TextSerializer


1   /*
2    * The Apache Software License, Version 1.1
3    *
4    *
5    * Copyright (c) 1999-2002 The Apache Software Foundation.  All rights
6    * reserved.
7    *
8    * Redistribution and use in source and binary forms, with or without
9    * modification, are permitted provided that the following conditions
10   * are met:
11   *
12   * 1. Redistributions of source code must retain the above copyright
13   *    notice, this list of conditions and the following disclaimer.
14   *
15   * 2. Redistributions in binary form must reproduce the above copyright
16   *    notice, this list of conditions and the following disclaimer in
17   *    the documentation and/or other materials provided with the
18   *    distribution.
19   *
20   * 3. The end-user documentation included with the redistribution,
21   *    if any, must include the following acknowledgment:
22   *       "This product includes software developed by the
23   *        Apache Software Foundation (http://www.apache.org/)."
24   *    Alternately, this acknowledgment may appear in the software itself,
25   *    if and wherever such third-party acknowledgments normally appear.
26   *
27   * 4. The names "Xerces" and "Apache Software Foundation" must
28   *    not be used to endorse or promote products derived from this
29   *    software without prior written permission. For written
30   *    permission, please contact apache@apache.org.
31   *
32   * 5. Products derived from this software may not be called "Apache",
33   *    nor may "Apache" appear in their name, without prior written
34   *    permission of the Apache Software Foundation.
35   *
36   * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
37   * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
38   * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
39   * DISCLAIMED.  IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
40   * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
41   * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
42   * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
43   * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
44   * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
45   * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
46   * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
47   * SUCH DAMAGE.
48   * ====================================================================
49   *
50   * This software consists of voluntary contributions made by many
51   * individuals on behalf of the Apache Software Foundation and was
52   * originally based on software copyright (c) 1999, International
53   * Business Machines, Inc., http://www.apache.org.  For more
54   * information on the Apache Software Foundation, please see
55   * <http://www.apache.org/>.
56   */
57  
58  
59  // Sep 14, 2000:
60  //  Fixed serializer to report IO exception directly, instead at
61  //  the end of document processing.
62  //  Reported by Patrick Higgins <phiggins@transzap.com>
63  
64  
65  package com.sun.org.apache.xml.internal.serialize;
66  
67  
68  import java.io.IOException  ;
69  
70  import org.w3c.dom.Element  ;
71  import org.w3c.dom.Node  ;
72  import org.xml.sax.AttributeList  ;
73  import org.xml.sax.Attributes  ;
74  import org.xml.sax.SAXException  ;
75  
76  
77  /**
78   * Implements a text serializer supporting both DOM and SAX
79   * serializing. For usage instructions see {@link Serializer}.
80   * <p>
81   * If an output stream is used, the encoding is taken from the
82   * output format (defaults to <tt>UTF-8</tt>). If a writer is
83   * used, make sure the writer uses the same encoding (if applies)
84   * as specified in the output format.
85   * <p>
86   * The serializer supports both DOM and SAX. DOM serializing is done
87   * by calling {@link #serialize} and SAX serializing is done by firing
88   * SAX events and using the serializer as a document handler.
89   * <p>
90   * If an I/O exception occurs while serializing, the serializer
91   * will not throw an exception directly, but only throw it
92   * at the end of serializing (either DOM or SAX's {@link
93   * org.xml.sax.DocumentHandler#endDocument}.
94   *
95   *
96   * @version $Revision: 1.13 $ $Date: 2003/05/13 13:23:49 $
97   * @author <a HREF="mailto:arkin@intalio.com">Assaf Arkin</a>
98   * @see Serializer
99   */
100 public class TextSerializer
101     extends BaseMarkupSerializer
102 {
103 
104 
105     /**
106      * Constructs a new serializer. The serializer cannot be used without
107      * calling {@link #setOutputCharStream} or {@link #setOutputByteStream}
108      * first.
109      */
110     public TextSerializer()
111     {
112         super( new OutputFormat( Method.TEXT, null, false ) );
113     }
114 
115 
116     public void setOutputFormat( OutputFormat format )
117     {
118         super.setOutputFormat( format != null ? format : new OutputFormat( Method.TEXT, null, false ) );
119     }
120 
121 
122     //-----------------------------------------//
123     // SAX content handler serializing methods //
124     //-----------------------------------------//
125 
126 
127     public void startElement( String   namespaceURI, String   localName,
128                               String   rawName, Attributes   attrs )
129         throws SAXException  
130     {
131         startElement( rawName == null ? localName : rawName, null );
132     }
133 
134 
135     public void endElement( String   namespaceURI, String   localName,
136                             String   rawName )
137         throws SAXException  
138     {
139         endElement( rawName == null ? localName : rawName );
140     }
141 
142 
143     //------------------------------------------//
144     // SAX document handler serializing methods //
145     //------------------------------000---------//
146 
147 
148     public void startElement( String   tagName, AttributeList   attrs )
149         throws SAXException  
150     {
151         boolean      preserveSpace;
152         ElementState state;
153 
154         try {
155             state = getElementState();
156             if ( isDocumentState() ) {
157                 // If this is the root element handle it differently.
158                 // If the first root element in the document, serialize
159                 // the document's DOCTYPE. Space preserving defaults
160                 // to that of the output format.
161                 if ( ! _started )
162                     startDocument( tagName );
163             }
164             // For any other element, if first in parent, then
165             // use the parnet's space preserving.
166             preserveSpace = state.preserveSpace;
167 
168             // Do not change the current element state yet.
169             // This only happens in endElement().
170 
171             // Ignore all other attributes of the element, only printing
172             // its contents.
173 
174             // Now it's time to enter a new element state
175             // with the tag name and space preserving.
176             // We still do not change the curent element state.
177             state = enterElementState( null, null, tagName, preserveSpace );
178         } catch ( IOException   except ) {
179             throw new SAXException  ( except );
180         }
181     }
182 
183 
184     public void endElement( String   tagName )
185         throws SAXException  
186     {
187         try {
188             endElementIO( tagName );
189         } catch ( IOException   except ) {
190             throw new SAXException  ( except );
191         }
192     }
193 
194 
195     public void endElementIO( String   tagName )
196         throws IOException  
197     {
198         ElementState state;
199 
200         // Works much like content() with additions for closing
201         // an element. Note the different checks for the closed
202         // element's state and the parent element's state.
203         state = getElementState();
204         // Leave the element state and update that of the parent
205         // (if we're not root) to not empty and after element.
206         state = leaveElementState();
207         state.afterElement = true;
208         state.empty = false;
209         if ( isDocumentState() )
210             _printer.flush();
211     }
212 
213 
214     public void processingInstructionIO( String   target, String   code ) throws IOException  
215     {
216     }
217 
218 
219     public void comment( String   text )
220     {
221     }
222 
223 
224     public void comment( char[] chars, int start, int length )
225     {
226     }
227 
228 
229     public void characters( char[] chars, int start, int length )
230         throws SAXException  
231     {
232         ElementState state;
233 
234         try {
235             state = content();
236             state.doCData = state.inCData = false;
237             printText( chars, start, length, true, true );
238         } catch ( IOException   except ) {
239             throw new SAXException  ( except );
240         }
241     }
242 
243 
244     protected void characters( String   text, boolean unescaped )
245         throws IOException  
246     {
247         ElementState state;
248 
249         state = content();
250         state.doCData = state.inCData = false;
251         printText( text, true, true );
252     }
253 
254 
255     //------------------------------------------//
256     // Generic node serializing methods methods //
257     //------------------------------------------//
258 
259 
260     /**
261      * Called to serialize the document's DOCTYPE by the root element.
262      * <p>
263      * This method will check if it has not been called before ({@link #_started}),
264      * will serialize the document type declaration, and will serialize all
265      * pre-root comments and PIs that were accumulated in the document
266      * (see {@link #serializePreRoot}). Pre-root will be serialized even if
267      * this is not the first root element of the document.
268      */
269     protected void startDocument( String   rootTagName )
270         throws IOException  
271     {
272         // Required to stop processing the DTD, even though the DTD
273         // is not printed.
274         _printer.leaveDTD();
275 
276         _started = true;
277         // Always serialize these, even if not te first root element.
278         serializePreRoot();
279     }
280 
281 
282     /**
283      * Called to serialize a DOM element. Equivalent to calling {@link
284      * #startElement}, {@link #endElement} and serializing everything
285      * inbetween, but better optimized.
286      */
287     protected void serializeElement( Element   elem )
288         throws IOException  
289     {
290         Node           child;
291         ElementState state;
292         boolean      preserveSpace;
293         String         tagName;
294 
295         tagName = elem.getTagName();
296         state = getElementState();
297         if ( isDocumentState() ) {
298             // If this is the root element handle it differently.
299             // If the first root element in the document, serialize
300             // the document's DOCTYPE. Space preserving defaults
301             // to that of the output format.
302             if ( ! _started )
303                 startDocument( tagName );
304         }
305         // For any other element, if first in parent, then
306         // use the parnet's space preserving.
307         preserveSpace = state.preserveSpace;
308 
309         // Do not change the current element state yet.
310         // This only happens in endElement().
311 
312         // Ignore all other attributes of the element, only printing
313         // its contents.
314 
315         // If element has children, then serialize them, otherwise
316         // serialize en empty tag.
317         if ( elem.hasChildNodes() ) {
318             // Enter an element state, and serialize the children
319             // one by one. Finally, end the element.
320             state = enterElementState( null, null, tagName, preserveSpace );
321             child = elem.getFirstChild();
322             while ( child != null ) {
323                 serializeNode( child );
324                 child = child.getNextSibling();
325             }
326             endElementIO( tagName );
327         } else {
328             if ( ! isDocumentState() ) {
329                 // After element but parent element is no longer empty.
330                 state.afterElement = true;
331                 state.empty = false;
332             }
333         }
334     }
335 
336 
337     /**
338      * Serialize the DOM node. This method is unique to the Text serializer.
339      *
340      * @param node The node to serialize
341      */
342     protected void serializeNode( Node   node )
343         throws IOException  
344     {
345         // Based on the node type call the suitable SAX handler.
346         // Only comments entities and documents which are not
347         // handled by SAX are serialized directly.
348         switch ( node.getNodeType() ) {
349         case Node.TEXT_NODE : {
350             String   text;
351 
352             text = node.getNodeValue();
353             if ( text != null )
354                 characters( node.getNodeValue(), true );
355             break;
356         }
357 
358         case Node.CDATA_SECTION_NODE : {
359             String   text;
360 
361             text = node.getNodeValue();
362             if ( text != null )
363                 characters( node.getNodeValue(), true );
364             break;
365         }
366 
367         case Node.COMMENT_NODE :
368             break;
369 
370         case Node.ENTITY_REFERENCE_NODE :
371             // Ignore.
372             break;
373 
374         case Node.PROCESSING_INSTRUCTION_NODE :
375             break;
376 
377         case Node.ELEMENT_NODE :
378             serializeElement( (Element  ) node );
379             break;
380 
381         case Node.DOCUMENT_NODE :
382             // !!! Fall through
383         case Node.DOCUMENT_FRAGMENT_NODE : {
384             Node           child;
385 
386             // By definition this will happen if the node is a document,
387             // document fragment, etc. Just serialize its contents. It will
388             // work well for other nodes that we do not know how to serialize.
389             child = node.getFirstChild();
390             while ( child != null ) {
391                 serializeNode( child );
392                 child = child.getNextSibling();
393             }
394             break;
395         }
396 
397         default:
398             break;
399         }
400     }
401 
402 
403     protected ElementState content(boolean ignorable)
404     {
405         ElementState state;
406 
407         state = getElementState();
408         if ( ! isDocumentState() ) {
409             // If this is the first content in the element,
410             // change the state to not-empty.
411             if ( state.empty )
412                 state.empty = false;
413             // Except for one content type, all of them
414             // are not last element. That one content
415             // type will take care of itself.
416             state.afterElement = false;
417         }
418         return state;
419     }
420 
421 
422     protected String   getEntityRef( int ch )
423     {
424         return null;
425     }
426 
427 
428 }
429 
430 
431
A to Z: JavaDoc & Examples Daily Java News & Articles Open Source Projects Open Source Codes Free Computer Books Remove Frame
Popular Tags