1   /*
2    *  XmlDocumentFormat.java
3    *
4    *  Copyright (c) 1998-2001, The University of Sheffield.
5    *
6    *  This file is part of GATE (see http://gate.ac.uk/), and is free
7    *  software, licenced under the GNU Library General Public License,
8    *  Version 2, June 1991 (in the distribution as file licence.html,
9    *  and also available at http://gate.ac.uk/gate/licence.html).
10   *
11   *  Cristian URSU, 26/May/2000
12   *
13   *  $Id: XmlDocumentFormat.java,v 1.41 2002/06/10 12:51:31 nasso Exp $
14   */
15  
16  package gate.corpora;
17  
18  //import com.sun.xml.parser.* ;
19  import java.util.*;
20  import java.io.*;
21  import java.net.*;
22  
23  import gate.util.*;
24  import gate.*;
25  import gate.xml.*;
26  import gate.event.*;
27  import gate.creole.*;
28  
29  // xml tools
30  import javax.xml.parsers.*;
31  import org.xml.sax.*;
32  import org.xml.sax.helpers.*;
33  import org.w3c.www.mime.*;
34  
35  /** The format of Documents. Subclasses of DocumentFormat know about
36    * particular MIME types and how to unpack the information in any
37    * markup or formatting they contain into GATE annotations. Each MIME
38    * type has its own subclass of DocumentFormat, e.g. XmlDocumentFormat,
39    * RtfDocumentFormat, MpegDocumentFormat. These classes register themselves
40    * with a static index residing here when they are constructed. Static
41    * getDocumentFormat methods can then be used to get the appropriate
42    * format class for a particular document.
43    */
44  public class XmlDocumentFormat extends TextualDocumentFormat
45  {
46    /** Debug flag */
47    private static final boolean DEBUG = false;
48  
49    /** Default construction */
50    public XmlDocumentFormat() { super(); }
51  
52    /** We could collect repositioning information during XML parsing */
53    public Boolean supportsRepositioning() {
54      return new Boolean(true);
55    } // supportsRepositioning
56  
57    /** Old style of unpackMarkup (without collecting of RepositioningInfo) */
58    public void unpackMarkup(Document doc) throws DocumentFormatException {
59      unpackMarkup(doc, (RepositioningInfo) null, (RepositioningInfo) null);
60    } // unpackMarkup
61  
62  
63    /** Unpack the markup in the document. This converts markup from the
64      * native format (e.g. XML) into annotations in GATE format.
65      * Uses the markupElementsMap to determine which elements to convert, and
66      * what annotation type names to use. If the document was created from a
67      * String, then is recomandable to set the doc's sourceUrl to <b>null</b>.
68      * So, if the document has a valid URL, then the parser will try to
69      * parse the XML document pointed by the URL.If the URL is not valid, or
70      * is null, then the doc's content will be parsed. If the doc's content is
71      * not a valid XML then the parser might crash.
72      *
73      * @param Document doc The gate document you want to parse. If
74      * <code>doc.getSourceUrl()</code> returns <b>null</b> then the content of
75      * doc will be parsed. Using a URL is recomended because the parser will
76      * report errors corectlly if the XML document is not well formed.
77      */
78    public void unpackMarkup(Document doc, RepositioningInfo repInfo,
79                RepositioningInfo ampCodingInfo) throws DocumentFormatException {
80      if( (doc == null) ||
81          (doc.getSourceUrl() == null && doc.getContent() == null)){
82  
83        throw new DocumentFormatException(
84                 "GATE document is null or no content found. Nothing to parse!");
85      }// End if
86  
87      boolean docHasContentButNoValidURL = false;
88      // This is a test to see if the GATE document has a valid URL or a valid
89      // content. If doesn't has a valid URL then try to parse its content as XML
90      try{
91        if (doc.getSourceUrl() == null && doc.getContent() != null){
92          // The doc's url is null but there is a content.
93          docHasContentButNoValidURL = true;
94        }else {URLConnection conn = doc.getSourceUrl().openConnection();}
95      }catch (IOException ex1){
96        // The URL is not null but is not valid.
97        if(doc.getContent() == null)
98          // The document content is also null. There is nothing we can do.
99          throw new DocumentFormatException("The document doesn't have a" +
100         " valid URL and also no content");
101       docHasContentButNoValidURL = true;
102     }// End try
103 
104     // Create a status listener
105     StatusListener statusListener = new StatusListener(){
106           public void statusChanged(String text){
107             // This is implemented in DocumentFormat.java and inherited here
108             fireStatusChanged(text);
109           }
110     };
111     GateFormatXmlDocumentHandler gateXmlHandler = null;
112     XmlDocumentHandler xmlDocHandler = null;
113     if (docHasContentButNoValidURL)
114       parseDocumentWithoutURL(doc, repInfo, ampCodingInfo);
115     else try {
116       // use Excerces XML parser with JAXP
117       // System.setProperty("javax.xml.parsers.SAXParserFactory",
118       //                         "org.apache.xerces.jaxp.SAXParserFactoryImpl");
119       // Get a parser factory.
120       SAXParserFactory saxParserFactory = SAXParserFactory.newInstance();
121       // Set up the factory to create the appropriate type of parser
122       // non validating one
123       saxParserFactory.setValidating(false);
124       // non namesapace aware one
125       saxParserFactory.setNamespaceAware(true);
126       // create it
127       SAXParser xmlParser = saxParserFactory.newSAXParser();
128       if (isGateXmlDocument){
129         // Construct the appropiate xml handler for the job.
130         gateXmlHandler = new GateFormatXmlDocumentHandler(doc);
131         // Register a status listener
132         gateXmlHandler.addStatusListener(statusListener);
133         // Parse the Gate Document
134         xmlParser.parse(doc.getSourceUrl().toString(), gateXmlHandler);
135         gateXmlHandler.removeStatusListener(statusListener);
136       }else{
137         // Create a new Xml document handler
138         xmlDocHandler =  new XmlDocumentHandler( doc,
139                                                  this.markupElementsMap,
140                                                  this.element2StringMap);
141         // Register a status listener with it
142         xmlDocHandler.addStatusListener(statusListener);
143         // set repositioning object
144         xmlDocHandler.setRepositioningInfo(repInfo);
145         // set the object with ampersand coding positions
146         xmlDocHandler.setAmpCodingInfo(ampCodingInfo);
147 
148         // Parse the document handler
149 /* Angel
150         xmlParser.parse(doc.getSourceUrl().toString(), xmlDocHandler );
151 Angel */
152       // try to choose concret parser (Xerces)
153 // Angel - start
154       org.apache.xerces.parsers.SAXParser newxmlParser =
155           new org.apache.xerces.parsers.SAXParser();
156       // Set up the factory to create the appropriate type of parser
157       // non validating one
158       // http://xml.org/sax/features/validation set to false
159       newxmlParser.setFeature("http://xml.org/sax/features/validation", false);
160       // namesapace aware one
161       // http://xml.org/sax/features/namespaces set to true
162       newxmlParser.setFeature("http://xml.org/sax/features/namespaces", true);
163       newxmlParser.setContentHandler(xmlDocHandler);
164       newxmlParser.setErrorHandler(xmlDocHandler);
165       newxmlParser.setDTDHandler(xmlDocHandler);
166       newxmlParser.setEntityResolver(xmlDocHandler);
167       newxmlParser.setReaderFactory(new StreamingCharFactory());
168       newxmlParser.parse(doc.getSourceUrl().toString());
169 // Angel - end
170         ((DocumentImpl) doc).setNextAnnotationId(
171                                           xmlDocHandler.getCustomObjectsId());
172         xmlDocHandler.removeStatusListener(statusListener);
173       }// End if
174     } catch (ParserConfigurationException e){
175         throw
176         new DocumentFormatException("XML parser configuration exception ", e);
177     } catch (SAXException e){
178       doc.getFeatures().put("parsingError", new Boolean(true));
179 
180       Boolean bThrow = (Boolean)
181         doc.getFeatures().get(GateConstants.THROWEX_FORMAT_PROPERTY_NAME);
182         
183       if(bThrow != null && bThrow.booleanValue()) {
184         // the next line is commented to avoid Document creation fail on error
185         throw new DocumentFormatException(e);
186       }
187       else {
188         Out.println("Warning: Document remains unparsed. \n"
189               +"\n  Stack Dump: ");
190         e.printStackTrace(Out.getPrintWriter());
191       } // if
192 
193     } catch (IOException e){
194         throw new DocumentFormatException("I/O exception for " +
195                                       doc.getSourceUrl().toString());
196     }finally{
197       if(gateXmlHandler != null)
198         gateXmlHandler.removeStatusListener(statusListener);
199       if (xmlDocHandler != null)
200         xmlDocHandler.removeStatusListener(statusListener);
201     }// End if else try
202   }// unpackMarkup
203 
204   /** Called from unpackMarkup() if the document have been created from a
205    *  string
206    */
207   private void parseDocumentWithoutURL(gate.Document aDocument,
208                                         RepositioningInfo repInfo,
209                                         RepositioningInfo ampCodingInfo)
210                                               throws DocumentFormatException {
211 
212     XmlDocumentHandler xmlDocHandler = null;
213     // Create a status listener
214     StatusListener statusList = new StatusListener(){
215         public void statusChanged(String text){
216           // this is implemented in DocumentFormat.java and inherited here
217           fireStatusChanged(text);
218         }
219     };
220     try{
221       Reader reader = new InputStreamReader(
222         new ByteArrayInputStream(aDocument.getContent().toString().getBytes()),
223         "UTF-8");
224       InputSource is = new InputSource(reader);
225 
226 
227       // use Excerces XML parser with JAXP
228       // System.setProperty("javax.xml.parsers.SAXParserFactory",
229       //                         "org.apache.xerces.jaxp.SAXParserFactoryImpl");
230       // Get a parser factory.
231       SAXParserFactory saxParserFactory = SAXParserFactory.newInstance();
232       // Set up the factory to create the appropriate type of parser
233       // non validating one
234       saxParserFactory.setValidating(false);
235       // non namesapace aware one
236       saxParserFactory.setNamespaceAware(true);
237       // create it
238       SAXParser xmlParser = saxParserFactory.newSAXParser();
239 
240       // create a new Xml document handler
241       xmlDocHandler =  new XmlDocumentHandler(aDocument,
242                                               this.markupElementsMap,
243                                               this.element2StringMap);
244       // Regsiter the statusListener with xmlDocHandler
245       xmlDocHandler.addStatusListener(statusList);
246       // set repositioning object
247       xmlDocHandler.setRepositioningInfo(repInfo);
248       // set the object with ampersand coding positions
249       xmlDocHandler.setAmpCodingInfo(ampCodingInfo);
250       // Parse the document handler
251 /* Angel
252 //      xmlParser.parse(is, xmlDocHandler);
253 Angel */
254 
255 // Angel - start
256       // try to choose concret parser
257       org.apache.xerces.parsers.SAXParser newxmlParser =
258           new org.apache.xerces.parsers.SAXParser();
259       // Set up the factory to create the appropriate type of parser
260       // non validating one
261       // http://xml.org/sax/features/validation set to false
262       newxmlParser.setFeature("http://xml.org/sax/features/validation", false);
263       // namesapace aware one
264       // http://xml.org/sax/features/namespaces set to true
265       newxmlParser.setFeature("http://xml.org/sax/features/namespaces", true);
266       newxmlParser.setContentHandler(xmlDocHandler);
267       newxmlParser.setErrorHandler(xmlDocHandler);
268       newxmlParser.setDTDHandler(xmlDocHandler);
269       newxmlParser.setEntityResolver(xmlDocHandler);
270       newxmlParser.setReaderFactory(new StreamingCharFactory());
271       newxmlParser.parse(is);
272 // Angel - end
273 
274       ((DocumentImpl) aDocument).setNextAnnotationId(
275                                           xmlDocHandler.getCustomObjectsId());
276     } catch (ParserConfigurationException e){
277         throw new DocumentFormatException(
278                         "XML parser configuration exception ", e);
279     } catch (SAXException e){
280         throw new DocumentFormatException(e);
281     } catch (IOException e){
282         throw new DocumentFormatException(e);
283     }finally{
284       // Remove the statusListener with xmlDocHandler
285       xmlDocHandler.removeStatusListener(statusList);
286     }// End try
287   }// End parseDocumentWithoutURL()
288 
289   /** Initialise this resource, and return it. */
290   public Resource init() throws ResourceInstantiationException{
291     // Register XML mime type
292     MimeType mime = new MimeType("text","xml");
293     // Register the class handler for this mime type
294     mimeString2ClassHandlerMap.put(mime.getType()+ "/" + mime.getSubtype(),
295                                                                           this);
296     // Register the mime type with mine string
297     mimeString2mimeTypeMap.put(mime.getType() + "/" + mime.getSubtype(), mime);
298     // Register file sufixes for this mime type
299     suffixes2mimeTypeMap.put("xml",mime);
300     suffixes2mimeTypeMap.put("xhtm",mime);
301     suffixes2mimeTypeMap.put("xhtml",mime);
302     // Register magic numbers for this mime type
303     magic2mimeTypeMap.put("<?xml",mime);
304     // Set the mimeType for this language resource
305     setMimeType(mime);
306     return this;
307   }// init()
308 
309 }//class XmlDocumentFormat
310