
 * Loads an XML file from an InputStream.
 * <br>
 * Note: the source stream is closed internally 
 *
 * @param configurationFileStream the source stream
 * @return the loaded XML document
 * @throws IOException for IO error
 * @throws SAXException for parsing exception
 */
public static Document loadXMLFile( InputStream configurationFileStream ) throws IOException,
                                                                                 SAXException {

    try {
        DOMParser parser = getDomParser();
        parser.parse(new InputSource(configurationFileStream));

        return parser.getDocument();
    } finally {
        IoUtils.closeStream(configurationFileStream);
    }
}
 

public void loadFromXmlFile(
                             InputStream resourceStream,
                             String resourceIdentifier ) {

    try {
        DOMParser parser = new DOMParser();

        // Required settings from the DomParser
        parser.setFeature("http://apache.org/xml/features/dom/defer-node-expansion", false); // otherwise
        parser.setFeature("http://apache.org/xml/features/continue-after-fatal-error", true);
        parser.setFeature("http://apache.org/xml/features/allow-java-encodings", true);
        parser.parse(new InputSource(resourceStream));

        Document doc = parser.getDocument();
        Element rootElement = doc.getDocumentElement();

        //cleanup the properties
        properties.clear();

        //init the current element path
        LinkedList<String> currentElementPath = new LinkedList<String>();

        //start reading the DOM
        NodeList rootElementChildren = rootElement.getChildNodes();
        for (int i = 0; i < rootElementChildren.getLength(); i++) {
            Node rootElementChild = rootElementChildren.item(i);
            if (rootElementChild.getNodeType() == Node.ELEMENT_NODE) {
                readXmlElement(currentElementPath, (Element) rootElementChild);
            }
        }
    } catch (SAXException e) {
        throw new ConfigurationException("Error while parsing config file '" + resourceIdentifier + "'",
                                         e);
    } catch (IOException ioe) {
        throw new ConfigurationException("Error while parsing config file '" + resourceIdentifier + "'",
                                         ioe);
    }
}
 

protected static final Element parseEquipmentConfiguration(final String confXML) {
  Element result = null;
  try {
    DOMParser parser = new DOMParser();
    parser.parse(confXML);
    result = parser.getDocument().getDocumentElement();
  }
  catch (Exception ex) {
    log.error("could not parse configuration document", ex);
  }
  return result;
}
 

public Document parseDocument(InputSource input, boolean validate)
  throws SAXException, IOException
{
  DOMParser parser = new DOMParser();

  parser.setFeature("http://xml.org/sax/features/validation", validate);
  parser.setEntityResolver(this);
  parser.parse(input);

  return parser.getDocument();
}
 

/**
 * Loads an XML file from a String.
 *
 * @param xmlContentsStr the source file as String
 * @return the loaded XML document
 * @throws IOException for IO error
 * @throws SAXException for parsing exception
 */
public static Document loadXML( String xmlContentsStr ) throws IOException,
                                                               SAXException {

    DOMParser parser = getDomParser();
    parser.parse(xmlContentsStr);

    return parser.getDocument();

}