1 package org.apache.maven.doxia.parser;
2
3 /*
4 * Licensed to the Apache Software Foundation (ASF) under one
5 * or more contributor license agreements. See the NOTICE file
6 * distributed with this work for additional information
7 * regarding copyright ownership. The ASF licenses this file
8 * to you under the Apache License, Version 2.0 (the
9 * "License"); you may not use this file except in compliance
10 * with the License. You may obtain a copy of the License at
11 *
12 * http://www.apache.org/licenses/LICENSE-2.0
13 *
14 * Unless required by applicable law or agreed to in writing,
15 * software distributed under the License is distributed on an
16 * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
17 * KIND, either express or implied. See the License for the
18 * specific language governing permissions and limitations
19 * under the License.
20 */
21
22 import java.io.BufferedReader;
23 import java.io.ByteArrayInputStream;
24 import java.io.File;
25 import java.io.FileOutputStream;
26 import java.io.IOException;
27 import java.io.InputStream;
28 import java.io.OutputStream;
29 import java.io.Reader;
30 import java.io.StringReader;
31 import java.net.URL;
32 import java.util.Hashtable;
33 import java.util.LinkedHashMap;
34 import java.util.Locale;
35 import java.util.Map;
36 import java.util.regex.Matcher;
37 import java.util.regex.Pattern;
38
39 import org.apache.http.HttpEntity;
40 import org.apache.http.HttpResponse;
41 import org.apache.http.HttpStatus;
42 import org.apache.http.client.ClientProtocolException;
43 import org.apache.http.client.HttpRequestRetryHandler;
44 import org.apache.http.client.methods.HttpGet;
45 import org.apache.http.impl.client.DefaultHttpClient;
46 import org.apache.http.impl.client.DefaultHttpRequestRetryHandler;
47 import org.apache.http.util.EntityUtils;
48
49 import org.apache.maven.doxia.macro.MacroExecutionException;
50 import org.apache.maven.doxia.markup.XmlMarkup;
51 import org.apache.maven.doxia.sink.Sink;
52 import org.apache.maven.doxia.sink.impl.SinkEventAttributeSet;
53 import org.apache.maven.doxia.util.HtmlTools;
54 import org.apache.maven.doxia.util.XmlValidator;
55
56 import org.codehaus.plexus.util.FileUtils;
57 import org.codehaus.plexus.util.IOUtil;
58 import org.codehaus.plexus.util.StringUtils;
59 import org.codehaus.plexus.util.xml.pull.MXParser;
60 import org.codehaus.plexus.util.xml.pull.XmlPullParser;
61 import org.codehaus.plexus.util.xml.pull.XmlPullParserException;
62
63 import org.xml.sax.EntityResolver;
64 import org.xml.sax.InputSource;
65 import org.xml.sax.SAXException;
66
67 /**
68 * An abstract class that defines some convenience methods for <code>XML</code> parsers.
69 *
70 * @author <a href="mailto:vincent.siveton@gmail.com">Vincent Siveton</a>
71 * @version $Id$
72 * @since 1.0
73 */
74 public abstract class AbstractXmlParser
75 extends AbstractParser
76 implements XmlMarkup
77 {
78 /**
79 * Entity pattern for HTML entity, i.e. &nbsp;
80 * "<!ENTITY(\\s)+([^>|^\\s]+)(\\s)+\"(\\s)*(&[a-zA-Z]{2,6};)(\\s)*\"(\\s)*>
81 * <br/>
82 * see <a href="http://www.w3.org/TR/REC-xml/#NT-EntityDecl">http://www.w3.org/TR/REC-xml/#NT-EntityDecl</a>.
83 */
84 private static final Pattern PATTERN_ENTITY_1 =
85 Pattern.compile( ENTITY_START + "(\\s)+([^>|^\\s]+)(\\s)+\"(\\s)*(&[a-zA-Z]{2,6};)(\\s)*\"(\\s)*>" );
86
87 /**
88 * Entity pattern for Unicode entity, i.e. &#38;
89 * "<!ENTITY(\\s)+([^>|^\\s]+)(\\s)+\"(\\s)*(&(#x?[0-9a-fA-F]{1,5};)*)(\\s)*\"(\\s)*>"
90 * <br/>
91 * see <a href="http://www.w3.org/TR/REC-xml/#NT-EntityDecl">http://www.w3.org/TR/REC-xml/#NT-EntityDecl</a>.
92 */
93 private static final Pattern PATTERN_ENTITY_2 =
94 Pattern.compile( ENTITY_START + "(\\s)+([^>|^\\s]+)(\\s)+\"(\\s)*(&(#x?[0-9a-fA-F]{1,5};)*)(\\s)*\"(\\s)*>" );
95
96 private boolean ignorableWhitespace;
97
98 private boolean collapsibleWhitespace;
99
100 private boolean trimmableWhitespace;
101
102 private Map<String, String> entities;
103
104 private boolean validate = false;
105
106 /** {@inheritDoc} */
107 public void parse( Reader source, Sink sink )
108 throws ParseException
109 {
110 init();
111
112 Reader src = source;
113
114 // 1 first parsing if validation is required
115 if ( isValidate() )
116 {
117 String content;
118 try
119 {
120 content = IOUtil.toString( new BufferedReader( src ) );
121 }
122 catch ( IOException e )
123 {
124 throw new ParseException( "Error reading the model: " + e.getMessage(), e );
125 }
126
127 new XmlValidator( getLog() ).validate( content );
128
129 src = new StringReader( content );
130 }
131
132 // 2 second parsing to process
133 try
134 {
135 XmlPullParser parser = new MXParser();
136
137 parser.setInput( src );
138
139 // allow parser initialization, e.g. for additional entities in XHTML
140 // Note: do it after input is set, otherwise values are reset
141 initXmlParser( parser );
142
143 sink.enableLogging( getLog() );
144
145 parseXml( parser, sink );
146 }
147 catch ( XmlPullParserException ex )
148 {
149 throw new ParseException( "Error parsing the model: " + ex.getMessage(), ex, ex.getLineNumber(),
150 ex.getColumnNumber() );
151 }
152 catch ( MacroExecutionException ex )
153 {
154 throw new ParseException( "Macro execution failed: " + ex.getMessage(), ex );
155 }
156
157 setSecondParsing( false );
158 init();
159 }
160
161 /**
162 * Initializes the parser with custom entities or other options.
163 *
164 * @param parser A parser, not null.
165 * @throws org.codehaus.plexus.util.xml.pull.XmlPullParserException if there's a problem initializing the parser
166 */
167 protected void initXmlParser( XmlPullParser parser )
168 throws XmlPullParserException
169 {
170 // nop
171 }
172
173 /**
174 * {@inheritDoc}
175 *
176 * Convenience method to parse an arbitrary string and emit any xml events into the given sink.
177 */
178 @Override
179 public void parse( String string, Sink sink )
180 throws ParseException
181 {
182 super.parse( string, sink );
183 }
184
185 /** {@inheritDoc} */
186 @Override
187 public final int getType()
188 {
189 return XML_TYPE;
190 }
191
192 /**
193 * Converts the attributes of the current start tag of the given parser to a SinkEventAttributeSet.
194 *
195 * @param parser A parser, not null.
196 * @return a SinkEventAttributeSet or null if the current parser event is not a start tag.
197 * @since 1.1
198 */
199 protected SinkEventAttributeSet getAttributesFromParser( XmlPullParser parser )
200 {
201 int count = parser.getAttributeCount();
202
203 if ( count < 0 )
204 {
205 return null;
206 }
207
208 SinkEventAttributeSet atts = new SinkEventAttributeSet( count );
209
210 for ( int i = 0; i < count; i++ )
211 {
212 atts.addAttribute( parser.getAttributeName( i ), parser.getAttributeValue( i ) );
213 }
214
215 return atts;
216 }
217
218 /**
219 * Parse the model from the XmlPullParser into the given sink.
220 *
221 * @param parser A parser, not null.
222 * @param sink the sink to receive the events.
223 * @throws org.codehaus.plexus.util.xml.pull.XmlPullParserException if there's a problem parsing the model
224 * @throws org.apache.maven.doxia.macro.MacroExecutionException if there's a problem executing a macro
225 */
226 private void parseXml( XmlPullParser parser, Sink sink )
227 throws XmlPullParserException, MacroExecutionException
228 {
229 int eventType = parser.getEventType();
230
231 while ( eventType != XmlPullParser.END_DOCUMENT )
232 {
233 if ( eventType == XmlPullParser.START_TAG )
234 {
235 handleStartTag( parser, sink );
236 }
237 else if ( eventType == XmlPullParser.END_TAG )
238 {
239 handleEndTag( parser, sink );
240 }
241 else if ( eventType == XmlPullParser.TEXT )
242 {
243 String text = getText( parser );
244
245 if ( isIgnorableWhitespace() )
246 {
247 if ( text.trim().length() != 0 )
248 {
249 handleText( parser, sink );
250 }
251 }
252 else
253 {
254 handleText( parser, sink );
255 }
256 }
257 else if ( eventType == XmlPullParser.CDSECT )
258 {
259 handleCdsect( parser, sink );
260 }
261 else if ( eventType == XmlPullParser.COMMENT )
262 {
263 handleComment( parser, sink );
264 }
265 else if ( eventType == XmlPullParser.ENTITY_REF )
266 {
267 handleEntity( parser, sink );
268 }
269 else if ( eventType == XmlPullParser.IGNORABLE_WHITESPACE )
270 {
271 // nop
272 }
273 else if ( eventType == XmlPullParser.PROCESSING_INSTRUCTION )
274 {
275 // nop
276 }
277 else if ( eventType == XmlPullParser.DOCDECL )
278 {
279 addLocalEntities( parser, parser.getText() );
280
281 for ( byte[] res : CachedFileEntityResolver.ENTITY_CACHE.values() )
282 {
283 addDTDEntities( parser, new String( res ) );
284 }
285 }
286
287 try
288 {
289 eventType = parser.nextToken();
290 }
291 catch ( IOException io )
292 {
293 throw new XmlPullParserException( "IOException: " + io.getMessage(), parser, io );
294 }
295 }
296 }
297
298 /**
299 * Goes through the possible start tags.
300 *
301 * @param parser A parser, not null.
302 * @param sink the sink to receive the events.
303 * @throws org.codehaus.plexus.util.xml.pull.XmlPullParserException if there's a problem parsing the model
304 * @throws org.apache.maven.doxia.macro.MacroExecutionException if there's a problem executing a macro
305 */
306 protected abstract void handleStartTag( XmlPullParser parser, Sink sink )
307 throws XmlPullParserException, MacroExecutionException;
308
309 /**
310 * Goes through the possible end tags.
311 *
312 * @param parser A parser, not null.
313 * @param sink the sink to receive the events.
314 * @throws org.codehaus.plexus.util.xml.pull.XmlPullParserException if there's a problem parsing the model
315 * @throws org.apache.maven.doxia.macro.MacroExecutionException if there's a problem executing a macro
316 */
317 protected abstract void handleEndTag( XmlPullParser parser, Sink sink )
318 throws XmlPullParserException, MacroExecutionException;
319
320 /**
321 * Handles text events.
322 *
323 * <p>This is a default implementation, if the parser points to a non-empty text element,
324 * it is emitted as a text event into the specified sink.</p>
325 *
326 * @param parser A parser, not null.
327 * @param sink the sink to receive the events. Not null.
328 * @throws org.codehaus.plexus.util.xml.pull.XmlPullParserException if there's a problem parsing the model
329 */
330 protected void handleText( XmlPullParser parser, Sink sink )
331 throws XmlPullParserException
332 {
333 String text = getText( parser );
334
335 /*
336 * NOTE: Don't do any whitespace trimming here. Whitespace normalization has already been performed by the
337 * parser so any whitespace that makes it here is significant.
338 */
339 if ( StringUtils.isNotEmpty( text ) )
340 {
341 sink.text( text );
342 }
343 }
344
345 /**
346 * Handles CDATA sections.
347 *
348 * <p>This is a default implementation, all data are emitted as text
349 * events into the specified sink.</p>
350 *
351 * @param parser A parser, not null.
352 * @param sink the sink to receive the events. Not null.
353 * @throws org.codehaus.plexus.util.xml.pull.XmlPullParserException if there's a problem parsing the model
354 */
355 protected void handleCdsect( XmlPullParser parser, Sink sink )
356 throws XmlPullParserException
357 {
358 sink.text( getText( parser ) );
359 }
360
361 /**
362 * Handles comments.
363 *
364 * <p>This is a default implementation, all data are emitted as comment
365 * events into the specified sink.</p>
366 *
367 * @param parser A parser, not null.
368 * @param sink the sink to receive the events. Not null.
369 * @throws org.codehaus.plexus.util.xml.pull.XmlPullParserException if there's a problem parsing the model
370 */
371 protected void handleComment( XmlPullParser parser, Sink sink )
372 throws XmlPullParserException
373 {
374 if ( isEmitComments() )
375 {
376 sink.comment( getText( parser ) );
377 }
378 }
379
380 /**
381 * Handles entities.
382 *
383 * <p>This is a default implementation, all entities are resolved and emitted as text
384 * events into the specified sink, except:</p>
385 * <ul>
386 * <li>the entities with names <code>#160</code>, <code>nbsp</code> and <code>#x00A0</code>
387 * are emitted as <code>nonBreakingSpace()</code> events.</li>
388 * </ul>
389 *
390 * @param parser A parser, not null.
391 * @param sink the sink to receive the events. Not null.
392 * @throws org.codehaus.plexus.util.xml.pull.XmlPullParserException if there's a problem parsing the model
393 */
394 protected void handleEntity( XmlPullParser parser, Sink sink )
395 throws XmlPullParserException
396 {
397 String text = getText( parser );
398
399 String name = parser.getName();
400
401 if ( "#160".equals( name ) || "nbsp".equals( name ) || "#x00A0".equals( name ) )
402 {
403 sink.nonBreakingSpace();
404 }
405 else
406 {
407 String unescaped = HtmlTools.unescapeHTML( text );
408
409 sink.text( unescaped );
410 }
411 }
412
413 /**
414 * Handles an unknown event.
415 *
416 * <p>This is a default implementation, all events are emitted as unknown
417 * events into the specified sink.</p>
418 *
419 * @param parser the parser to get the event from.
420 * @param sink the sink to receive the event.
421 * @param type the tag event type. This should be one of HtmlMarkup.TAG_TYPE_SIMPLE,
422 * HtmlMarkup.TAG_TYPE_START, HtmlMarkup.TAG_TYPE_END or HtmlMarkup.ENTITY_TYPE.
423 * It will be passed as the first argument of the required parameters to the Sink
424 * {@link
425 * org.apache.maven.doxia.sink.Sink#unknown(String, Object[], org.apache.maven.doxia.sink.SinkEventAttributes)}
426 * method.
427 */
428 protected void handleUnknown( XmlPullParser parser, Sink sink, int type )
429 {
430 Object[] required = new Object[] { Integer.valueOf( type ) };
431
432 SinkEventAttributeSet attribs = getAttributesFromParser( parser );
433
434 sink.unknown( parser.getName(), required, attribs );
435 }
436
437 /**
438 * <p>isIgnorableWhitespace.</p>
439 *
440 * @return <code>true</code> if whitespace will be ignored, <code>false</code> otherwise.
441 * @see #setIgnorableWhitespace(boolean)
442 * @since 1.1
443 */
444 protected boolean isIgnorableWhitespace()
445 {
446 return ignorableWhitespace;
447 }
448
449 /**
450 * Specify that whitespace will be ignored. I.e.:
451 * <pre><tr> <td/> </tr></pre>
452 * is equivalent to
453 * <pre><tr><td/></tr></pre>
454 *
455 * @param ignorable <code>true</code> to ignore whitespace, <code>false</code> otherwise.
456 * @since 1.1
457 */
458 protected void setIgnorableWhitespace( boolean ignorable )
459 {
460 this.ignorableWhitespace = ignorable;
461 }
462
463 /**
464 * <p>isCollapsibleWhitespace.</p>
465 *
466 * @return <code>true</code> if text will collapse, <code>false</code> otherwise.
467 * @see #setCollapsibleWhitespace(boolean)
468 * @since 1.1
469 */
470 protected boolean isCollapsibleWhitespace()
471 {
472 return collapsibleWhitespace;
473 }
474
475 /**
476 * Specify that text will be collapsed. I.e.:
477 * <pre>Text Text</pre>
478 * is equivalent to
479 * <pre>Text Text</pre>
480 *
481 * @param collapsible <code>true</code> to allow collapsible text, <code>false</code> otherwise.
482 * @since 1.1
483 */
484 protected void setCollapsibleWhitespace( boolean collapsible )
485 {
486 this.collapsibleWhitespace = collapsible;
487 }
488
489 /**
490 * <p>isTrimmableWhitespace.</p>
491 *
492 * @return <code>true</code> if text will be trim, <code>false</code> otherwise.
493 * @see #setTrimmableWhitespace(boolean)
494 * @since 1.1
495 */
496 protected boolean isTrimmableWhitespace()
497 {
498 return trimmableWhitespace;
499 }
500
501 /**
502 * Specify that text will be collapsed. I.e.:
503 * <pre><p> Text </p></pre>
504 * is equivalent to
505 * <pre><p>Text</p></pre>
506 *
507 * @param trimmable <code>true</code> to allow trimmable text, <code>false</code> otherwise.
508 * @since 1.1
509 */
510 protected void setTrimmableWhitespace( boolean trimmable )
511 {
512 this.trimmableWhitespace = trimmable;
513 }
514
515 /**
516 * <p>getText.</p>
517 *
518 * @param parser A parser, not null.
519 * @return the {@link XmlPullParser#getText()} taking care of trimmable or collapsible configuration.
520 * @see XmlPullParser#getText()
521 * @see #isCollapsibleWhitespace()
522 * @see #isTrimmableWhitespace()
523 * @since 1.1
524 */
525 protected String getText( XmlPullParser parser )
526 {
527 String text = parser.getText();
528
529 if ( isTrimmableWhitespace() )
530 {
531 text = text.trim();
532 }
533
534 if ( isCollapsibleWhitespace() )
535 {
536 StringBuilder newText = new StringBuilder();
537 String[] elts = StringUtils.split( text, " \r\n" );
538 for ( int i = 0; i < elts.length; i++ )
539 {
540 newText.append( elts[i] );
541 if ( ( i + 1 ) < elts.length )
542 {
543 newText.append( " " );
544 }
545 }
546 text = newText.toString();
547 }
548
549 return text;
550 }
551
552 /**
553 * Return the defined entities in a local doctype. I.e.:
554 * <pre>
555 * <!DOCTYPE foo [
556 * <!ENTITY bar "&#x160;">
557 * <!ENTITY bar1 "&#x161;">
558 * ]>
559 * </pre>
560 *
561 * @return a map of the defined entities in a local doctype.
562 * @since 1.1
563 */
564 protected Map<String, String> getLocalEntities()
565 {
566 if ( entities == null )
567 {
568 entities = new LinkedHashMap<String, String>();
569 }
570
571 return entities;
572 }
573
574 /**
575 * <p>isValidate.</p>
576 *
577 * @return <code>true</code> if XML content will be validate, <code>false</code> otherwise.
578 * @since 1.1
579 */
580 public boolean isValidate()
581 {
582 return validate;
583 }
584
585 /**
586 * Specify a flag to validate or not the XML content.
587 *
588 * @param validate the validate to set
589 * @see #parse(Reader, Sink)
590 * @since 1.1
591 */
592 public void setValidate( boolean validate )
593 {
594 this.validate = validate;
595 }
596
597 // ----------------------------------------------------------------------
598 // Private methods
599 // ----------------------------------------------------------------------
600
601 /**
602 * Add an entity given by <code>entityName</code> and <code>entityValue</code> to {@link #entities}.
603 * <br/>
604 * By default, we exclude the default XML entities: &amp;, &lt;, &gt;, &quot; and &apos;.
605 *
606 * @param parser not null
607 * @param entityName not null
608 * @param entityValue not null
609 * @throws XmlPullParserException if any
610 * @see {@link XmlPullParser#defineEntityReplacementText(String, String)}
611 */
612 private void addEntity( XmlPullParser parser, String entityName, String entityValue )
613 throws XmlPullParserException
614 {
615 if ( entityName.endsWith( "amp" ) || entityName.endsWith( "lt" ) || entityName.endsWith( "gt" )
616 || entityName.endsWith( "quot" ) || entityName.endsWith( "apos" ) )
617 {
618 return;
619 }
620
621 parser.defineEntityReplacementText( entityName, entityValue );
622 getLocalEntities().put( entityName, entityValue );
623 }
624
625 /**
626 * Handle entities defined in a local doctype as the following:
627 * <pre>
628 * <!DOCTYPE foo [
629 * <!ENTITY bar "&#x160;">
630 * <!ENTITY bar1 "&#x161;">
631 * ]>
632 * </pre>
633 *
634 * @param parser not null
635 * @param text not null
636 * @throws XmlPullParserException if any
637 */
638 private void addLocalEntities( XmlPullParser parser, String text )
639 throws XmlPullParserException
640 {
641 int entitiesCount = StringUtils.countMatches( text, ENTITY_START );
642 if ( entitiesCount > 0 )
643 {
644 // text should be foo [...]
645 int start = text.indexOf( '[' );
646 int end = text.lastIndexOf( ']' );
647 if ( start != -1 && end != -1 )
648 {
649 addDTDEntities( parser, text.substring( start + 1, end ) );
650 }
651 }
652 }
653
654 /**
655 * Handle entities defined in external doctypes as the following:
656 * <pre>
657 * <!DOCTYPE foo [
658 * <!-- These are the entity sets for ISO Latin 1 characters for the XHTML -->
659 * <!ENTITY % HTMLlat1 PUBLIC "-//W3C//ENTITIES Latin 1 for XHTML//EN"
660 * "http://www.w3.org/TR/xhtml1/DTD/xhtml-lat1.ent">
661 * %HTMLlat1;
662 * ]>
663 * </pre>
664 *
665 * @param parser not null
666 * @param text not null
667 * @throws XmlPullParserException if any
668 */
669 private void addDTDEntities( XmlPullParser parser, String text )
670 throws XmlPullParserException
671 {
672 int entitiesCount = StringUtils.countMatches( text, ENTITY_START );
673 if ( entitiesCount > 0 )
674 {
675 final String txt = StringUtils.replace( text, ENTITY_START, "\n" + ENTITY_START );
676 BufferedReader reader = new BufferedReader( new StringReader( txt ) );
677 String line;
678 String tmpLine = "";
679 try
680 {
681 Matcher matcher;
682 while ( ( line = reader.readLine() ) != null )
683 {
684 tmpLine += "\n" + line;
685 matcher = PATTERN_ENTITY_1.matcher( tmpLine );
686 if ( matcher.find() && matcher.groupCount() == 7 )
687 {
688 String entityName = matcher.group( 2 );
689 String entityValue = matcher.group( 5 );
690
691 addEntity( parser, entityName, entityValue );
692 tmpLine = "";
693 }
694 else
695 {
696 matcher = PATTERN_ENTITY_2.matcher( tmpLine );
697 if ( matcher.find() && matcher.groupCount() == 8 )
698 {
699 String entityName = matcher.group( 2 );
700 String entityValue = matcher.group( 5 );
701
702 addEntity( parser, entityName, entityValue );
703 tmpLine = "";
704 }
705 }
706 }
707 }
708 catch ( IOException e )
709 {
710 // nop
711 }
712 finally
713 {
714 IOUtil.close( reader );
715 }
716 }
717 }
718
719 /**
720 * Implementation of the callback mechanism <code>EntityResolver</code>.
721 * Using a mechanism of cached files in temp dir to improve performance when using the <code>XMLReader</code>.
722 */
723 public static class CachedFileEntityResolver
724 implements EntityResolver
725 {
726 /** Map with systemId as key and the content of systemId as byte[]. */
727 protected static final Map<String, byte[]> ENTITY_CACHE = new Hashtable<String, byte[]>();
728
729 /** {@inheritDoc} */
730 public InputSource resolveEntity( String publicId, String systemId )
731 throws SAXException, IOException
732 {
733 byte[] res = ENTITY_CACHE.get( systemId );
734 // already cached?
735 if ( res == null )
736 {
737 String systemName = FileUtils.getFile( systemId ).getName();
738 File temp = new File( System.getProperty( "java.io.tmpdir" ), systemName );
739 // maybe already as a temp file?
740 if ( !temp.exists() )
741 {
742 // is systemId a file or an url?
743 if ( systemId.toLowerCase( Locale.ENGLISH ).startsWith( "file" ) )
744 {
745 // Doxia XSDs are included in the jars, so try to find the resource systemName from
746 // the classpath...
747 String resource = "/" + systemName;
748 URL url = getClass().getResource( resource );
749 if ( url != null )
750 {
751 res = toByteArray( url );
752 }
753 else
754 {
755 throw new SAXException( "Could not find the SYSTEM entity: " + systemId
756 + " because '" + resource + "' is not available of the classpath." );
757 }
758 }
759 else
760 {
761 res = toByteArray( new URL( systemId ) );
762 }
763
764 // write systemId as temp file
765 copy( res, temp );
766 }
767 else
768 {
769 // TODO How to refresh Doxia XSDs from temp dir?
770 res = toByteArray( temp.toURI().toURL() );
771 }
772
773 ENTITY_CACHE.put( systemId, res );
774 }
775
776 InputSource is = new InputSource( new ByteArrayInputStream( res ) );
777 is.setPublicId( publicId );
778 is.setSystemId( systemId );
779
780 return is;
781 }
782
783 /**
784 * If url is not an http/https urls, call {@link IOUtil#toByteArray(java.io.InputStream)} to get the url
785 * content.
786 * Otherwise, use HttpClient to get the http content.
787 * Wrap all internal exceptions to throw SAXException.
788 *
789 * @param url not null
790 * @return return an array of byte
791 * @throws SAXException if any
792 */
793 private static byte[] toByteArray( URL url )
794 throws SAXException
795 {
796 if ( !( url.getProtocol().equalsIgnoreCase( "http" ) || url.getProtocol().equalsIgnoreCase( "https" ) ) )
797 {
798 InputStream is = null;
799 try
800 {
801 is = url.openStream();
802 if ( is == null )
803 {
804 throw new SAXException( "Cannot open stream from the url: " + url.toString() );
805 }
806 return IOUtil.toByteArray( is );
807 }
808 catch ( IOException e )
809 {
810 throw new SAXException( "IOException: " + e.getMessage(), e );
811 }
812 finally
813 {
814 IOUtil.close( is );
815 }
816 }
817
818 // it is an HTTP url, using HttpClient...
819 DefaultHttpClient client = new DefaultHttpClient();
820 HttpGet method = new HttpGet( url.toString() );
821 // Set a user-agent that doesn't contain the word "java", otherwise it will be blocked by the W3C
822 // The default user-agent is "Apache-HttpClient/4.0.2 (java 1.5)"
823 method.setHeader( "user-agent", "Apache-Doxia/" + doxiaVersion() );
824
825 HttpRequestRetryHandler retryHandler = new DefaultHttpRequestRetryHandler( 3, false );
826 client.setHttpRequestRetryHandler( retryHandler );
827
828 HttpEntity entity = null;
829 try
830 {
831 HttpResponse response = client.execute( method );
832 int statusCode = response.getStatusLine().getStatusCode();
833 if ( statusCode != HttpStatus.SC_OK )
834 {
835 throw new IOException( "The status code when accessing the URL '" + url.toString() + "' was "
836 + statusCode + ", which is not allowed. The server gave this reason for the failure '"
837 + response.getStatusLine().getReasonPhrase() + "'." );
838 }
839
840 entity = response.getEntity();
841 return EntityUtils.toByteArray( entity );
842 }
843 catch ( ClientProtocolException e )
844 {
845 throw new SAXException( "ClientProtocolException: Fatal protocol violation: " + e.getMessage(), e );
846 }
847 catch ( IOException e )
848 {
849 throw new SAXException( "IOException: Fatal transport error: " + e.getMessage(), e );
850 }
851 finally
852 {
853 if ( entity != null )
854 {
855 try
856 {
857 entity.consumeContent();
858 }
859 catch ( IOException e )
860 {
861 // Ignore
862 }
863 }
864 }
865 }
866
867 /**
868 * Wrap {@link IOUtil#copy(byte[], OutputStream)} to throw SAXException.
869 *
870 * @param res not null array of byte
871 * @param f the file where to write the bytes
872 * @throws SAXException if any
873 * @see {@link IOUtil#copy(byte[], OutputStream)}
874 */
875 private void copy( byte[] res, File f )
876 throws SAXException
877 {
878 if ( f.isDirectory() )
879 {
880 throw new SAXException( "'" + f.getAbsolutePath() + "' is a directory, can not write it." );
881 }
882
883 OutputStream os = null;
884 try
885 {
886 os = new FileOutputStream( f );
887 IOUtil.copy( res, os );
888 }
889 catch ( IOException e )
890 {
891 throw new SAXException( "IOException: " + e.getMessage(), e );
892 }
893 finally
894 {
895 IOUtil.close( os );
896 }
897 }
898 }
899 }