src/java/org/apache/log4j/xml/XMLDecoder.java

/*
 * Copyright 1999,2004 The Apache Software Foundation.
 * 
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 * 
 *      http://www.apache.org/licenses/LICENSE-2.0
 * 
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package org.apache.log4j.xml;

import java.awt.Component;
import java.io.IOException;
import java.io.InputStreamReader;
import java.io.LineNumberReader;
import java.io.StringReader;
import java.net.URL;
import java.util.HashMap;
import java.util.Hashtable;
import java.util.Iterator;
import java.util.Map;
import java.util.Vector;

import javax.swing.ProgressMonitorInputStream;
import javax.xml.parsers.DocumentBuilder;
import javax.xml.parsers.DocumentBuilderFactory;
import javax.xml.parsers.ParserConfigurationException;

import org.apache.log4j.Level;
import org.apache.log4j.Logger;
import org.apache.log4j.spi.Decoder;
import org.apache.log4j.spi.LoggingEvent;
import org.apache.log4j.spi.ThrowableInformation;
import org.apache.log4j.spi.location.LocationInfo;
import org.w3c.dom.Document;
import org.w3c.dom.Node;
import org.w3c.dom.NodeList;
import org.xml.sax.InputSource;


/**
 * Decodes Logging Events in XML formated into elements that are used by
 * Chainsaw.
 *
 * This decoder can process a collection of log4j:event nodes ONLY 
 * (no XML declaration nor eventSet node)
 * 
 * NOTE:  Only a single LoggingEvent is returned from the decode method
 * even though the DTD supports multiple events nested in an eventSet.
 *
 * @author Scott Deboy <sdeboy@apache.org>
 * @author Paul Smith <psmith@apache.org>
 *
 */
public class XMLDecoder implements Decoder {
  private static final String BEGINPART =
    "<?xml version=\"1.0\" encoding=\"UTF-8\" ?><!DOCTYPE log4j:eventSet SYSTEM \"log4j.dtd\"><log4j:eventSet version=\"1.2\" xmlns:log4j=\"http://jakarta.apache.org/log4j/\">";
  private static final String ENDPART = "</log4j:eventSet>";
  private static final String RECORD_END = "</log4j:event>";
  private DocumentBuilderFactory dbf;
  private DocumentBuilder docBuilder;
  private Map additionalProperties = new HashMap();
  private String partialEvent;
  private Component owner = null;

  public XMLDecoder(Component owner) {
      this();
      this.owner = owner;
  }
  public XMLDecoder() {
    dbf = DocumentBuilderFactory.newInstance();
    dbf.setValidating(false);

    try {
      docBuilder = dbf.newDocumentBuilder();
      docBuilder.setErrorHandler(new SAXErrorHandler());
      docBuilder.setEntityResolver(new Log4jEntityResolver());
    } catch (ParserConfigurationException pce) {
      System.err.println("Unable to get document builder");
    }
  }

  /**
   * Sets an additionalProperty map, where each Key/Value pair is
   * automatically added to each LoggingEvent as it is decoded.
   *
   * This is useful, say, to include the source file name of the Logging events
   * @param additionalProperties
   */
  public void setAdditionalProperties(Map additionalProperties) {
    this.additionalProperties = additionalProperties;
  }

  /**
   * Converts the LoggingEvent data in XML string format into an actual
   * XML Document class instance.
   * @param data
   * @return dom document
   */
  private Document parse(String data) {
    if (docBuilder == null || data == null) {
      return null;
    }
    Document document = null;

    try {
      // we change the system ID to a valid URI so that Crimson won't
      // complain. Indeed, "log4j.dtd" alone is not a valid URI which
      // causes Crimson to barf. The Log4jEntityResolver only cares
      // about the "log4j.dtd" ending.
      //      buf.setLength(0);

      /**
       * resetting the length of the StringBuffer is dangerous, particularly
       * on some JDK 1.4 impls, there's a known Bug that causes a memory leak
       */
      StringBuffer buf = new StringBuffer(1024);

      buf.append(BEGINPART);
      buf.append(data);
      buf.append(ENDPART);

      InputSource inputSource =
        new InputSource(new StringReader(buf.toString()));
      inputSource.setSystemId("dummy://log4j.dtd");
      document = docBuilder.parse(inputSource);
    } catch (Exception e) {
      e.printStackTrace();
    }

    return document;
  }

  /**
   * Decodes a File into a Vector of LoggingEvents
   * @param url the url of a file containing events to decode
   * @return Vector of LoggingEvents
   * @throws IOException
   */
  public Vector decode(URL url) throws IOException {
    LineNumberReader reader = null;
    if (owner != null) {
      reader = new LineNumberReader(new InputStreamReader(new ProgressMonitorInputStream(owner, "Loading " + url , url.openStream())));
    } else {
      reader = new LineNumberReader(new InputStreamReader(url.openStream()));
    }

    Vector v = new Vector();

    String line = null;
	Vector events = null;
    try {
        while ((line = reader.readLine()) != null) {
            StringBuffer buffer = new StringBuffer(line);
            for (int i = 0;i<1000;i++) {
                buffer.append(reader.readLine()).append("\n");
            }
			events = decodeEvents(buffer.toString());
            if (events != null)
            	v.addAll(events);
        }
    } finally {
      partialEvent = null;
      try {
        if (reader != null) {
          reader.close();
        }
      } catch (Exception e) {
        e.printStackTrace();
      }
    }
    return v;
  }

  public Vector decodeEvents(String document) {
    if (document != null) {
      document = document.trim();
      if (document.equals("")) {
        return null;
      }
	  	String newDoc=null;
	  	String newPartialEvent=null;
	  	//separate the string into the last portion ending with </log4j:event> (which will
	  	//be processed) and the partial event which will be combined and processed in the next section
	
		//if the document does not contain a record end, append it to the partial event string
	  	if (document.lastIndexOf(RECORD_END) == -1) {
			partialEvent = partialEvent + document;
	  		return null;
	  	}
	
	  	if (document.lastIndexOf(RECORD_END) + RECORD_END.length() < document.length()) {
	      	newDoc = document.substring(0, document.lastIndexOf(RECORD_END) + RECORD_END.length());
			newPartialEvent = document.substring(document.lastIndexOf(RECORD_END) + RECORD_END.length());
	 	} else {
	  		newDoc = document;
	  	}
		if (partialEvent != null) {
			newDoc=partialEvent + newDoc;
		}	      		
	  	partialEvent=newPartialEvent;
	    Document doc = parse(newDoc);
	    if (doc == null) {
	      return null;
	    }
	    return decodeEvents(doc);
    }
    return null;
  }

  /**
   * Converts the string data into an XML Document, and then soaks out the
   * relevant bits to form a new LoggingEvent instance which can be used
   * by any Log4j element locally.
   * @param data
   * @return a single LoggingEvent
   */
  public LoggingEvent decode(String data) {
    Document document = parse(data);

    if (document == null) {
      return null;
    }

    Vector events = decodeEvents(document);

    if (events.size() > 0) {
      return (LoggingEvent) events.firstElement();
    }

    return null;
  }

  /**
   * Given a Document, converts the XML into a Vector of LoggingEvents
   * @param document
   * @return Vector of LoggingEvents
   */
  private Vector decodeEvents(Document document) {
    Vector events = new Vector();

    Logger logger = null;
    long timeStamp = 0L;
    String level = null;
    String threadName = null;
    Object message = null;
    String ndc = null;
    String[] exception = null;
    String className = null;
    String methodName = null;
    String fileName = null;
    String lineNumber = null;
    Hashtable properties = null;

    NodeList nl = document.getElementsByTagName("log4j:eventSet");
    Node eventSet = nl.item(0);

    NodeList eventList = eventSet.getChildNodes();

    for (int eventIndex = 0; eventIndex < eventList.getLength();
        eventIndex++) {
      Node eventNode = eventList.item(eventIndex);
      //ignore carriage returns in xml 
	  if(eventNode.getNodeType() != Node.ELEMENT_NODE) {
	  	continue;
	  }
      logger =
        Logger.getLogger(
          eventNode.getAttributes().getNamedItem("logger").getNodeValue());
      timeStamp =
        Long.parseLong(
          eventNode.getAttributes().getNamedItem("timestamp").getNodeValue());
      level =eventNode.getAttributes().getNamedItem("level").getNodeValue();
      threadName =
        eventNode.getAttributes().getNamedItem("thread").getNodeValue();

      NodeList list = eventNode.getChildNodes();
      int listLength = list.getLength();

      for (int y = 0; y < listLength; y++) {
        String tagName = list.item(y).getNodeName();

        if (tagName.equalsIgnoreCase("log4j:message")) {
          message = getCData(list.item(y));
        }

        if (tagName.equalsIgnoreCase("log4j:NDC")) {
          ndc = getCData(list.item(y));
        }
        //still support receiving of MDC and convert to properties
        if (tagName.equalsIgnoreCase("log4j:MDC")) {
          properties = new Hashtable();
          NodeList propertyList = list.item(y).getChildNodes();
          int propertyLength = propertyList.getLength();

          for (int i = 0; i < propertyLength; i++) {
            String propertyTag = propertyList.item(i).getNodeName();

            if (propertyTag.equalsIgnoreCase("log4j:data")) {
              Node property = propertyList.item(i);
              String name =
                property.getAttributes().getNamedItem("name").getNodeValue();
              String value =
                property.getAttributes().getNamedItem("value").getNodeValue();
              properties.put(name, value);
            }
          }
        }

        if (tagName.equalsIgnoreCase("log4j:throwable")) {
          exception = new String[] { getCData(list.item(y)) };
        }

        if (tagName.equalsIgnoreCase("log4j:locationinfo")) {
          className =
            list.item(y).getAttributes().getNamedItem("class").getNodeValue();
          methodName =
            list.item(y).getAttributes().getNamedItem("method").getNodeValue();
          fileName =
            list.item(y).getAttributes().getNamedItem("file").getNodeValue();
          lineNumber =
            list.item(y).getAttributes().getNamedItem("line").getNodeValue();
        }

        if (tagName.equalsIgnoreCase("log4j:properties")) {
          if (properties == null) {
              properties = new Hashtable();
          }
          NodeList propertyList = list.item(y).getChildNodes();
          int propertyLength = propertyList.getLength();

          for (int i = 0; i < propertyLength; i++) {
            String propertyTag = propertyList.item(i).getNodeName();

            if (propertyTag.equalsIgnoreCase("log4j:data")) {
              Node property = propertyList.item(i);
              String name =
                property.getAttributes().getNamedItem("name").getNodeValue();
              String value =
                property.getAttributes().getNamedItem("value").getNodeValue();
              properties.put(name, value);
            }
          }
        }

        /**
         * We add all the additional properties to the properties
         * hashtable.  Don't override properties that already exist
         */
        if (additionalProperties.size() > 0) {
          if (properties == null) {
            properties = new Hashtable(additionalProperties);
          } else {
            Iterator i = additionalProperties.entrySet().iterator();
            while (i.hasNext()) {
              Map.Entry e = (Map.Entry) i.next();
              if (!(properties.containsKey(e.getKey()))) {
              	properties.put(e.getKey(), e.getValue());
              }
            }
          }
        }
      }
      Level levelImpl = Level.toLevel(level);
      
      LocationInfo info = null;
      if ((fileName != null) || (className != null) || (methodName != null) || (lineNumber != null)) {
          info = new LocationInfo(fileName, className, methodName, lineNumber);
      } else {
        info = LocationInfo.NA_LOCATION_INFO;
      }
      if (exception == null) {
          exception = new String[]{""};
      }
      
      LoggingEvent loggingEvent = new LoggingEvent();
      loggingEvent.setLogger(logger);
      loggingEvent.setTimeStamp(timeStamp);
      loggingEvent.setLevel(levelImpl);
      loggingEvent.setThreadName(threadName);
      loggingEvent.setMessage(message);
      loggingEvent.setNDC(ndc);
      loggingEvent.setThrowableInformation(new ThrowableInformation(exception));
      loggingEvent.setLocationInformation(info);
      loggingEvent.setProperties(properties);
      
      events.add(loggingEvent);
     
      logger = null;
      timeStamp = 0L;
      level = null;
      threadName = null;
      message = null;
      ndc = null;
      exception = null;
      className = null;
      methodName = null;
      fileName = null;
      lineNumber = null;
      properties = null;
    }

    return events;
  }

  private String getCData(Node n) {
    StringBuffer buf = new StringBuffer();
    NodeList nl = n.getChildNodes();

    for (int x = 0; x < nl.getLength(); x++) {
      Node innerNode = nl.item(x);

      if (
        (innerNode.getNodeType() == Node.TEXT_NODE)
          || (innerNode.getNodeType() == Node.CDATA_SECTION_NODE)) {
        buf.append(innerNode.getNodeValue());
      }
    }

    return buf.toString();
  }
}