1 /*
2 * Copyright (c) 2009, 2017, Oracle and/or its affiliates. All rights reserved.
3 * @LastModified: Oct 2017
4 */
5 /*
6 * Licensed to the Apache Software Foundation (ASF) under one or more
7 * contributor license agreements. See the NOTICE file distributed with
8 * this work for additional information regarding copyright ownership.
9 * The ASF licenses this file to You under the Apache License, Version 2.0
10 * (the "License"); you may not use this file except in compliance with
11 * the License. You may obtain a copy of the License at
12 *
13 * http://www.apache.org/licenses/LICENSE-2.0
14 *
15 * Unless required by applicable law or agreed to in writing, software
16 * distributed under the License is distributed on an "AS IS" BASIS,
17 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
18 * See the License for the specific language governing permissions and
19 * limitations under the License.
20 */
21
22 package com.sun.org.apache.xerces.internal.impl ;
23
24 import com.sun.org.apache.xerces.internal.impl.io.ASCIIReader;
25 import com.sun.org.apache.xerces.internal.impl.io.UCSReader;
26 import com.sun.org.apache.xerces.internal.impl.io.UTF8Reader;
27 import com.sun.org.apache.xerces.internal.impl.msg.XMLMessageFormatter;
28 import com.sun.org.apache.xerces.internal.impl.validation.ValidationManager;
29 import com.sun.org.apache.xerces.internal.util.*;
30 import com.sun.org.apache.xerces.internal.util.URI;
31 import com.sun.org.apache.xerces.internal.utils.XMLLimitAnalyzer;
32 import com.sun.org.apache.xerces.internal.utils.XMLSecurityManager;
33 import com.sun.org.apache.xerces.internal.utils.XMLSecurityPropertyManager;
34 import com.sun.org.apache.xerces.internal.xni.Augmentations;
35 import com.sun.org.apache.xerces.internal.xni.XMLResourceIdentifier;
36 import com.sun.org.apache.xerces.internal.xni.XNIException;
37 import com.sun.org.apache.xerces.internal.xni.parser.*;
38 import com.sun.xml.internal.stream.Entity;
39 import com.sun.xml.internal.stream.StaxEntityResolverWrapper;
40 import com.sun.xml.internal.stream.StaxXMLInputSource;
41 import com.sun.xml.internal.stream.XMLEntityStorage;
42 import java.io.*;
43 import java.net.HttpURLConnection;
44 import java.net.URISyntaxException;
45 import java.net.URL;
46 import java.net.URLConnection;
47 import java.util.HashMap;
48 import java.util.Iterator;
49 import java.util.Locale;
50 import java.util.Map;
51 import java.util.Stack;
52 import java.util.StringTokenizer;
53 import javax.xml.XMLConstants;
54 import javax.xml.catalog.CatalogException;
55 import javax.xml.catalog.CatalogFeatures.Feature;
56 import javax.xml.catalog.CatalogFeatures;
57 import javax.xml.catalog.CatalogManager;
58 import javax.xml.catalog.CatalogResolver;
59 import javax.xml.stream.XMLInputFactory;
60 import javax.xml.transform.Source;
61 import jdk.xml.internal.JdkXmlUtils;
62 import jdk.xml.internal.SecuritySupport;
63 import org.xml.sax.InputSource;
64
65
66 /**
67 * Will keep track of current entity.
68 *
69 * The entity manager handles the registration of general and parameter
70 * entities; resolves entities; and starts entities. The entity manager
71 * is a central component in a standard parser configuration and this
72 * class works directly with the entity scanner to manage the underlying
73 * xni.
74 * <p>
75 * This component requires the following features and properties from the
76 * component manager that uses it:
77 * <ul>
78 * <li>http://xml.org/sax/features/validation</li>
79 * <li>http://xml.org/sax/features/external-general-entities</li>
80 * <li>http://xml.org/sax/features/external-parameter-entities</li>
81 * <li>http://apache.org/xml/features/allow-java-encodings</li>
82 * <li>http://apache.org/xml/properties/internal/symbol-table</li>
83 * <li>http://apache.org/xml/properties/internal/error-reporter</li>
84 * <li>http://apache.org/xml/properties/internal/entity-resolver</li>
85 * </ul>
86 *
87 *
88 * @author Andy Clark, IBM
89 * @author Arnaud Le Hors, IBM
90 * @author K.Venugopal SUN Microsystems
91 * @author Neeraj Bajaj SUN Microsystems
92 * @author Sunitha Reddy SUN Microsystems
93 */
94 public class XMLEntityManager implements XMLComponent, XMLEntityResolver {
95
96 //
97 // Constants
98 //
99
100 /** Default buffer size (2048). */
101 public static final int DEFAULT_BUFFER_SIZE = 8192;
102
103 /** Default buffer size before we've finished with the XMLDecl: */
104 public static final int DEFAULT_XMLDECL_BUFFER_SIZE = 64;
105
106 /** Default internal entity buffer size (1024). */
107 public static final int DEFAULT_INTERNAL_BUFFER_SIZE = 1024;
108
109 // feature identifiers
110
111 /** Feature identifier: validation. */
112 protected static final String VALIDATION =
113 Constants.SAX_FEATURE_PREFIX + Constants.VALIDATION_FEATURE;
114
115 /**
116 * standard uri conformant (strict uri).
117 * http://apache.org/xml/features/standard-uri-conformant
118 */
119 protected boolean fStrictURI;
120
121
122 /** Feature identifier: external general entities. */
123 protected static final String EXTERNAL_GENERAL_ENTITIES =
124 Constants.SAX_FEATURE_PREFIX + Constants.EXTERNAL_GENERAL_ENTITIES_FEATURE;
125
126 /** Feature identifier: external parameter entities. */
127 protected static final String EXTERNAL_PARAMETER_ENTITIES =
128 Constants.SAX_FEATURE_PREFIX + Constants.EXTERNAL_PARAMETER_ENTITIES_FEATURE;
129
130 /** Feature identifier: allow Java encodings. */
131 protected static final String ALLOW_JAVA_ENCODINGS =
132 Constants.XERCES_FEATURE_PREFIX + Constants.ALLOW_JAVA_ENCODINGS_FEATURE;
133
134 /** Feature identifier: warn on duplicate EntityDef */
135 protected static final String WARN_ON_DUPLICATE_ENTITYDEF =
136 Constants.XERCES_FEATURE_PREFIX +Constants.WARN_ON_DUPLICATE_ENTITYDEF_FEATURE;
137
138 /** Feature identifier: load external DTD. */
139 protected static final String LOAD_EXTERNAL_DTD =
140 Constants.XERCES_FEATURE_PREFIX + Constants.LOAD_EXTERNAL_DTD_FEATURE;
141
142 // property identifiers
143
144 /** Property identifier: symbol table. */
145 protected static final String SYMBOL_TABLE =
146 Constants.XERCES_PROPERTY_PREFIX + Constants.SYMBOL_TABLE_PROPERTY;
147
148 /** Property identifier: error reporter. */
149 protected static final String ERROR_REPORTER =
150 Constants.XERCES_PROPERTY_PREFIX + Constants.ERROR_REPORTER_PROPERTY;
151
152 /** Feature identifier: standard uri conformant */
153 protected static final String STANDARD_URI_CONFORMANT =
154 Constants.XERCES_FEATURE_PREFIX +Constants.STANDARD_URI_CONFORMANT_FEATURE;
155
156 /** Property identifier: entity resolver. */
157 protected static final String ENTITY_RESOLVER =
158 Constants.XERCES_PROPERTY_PREFIX + Constants.ENTITY_RESOLVER_PROPERTY;
159
160 protected static final String STAX_ENTITY_RESOLVER =
161 Constants.XERCES_PROPERTY_PREFIX + Constants.STAX_ENTITY_RESOLVER_PROPERTY;
162
163 // property identifier: ValidationManager
164 protected static final String VALIDATION_MANAGER =
165 Constants.XERCES_PROPERTY_PREFIX + Constants.VALIDATION_MANAGER_PROPERTY;
166
167 /** property identifier: buffer size. */
168 protected static final String BUFFER_SIZE =
169 Constants.XERCES_PROPERTY_PREFIX + Constants.BUFFER_SIZE_PROPERTY;
170
171 /** property identifier: security manager. */
172 protected static final String SECURITY_MANAGER =
173 Constants.XERCES_PROPERTY_PREFIX + Constants.SECURITY_MANAGER_PROPERTY;
174
175 protected static final String PARSER_SETTINGS =
176 Constants.XERCES_FEATURE_PREFIX + Constants.PARSER_SETTINGS;
177
178 /** Property identifier: Security property manager. */
179 private static final String XML_SECURITY_PROPERTY_MANAGER =
180 Constants.XML_SECURITY_PROPERTY_MANAGER;
181
182 /** access external dtd: file protocol */
183 static final String EXTERNAL_ACCESS_DEFAULT = Constants.EXTERNAL_ACCESS_DEFAULT;
184
185 // recognized features and properties
186
187 /** Recognized features. */
188 private static final String[] RECOGNIZED_FEATURES = {
189 VALIDATION,
190 EXTERNAL_GENERAL_ENTITIES,
191 EXTERNAL_PARAMETER_ENTITIES,
192 ALLOW_JAVA_ENCODINGS,
193 WARN_ON_DUPLICATE_ENTITYDEF,
194 STANDARD_URI_CONFORMANT,
195 XMLConstants.USE_CATALOG
196 };
197
198 /** Feature defaults. */
199 private static final Boolean[] FEATURE_DEFAULTS = {
200 null,
201 Boolean.TRUE,
202 Boolean.TRUE,
203 Boolean.TRUE,
204 Boolean.FALSE,
205 Boolean.FALSE,
206 JdkXmlUtils.USE_CATALOG_DEFAULT
207 };
208
209 /** Recognized properties. */
210 private static final String[] RECOGNIZED_PROPERTIES = {
211 SYMBOL_TABLE,
212 ERROR_REPORTER,
213 ENTITY_RESOLVER,
214 VALIDATION_MANAGER,
215 BUFFER_SIZE,
216 SECURITY_MANAGER,
217 XML_SECURITY_PROPERTY_MANAGER,
218 JdkXmlUtils.CATALOG_DEFER,
219 JdkXmlUtils.CATALOG_FILES,
220 JdkXmlUtils.CATALOG_PREFER,
221 JdkXmlUtils.CATALOG_RESOLVE,
222 JdkXmlUtils.CDATA_CHUNK_SIZE
223 };
224
225 /** Property defaults. */
226 private static final Object[] PROPERTY_DEFAULTS = {
227 null,
228 null,
229 null,
230 null,
231 DEFAULT_BUFFER_SIZE,
232 null,
233 null,
234 null,
235 null,
236 null,
237 null,
238 JdkXmlUtils.CDATA_CHUNK_SIZE_DEFAULT
239 };
240
241 private static final String XMLEntity = "[xml]".intern();
242 private static final String DTDEntity = "[dtd]".intern();
243
244 // debugging
245
246 /**
247 * Debug printing of buffer. This debugging flag works best when you
248 * resize the DEFAULT_BUFFER_SIZE down to something reasonable like
249 * 64 characters.
250 */
251 private static final boolean DEBUG_BUFFER = false;
252
253 /** warn on duplicate Entity declaration.
254 * http://apache.org/xml/features/warn-on-duplicate-entitydef
255 */
256 protected boolean fWarnDuplicateEntityDef;
257
258 /** Debug some basic entities. */
259 private static final boolean DEBUG_ENTITIES = false;
260
261 /** Debug switching readers for encodings. */
262 private static final boolean DEBUG_ENCODINGS = false;
263
264 // should be diplayed trace resolving messages
265 private static final boolean DEBUG_RESOLVER = false ;
266
267 //
268 // Data
269 //
270
271 // features
272
273 /**
274 * Validation. This feature identifier is:
275 * http://xml.org/sax/features/validation
276 */
277 protected boolean fValidation;
278
279 /**
280 * External general entities. This feature identifier is:
281 * http://xml.org/sax/features/external-general-entities
282 */
283 protected boolean fExternalGeneralEntities;
284
285 /**
286 * External parameter entities. This feature identifier is:
287 * http://xml.org/sax/features/external-parameter-entities
288 */
289 protected boolean fExternalParameterEntities;
290
291 /**
292 * Allow Java encoding names. This feature identifier is:
293 * http://apache.org/xml/features/allow-java-encodings
294 */
295 protected boolean fAllowJavaEncodings = true ;
296
297 /** Load external DTD. */
298 protected boolean fLoadExternalDTD = true;
299
300 // properties
301
302 /**
303 * Symbol table. This property identifier is:
304 * http://apache.org/xml/properties/internal/symbol-table
305 */
306 protected SymbolTable fSymbolTable;
307
308 /**
309 * Error reporter. This property identifier is:
310 * http://apache.org/xml/properties/internal/error-reporter
311 */
312 protected XMLErrorReporter fErrorReporter;
313
314 /**
315 * Entity resolver. This property identifier is:
316 * http://apache.org/xml/properties/internal/entity-resolver
317 */
318 protected XMLEntityResolver fEntityResolver;
319
320 /** Stax Entity Resolver. This property identifier is XMLInputFactory.ENTITY_RESOLVER */
321
322 protected StaxEntityResolverWrapper fStaxEntityResolver;
323
324 /** Property Manager. This is used from Stax */
325 protected PropertyManager fPropertyManager ;
326
327 /** StAX properties */
328 boolean fSupportDTD = true;
329 boolean fReplaceEntityReferences = true;
330 boolean fSupportExternalEntities = true;
331
332 /** used to restrict external access */
333 protected String fAccessExternalDTD = EXTERNAL_ACCESS_DEFAULT;
334
335 // settings
336
337 /**
338 * Validation manager. This property identifier is:
339 * http://apache.org/xml/properties/internal/validation-manager
340 */
341 protected ValidationManager fValidationManager;
342
343 // settings
344
345 /**
346 * Buffer size. We get this value from a property. The default size
347 * is used if the input buffer size property is not specified.
348 * REVISIT: do we need a property for internal entity buffer size?
349 */
350 protected int fBufferSize = DEFAULT_BUFFER_SIZE;
351
352 /** Security Manager */
353 protected XMLSecurityManager fSecurityManager = null;
354
355 protected XMLLimitAnalyzer fLimitAnalyzer = null;
356
357 protected int entityExpansionIndex;
358
359 /**
360 * True if the document entity is standalone. This should really
361 * only be set by the document source (e.g. XMLDocumentScanner).
362 */
363 protected boolean fStandalone;
364
365 // are the entities being parsed in the external subset?
366 // NOTE: this *is not* the same as whether they're external entities!
367 protected boolean fInExternalSubset = false;
368
369
370 // handlers
371 /** Entity handler. */
372 protected XMLEntityHandler fEntityHandler;
373
374 /** Current entity scanner */
375 protected XMLEntityScanner fEntityScanner ;
376
377 /** XML 1.0 entity scanner. */
378 protected XMLEntityScanner fXML10EntityScanner;
379
380 /** XML 1.1 entity scanner. */
381 protected XMLEntityScanner fXML11EntityScanner;
382
383 /** count of entities expanded: */
384 protected int fEntityExpansionCount = 0;
385
386 // entities
387
388 /** Entities. */
389 protected Map<String, Entity> fEntities = new HashMap<>();
390
391 /** Entity stack. */
392 protected Stack<Entity> fEntityStack = new Stack<>();
393
394 /** Current entity. */
395 protected Entity.ScannedEntity fCurrentEntity = null;
396
397 /** identify if the InputSource is created by a resolver */
398 boolean fISCreatedByResolver = false;
399
400 // shared context
401
402 protected XMLEntityStorage fEntityStorage ;
403
404 protected final Object [] defaultEncoding = new Object[]{"UTF-8", null};
405
406
407 // temp vars
408
409 /** Resource identifer. */
410 private final XMLResourceIdentifierImpl fResourceIdentifier = new XMLResourceIdentifierImpl();
411
412 /** Augmentations for entities. */
413 private final Augmentations fEntityAugs = new AugmentationsImpl();
414
415 /** Pool of character buffers. */
416 private CharacterBufferPool fBufferPool = new CharacterBufferPool(fBufferSize, DEFAULT_INTERNAL_BUFFER_SIZE);
417
418 /** indicate whether Catalog should be used for resolving external resources */
419 private boolean fUseCatalog = true;
420 CatalogFeatures fCatalogFeatures;
421 CatalogResolver fCatalogResolver;
422
423 private String fCatalogFile;
424 private String fDefer;
425 private String fPrefer;
426 private String fResolve;
427
428 //
429 // Constructors
430 //
431
432 /**
433 * If this constructor is used to create the object, reset() should be invoked on this object
434 */
435 public XMLEntityManager() {
436 //for entity managers not created by parsers
437 fSecurityManager = new XMLSecurityManager(true);
438 fEntityStorage = new XMLEntityStorage(this) ;
439 setScannerVersion(Constants.XML_VERSION_1_0);
440 } // <init>()
441
442 /** Default constructor. */
443 public XMLEntityManager(PropertyManager propertyManager) {
444 fPropertyManager = propertyManager ;
445 //pass a reference to current entity being scanned
446 //fEntityStorage = new XMLEntityStorage(fCurrentEntity) ;
447 fEntityStorage = new XMLEntityStorage(this) ;
448 fEntityScanner = new XMLEntityScanner(propertyManager, this) ;
449 reset(propertyManager);
450 } // <init>()
451
452 /**
453 * Adds an internal entity declaration.
454 * <p>
455 * <strong>Note:</strong> This method ignores subsequent entity
456 * declarations.
457 * <p>
458 * <strong>Note:</strong> The name should be a unique symbol. The
459 * SymbolTable can be used for this purpose.
460 *
461 * @param name The name of the entity.
462 * @param text The text of the entity.
463 *
464 * @see SymbolTable
465 */
466 public void addInternalEntity(String name, String text) {
467 if (!fEntities.containsKey(name)) {
468 Entity entity = new Entity.InternalEntity(name, text, fInExternalSubset);
469 fEntities.put(name, entity);
470 } else{
471 if(fWarnDuplicateEntityDef){
472 fErrorReporter.reportError(XMLMessageFormatter.XML_DOMAIN,
473 "MSG_DUPLICATE_ENTITY_DEFINITION",
474 new Object[]{ name },
475 XMLErrorReporter.SEVERITY_WARNING );
476 }
477 }
478
479 } // addInternalEntity(String,String)
480
481 /**
482 * Adds an external entity declaration.
483 * <p>
484 * <strong>Note:</strong> This method ignores subsequent entity
485 * declarations.
486 * <p>
487 * <strong>Note:</strong> The name should be a unique symbol. The
488 * SymbolTable can be used for this purpose.
489 *
490 * @param name The name of the entity.
491 * @param publicId The public identifier of the entity.
492 * @param literalSystemId The system identifier of the entity.
493 * @param baseSystemId The base system identifier of the entity.
494 * This is the system identifier of the entity
495 * where <em>the entity being added</em> and
496 * is used to expand the system identifier when
497 * the system identifier is a relative URI.
498 * When null the system identifier of the first
499 * external entity on the stack is used instead.
500 *
501 * @see SymbolTable
502 */
503 public void addExternalEntity(String name,
504 String publicId, String literalSystemId,
505 String baseSystemId) throws IOException {
506 if (!fEntities.containsKey(name)) {
507 if (baseSystemId == null) {
508 // search for the first external entity on the stack
509 int size = fEntityStack.size();
510 if (size == 0 && fCurrentEntity != null && fCurrentEntity.entityLocation != null) {
511 baseSystemId = fCurrentEntity.entityLocation.getExpandedSystemId();
512 }
513 for (int i = size - 1; i >= 0 ; i--) {
514 Entity.ScannedEntity externalEntity =
515 (Entity.ScannedEntity)fEntityStack.get(i);
516 if (externalEntity.entityLocation != null && externalEntity.entityLocation.getExpandedSystemId() != null) {
517 baseSystemId = externalEntity.entityLocation.getExpandedSystemId();
518 break;
519 }
520 }
521 }
522 Entity entity = new Entity.ExternalEntity(name,
523 new XMLEntityDescriptionImpl(name, publicId, literalSystemId, baseSystemId,
524 expandSystemId(literalSystemId, baseSystemId, false)), null, fInExternalSubset);
525 fEntities.put(name, entity);
526 } else{
527 if(fWarnDuplicateEntityDef){
528 fErrorReporter.reportError(XMLMessageFormatter.XML_DOMAIN,
529 "MSG_DUPLICATE_ENTITY_DEFINITION",
530 new Object[]{ name },
531 XMLErrorReporter.SEVERITY_WARNING );
532 }
533 }
534
535 } // addExternalEntity(String,String,String,String)
536
537
538 /**
539 * Adds an unparsed entity declaration.
540 * <p>
541 * <strong>Note:</strong> This method ignores subsequent entity
542 * declarations.
543 * <p>
544 * <strong>Note:</strong> The name should be a unique symbol. The
545 * SymbolTable can be used for this purpose.
546 *
547 * @param name The name of the entity.
548 * @param publicId The public identifier of the entity.
549 * @param systemId The system identifier of the entity.
550 * @param notation The name of the notation.
551 *
552 * @see SymbolTable
553 */
554 public void addUnparsedEntity(String name,
555 String publicId, String systemId,
556 String baseSystemId, String notation) {
557 if (!fEntities.containsKey(name)) {
558 Entity.ExternalEntity entity = new Entity.ExternalEntity(name,
559 new XMLEntityDescriptionImpl(name, publicId, systemId, baseSystemId, null),
560 notation, fInExternalSubset);
561 fEntities.put(name, entity);
562 } else{
563 if(fWarnDuplicateEntityDef){
564 fErrorReporter.reportError(XMLMessageFormatter.XML_DOMAIN,
565 "MSG_DUPLICATE_ENTITY_DEFINITION",
566 new Object[]{ name },
567 XMLErrorReporter.SEVERITY_WARNING );
568 }
569 }
570 } // addUnparsedEntity(String,String,String,String)
571
572
573 /** get the entity storage object from entity manager */
574 public XMLEntityStorage getEntityStore(){
575 return fEntityStorage ;
576 }
577
578 /** return the entity responsible for reading the entity */
579 public XMLEntityScanner getEntityScanner(){
580 if(fEntityScanner == null) {
581 // default to 1.0
582 if(fXML10EntityScanner == null) {
583 fXML10EntityScanner = new XMLEntityScanner();
584 }
585 fXML10EntityScanner.reset(fSymbolTable, this, fErrorReporter);
586 fEntityScanner = fXML10EntityScanner;
587 }
588 return fEntityScanner;
589
590 }
591
592 public void setScannerVersion(short version) {
593
594 if(version == Constants.XML_VERSION_1_0) {
595 if(fXML10EntityScanner == null) {
596 fXML10EntityScanner = new XMLEntityScanner();
597 }
598 fXML10EntityScanner.reset(fSymbolTable, this, fErrorReporter);
599 fEntityScanner = fXML10EntityScanner;
600 fEntityScanner.setCurrentEntity(fCurrentEntity);
601 } else {
602 if(fXML11EntityScanner == null) {
603 fXML11EntityScanner = new XML11EntityScanner();
604 }
605 fXML11EntityScanner.reset(fSymbolTable, this, fErrorReporter);
606 fEntityScanner = fXML11EntityScanner;
607 fEntityScanner.setCurrentEntity(fCurrentEntity);
608 }
609
610 }
611
612 /**
613 * This method uses the passed-in XMLInputSource to make
614 * fCurrentEntity usable for reading.
615 *
616 * @param reference flag to indicate whether the entity is an Entity Reference.
617 * @param name name of the entity (XML is it's the document entity)
618 * @param xmlInputSource the input source, with sufficient information
619 * to begin scanning characters.
620 * @param literal True if this entity is started within a
621 * literal value.
622 * @param isExternal whether this entity should be treated as an internal or external entity.
623 * @throws IOException if anything can't be read
624 * XNIException If any parser-specific goes wrong.
625 * @return the encoding of the new entity or null if a character stream was employed
626 */
627 public String setupCurrentEntity(boolean reference, String name, XMLInputSource xmlInputSource,
628 boolean literal, boolean isExternal)
629 throws IOException, XNIException {
630 // get information
631
632 final String publicId = xmlInputSource.getPublicId();
633 String literalSystemId = xmlInputSource.getSystemId();
634 String baseSystemId = xmlInputSource.getBaseSystemId();
635 String encoding = xmlInputSource.getEncoding();
636 final boolean encodingExternallySpecified = (encoding != null);
637 Boolean isBigEndian = null;
638
639 // create reader
640 InputStream stream = null;
641 Reader reader = xmlInputSource.getCharacterStream();
642
643 // First chance checking strict URI
644 String expandedSystemId = expandSystemId(literalSystemId, baseSystemId, fStrictURI);
645 if (baseSystemId == null) {
646 baseSystemId = expandedSystemId;
647 }
648 if (reader == null) {
649 stream = xmlInputSource.getByteStream();
650 if (stream == null) {
651 URL location = new URL(expandedSystemId);
652 URLConnection connect = location.openConnection();
653 if (!(connect instanceof HttpURLConnection)) {
654 stream = connect.getInputStream();
655 }
656 else {
657 boolean followRedirects = true;
658
659 // setup URLConnection if we have an HTTPInputSource
660 if (xmlInputSource instanceof HTTPInputSource) {
661 final HttpURLConnection urlConnection = (HttpURLConnection) connect;
662 final HTTPInputSource httpInputSource = (HTTPInputSource) xmlInputSource;
663
664 // set request properties
665 Iterator<Map.Entry<String, String>> propIter = httpInputSource.getHTTPRequestProperties();
666 while (propIter.hasNext()) {
667 Map.Entry<String, String> entry = propIter.next();
668 urlConnection.setRequestProperty(entry.getKey(), entry.getValue());
669 }
670
671 // set preference for redirection
672 followRedirects = httpInputSource.getFollowHTTPRedirects();
673 if (!followRedirects) {
674 urlConnection.setInstanceFollowRedirects(followRedirects);
675 }
676 }
677
678 stream = connect.getInputStream();
679
680 // REVISIT: If the URLConnection has external encoding
681 // information, we should be reading it here. It's located
682 // in the charset parameter of Content-Type. -- mrglavas
683
684 if (followRedirects) {
685 String redirect = connect.getURL().toString();
686 // E43: Check if the URL was redirected, and then
687 // update literal and expanded system IDs if needed.
688 if (!redirect.equals(expandedSystemId)) {
689 literalSystemId = redirect;
690 expandedSystemId = redirect;
691 }
692 }
693 }
694 }
695
696 // wrap this stream in RewindableInputStream
697 stream = new RewindableInputStream(stream);
698
699 // perform auto-detect of encoding if necessary
700 if (encoding == null) {
701 // read first four bytes and determine encoding
702 final byte[] b4 = new byte[4];
703 int count = 0;
704 for (; count<4; count++ ) {
705 b4[count] = (byte)stream.read();
706 }
707 if (count == 4) {
708 Object [] encodingDesc = getEncodingName(b4, count);
709 encoding = (String)(encodingDesc[0]);
710 isBigEndian = (Boolean)(encodingDesc[1]);
711
712 stream.reset();
713 // Special case UTF-8 files with BOM created by Microsoft
714 // tools. It's more efficient to consume the BOM than make
715 // the reader perform extra checks. -Ac
716 if (count > 2 && encoding.equals("UTF-8")) {
717 int b0 = b4[0] & 0xFF;
718 int b1 = b4[1] & 0xFF;
719 int b2 = b4[2] & 0xFF;
720 if (b0 == 0xEF && b1 == 0xBB && b2 == 0xBF) {
721 // ignore first three bytes...
722 stream.skip(3);
723 }
724 }
725 reader = createReader(stream, encoding, isBigEndian);
726 } else {
727 reader = createReader(stream, encoding, isBigEndian);
728 }
729 }
730
731 // use specified encoding
732 else {
733 encoding = encoding.toUpperCase(Locale.ENGLISH);
734
735 // If encoding is UTF-8, consume BOM if one is present.
736 if (encoding.equals("UTF-8")) {
737 final int[] b3 = new int[3];
738 int count = 0;
739 for (; count < 3; ++count) {
740 b3[count] = stream.read();
741 if (b3[count] == -1)
742 break;
743 }
744 if (count == 3) {
745 if (b3[0] != 0xEF || b3[1] != 0xBB || b3[2] != 0xBF) {
746 // First three bytes are not BOM, so reset.
747 stream.reset();
748 }
749 } else {
750 stream.reset();
751 }
752 }
753 // If encoding is UTF-16, we still need to read the first four bytes
754 // in order to discover the byte order.
755 else if (encoding.equals("UTF-16")) {
756 final int[] b4 = new int[4];
757 int count = 0;
758 for (; count < 4; ++count) {
759 b4[count] = stream.read();
760 if (b4[count] == -1)
761 break;
762 }
763 stream.reset();
764
765 String utf16Encoding = "UTF-16";
766 if (count >= 2) {
767 final int b0 = b4[0];
768 final int b1 = b4[1];
769 if (b0 == 0xFE && b1 == 0xFF) {
770 // UTF-16, big-endian
771 utf16Encoding = "UTF-16BE";
772 isBigEndian = Boolean.TRUE;
773 }
774 else if (b0 == 0xFF && b1 == 0xFE) {
775 // UTF-16, little-endian
776 utf16Encoding = "UTF-16LE";
777 isBigEndian = Boolean.FALSE;
778 }
779 else if (count == 4) {
780 final int b2 = b4[2];
781 final int b3 = b4[3];
782 if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x3F) {
783 // UTF-16, big-endian, no BOM
784 utf16Encoding = "UTF-16BE";
785 isBigEndian = Boolean.TRUE;
786 }
787 if (b0 == 0x3C && b1 == 0x00 && b2 == 0x3F && b3 == 0x00) {
788 // UTF-16, little-endian, no BOM
789 utf16Encoding = "UTF-16LE";
790 isBigEndian = Boolean.FALSE;
791 }
792 }
793 }
794 reader = createReader(stream, utf16Encoding, isBigEndian);
795 }
796 // If encoding is UCS-4, we still need to read the first four bytes
797 // in order to discover the byte order.
798 else if (encoding.equals("ISO-10646-UCS-4")) {
799 final int[] b4 = new int[4];
800 int count = 0;
801 for (; count < 4; ++count) {
802 b4[count] = stream.read();
803 if (b4[count] == -1)
804 break;
805 }
806 stream.reset();
807
808 // Ignore unusual octet order for now.
809 if (count == 4) {
810 // UCS-4, big endian (1234)
811 if (b4[0] == 0x00 && b4[1] == 0x00 && b4[2] == 0x00 && b4[3] == 0x3C) {
812 isBigEndian = Boolean.TRUE;
813 }
814 // UCS-4, little endian (1234)
815 else if (b4[0] == 0x3C && b4[1] == 0x00 && b4[2] == 0x00 && b4[3] == 0x00) {
816 isBigEndian = Boolean.FALSE;
817 }
818 }
819 }
820 // If encoding is UCS-2, we still need to read the first four bytes
821 // in order to discover the byte order.
822 else if (encoding.equals("ISO-10646-UCS-2")) {
823 final int[] b4 = new int[4];
824 int count = 0;
825 for (; count < 4; ++count) {
826 b4[count] = stream.read();
827 if (b4[count] == -1)
828 break;
829 }
830 stream.reset();
831
832 if (count == 4) {
833 // UCS-2, big endian
834 if (b4[0] == 0x00 && b4[1] == 0x3C && b4[2] == 0x00 && b4[3] == 0x3F) {
835 isBigEndian = Boolean.TRUE;
836 }
837 // UCS-2, little endian
838 else if (b4[0] == 0x3C && b4[1] == 0x00 && b4[2] == 0x3F && b4[3] == 0x00) {
839 isBigEndian = Boolean.FALSE;
840 }
841 }
842 }
843
844 reader = createReader(stream, encoding, isBigEndian);
845 }
846
847 // read one character at a time so we don't jump too far
848 // ahead, converting characters from the byte stream in
849 // the wrong encoding
850 if (DEBUG_ENCODINGS) {
851 System.out.println("$$$ no longer wrapping reader in OneCharReader");
852 }
853 //reader = new OneCharReader(reader);
854 }
855
856 // We've seen a new Reader.
857 // Push it on the stack so we can close it later.
858 //fOwnReaders.add(reader);
859
860 // push entity on stack
861 if (fCurrentEntity != null) {
862 fEntityStack.push(fCurrentEntity);
863 }
864
865 // create entity
866 /* if encoding is specified externally, 'encoding' information present
867 * in the prolog of the XML document is not considered. Hence, prolog can
868 * be read in Chunks of data instead of byte by byte.
869 */
870 fCurrentEntity = new Entity.ScannedEntity(reference, name,
871 new XMLResourceIdentifierImpl(publicId, literalSystemId, baseSystemId, expandedSystemId),
872 stream, reader, encoding, literal, encodingExternallySpecified, isExternal);
873 fCurrentEntity.setEncodingExternallySpecified(encodingExternallySpecified);
874 fEntityScanner.setCurrentEntity(fCurrentEntity);
875 fResourceIdentifier.setValues(publicId, literalSystemId, baseSystemId, expandedSystemId);
876 if (fLimitAnalyzer != null) {
877 fLimitAnalyzer.startEntity(name);
878 }
879 return encoding;
880 } //setupCurrentEntity(String, XMLInputSource, boolean, boolean): String
881
882
883 /**
884 * Checks whether an entity given by name is external.
885 *
886 * @param entityName The name of the entity to check.
887 * @return True if the entity is external, false otherwise
888 * (including when the entity is not declared).
889 */
890 public boolean isExternalEntity(String entityName) {
891
892 Entity entity = fEntities.get(entityName);
893 if (entity == null) {
894 return false;
895 }
896 return entity.isExternal();
897 }
898
899 /**
900 * Checks whether the declaration of an entity given by name is
901 * // in the external subset.
902 *
903 * @param entityName The name of the entity to check.
904 * @return True if the entity was declared in the external subset, false otherwise
905 * (including when the entity is not declared).
906 */
907 public boolean isEntityDeclInExternalSubset(String entityName) {
908
909 Entity entity = fEntities.get(entityName);
910 if (entity == null) {
911 return false;
912 }
913 return entity.isEntityDeclInExternalSubset();
914 }
915
916
917
918 //
919 // Public methods
920 //
921
922 /**
923 * Sets whether the document entity is standalone.
924 *
925 * @param standalone True if document entity is standalone.
926 */
927 public void setStandalone(boolean standalone) {
928 fStandalone = standalone;
929 }
930 // setStandalone(boolean)
931
932 /** Returns true if the document entity is standalone. */
933 public boolean isStandalone() {
934 return fStandalone;
935 } //isStandalone():boolean
936
937 public boolean isDeclaredEntity(String entityName) {
938
939 Entity entity = fEntities.get(entityName);
940 return entity != null;
941 }
942
943 public boolean isUnparsedEntity(String entityName) {
944
945 Entity entity = fEntities.get(entityName);
946 if (entity == null) {
947 return false;
948 }
949 return entity.isUnparsed();
950 }
951
952
953
954 // this simply returns the fResourceIdentifier object;
955 // this should only be used with caution by callers that
956 // carefully manage the entity manager's behaviour, so that
957 // this doesn't returning meaningless or misleading data.
958 // @return a reference to the current fResourceIdentifier object
959 public XMLResourceIdentifier getCurrentResourceIdentifier() {
960 return fResourceIdentifier;
961 }
962
963 /**
964 * Sets the entity handler. When an entity starts and ends, the
965 * entity handler is notified of the change.
966 *
967 * @param entityHandler The new entity handler.
968 */
969
970 public void setEntityHandler(com.sun.org.apache.xerces.internal.impl.XMLEntityHandler entityHandler) {
971 fEntityHandler = entityHandler;
972 } // setEntityHandler(XMLEntityHandler)
973
974 //this function returns StaxXMLInputSource
975 public StaxXMLInputSource resolveEntityAsPerStax(XMLResourceIdentifier resourceIdentifier) throws java.io.IOException{
976
977 if(resourceIdentifier == null ) return null;
978
979 String publicId = resourceIdentifier.getPublicId();
980 String literalSystemId = resourceIdentifier.getLiteralSystemId();
981 String baseSystemId = resourceIdentifier.getBaseSystemId();
982 String expandedSystemId = resourceIdentifier.getExpandedSystemId();
983 // if no base systemId given, assume that it's relative
984 // to the systemId of the current scanned entity
985 // Sometimes the system id is not (properly) expanded.
986 // We need to expand the system id if:
987 // a. the expanded one was null; or
988 // b. the base system id was null, but becomes non-null from the current entity.
989 boolean needExpand = (expandedSystemId == null);
990 // REVISIT: why would the baseSystemId ever be null? if we
991 // didn't have to make this check we wouldn't have to reuse the
992 // fXMLResourceIdentifier object...
993 if (baseSystemId == null && fCurrentEntity != null && fCurrentEntity.entityLocation != null) {
994 baseSystemId = fCurrentEntity.entityLocation.getExpandedSystemId();
995 if (baseSystemId != null)
996 needExpand = true;
997 }
998 if (needExpand)
999 expandedSystemId = expandSystemId(literalSystemId, baseSystemId,false);
1000
1001 // give the entity resolver a chance
1002 StaxXMLInputSource staxInputSource = null;
1003 XMLInputSource xmlInputSource = null;
1004
1005 XMLResourceIdentifierImpl ri = null;
1006
1007 if (resourceIdentifier instanceof XMLResourceIdentifierImpl) {
1008 ri = (XMLResourceIdentifierImpl)resourceIdentifier;
1009 } else {
1010 fResourceIdentifier.clear();
1011 ri = fResourceIdentifier;
1012 }
1013 ri.setValues(publicId, literalSystemId, baseSystemId, expandedSystemId);
1014 if(DEBUG_RESOLVER){
1015 System.out.println("BEFORE Calling resolveEntity") ;
1016 }
1017
1018 fISCreatedByResolver = false;
1019 //either of Stax or Xerces would be null
1020 if(fStaxEntityResolver != null){
1021 staxInputSource = fStaxEntityResolver.resolveEntity(ri);
1022 if(staxInputSource != null) {
1023 fISCreatedByResolver = true;
1024 }
1025 }
1026
1027 if(fEntityResolver != null){
1028 xmlInputSource = fEntityResolver.resolveEntity(ri);
1029 if(xmlInputSource != null) {
1030 fISCreatedByResolver = true;
1031 }
1032 }
1033
1034 if(xmlInputSource != null){
1035 //wrap this XMLInputSource to StaxInputSource
1036 staxInputSource = new StaxXMLInputSource(xmlInputSource, fISCreatedByResolver);
1037 }
1038
1039 if (staxInputSource == null && fUseCatalog) {
1040 if (fCatalogFeatures == null) {
1041 fCatalogFeatures = JdkXmlUtils.getCatalogFeatures(fDefer, fCatalogFile, fPrefer, fResolve);
1042 }
1043 fCatalogFile = fCatalogFeatures.get(Feature.FILES);
1044 if (fCatalogFile != null) {
1045 try {
1046 if (fCatalogResolver == null) {
1047 fCatalogResolver = CatalogManager.catalogResolver(fCatalogFeatures);
1048 }
1049 InputSource is = fCatalogResolver.resolveEntity(publicId, literalSystemId);
1050 if (is != null && !is.isEmpty()) {
1051 staxInputSource = new StaxXMLInputSource(new XMLInputSource(is, true), true);
1052 }
1053 } catch (CatalogException e) {
1054 fErrorReporter.reportError(XMLMessageFormatter.XML_DOMAIN,"CatalogException",
1055 new Object[]{SecuritySupport.sanitizePath(fCatalogFile)},
1056 XMLErrorReporter.SEVERITY_FATAL_ERROR, e );
1057 }
1058 }
1059 }
1060
1061 // do default resolution
1062 //this works for both stax & Xerces, if staxInputSource is null,
1063 //it means parser need to revert to default resolution
1064 if (staxInputSource == null) {
1065 // REVISIT: when systemId is null, I think we should return null.
1066 // is this the right solution? -SG
1067 //if (systemId != null)
1068 staxInputSource = new StaxXMLInputSource(
1069 new XMLInputSource(publicId, literalSystemId, baseSystemId, true), false);
1070 }else if(staxInputSource.hasXMLStreamOrXMLEventReader()){
1071 //Waiting for the clarification from EG. - nb
1072 }
1073
1074 if (DEBUG_RESOLVER) {
1075 System.err.println("XMLEntityManager.resolveEntity(" + publicId + ")");
1076 System.err.println(" = " + xmlInputSource);
1077 }
1078
1079 return staxInputSource;
1080
1081 }
1082
1083 /**
1084 * Resolves the specified public and system identifiers. This
1085 * method first attempts to resolve the entity based on the
1086 * EntityResolver registered by the application. If no entity
1087 * resolver is registered or if the registered entity handler
1088 * is unable to resolve the entity, then default entity
1089 * resolution will occur.
1090 *
1091 * @param publicId The public identifier of the entity.
1092 * @param systemId The system identifier of the entity.
1093 * @param baseSystemId The base system identifier of the entity.
1094 * This is the system identifier of the current
1095 * entity and is used to expand the system
1096 * identifier when the system identifier is a
1097 * relative URI.
1098 *
1099 * @return Returns an input source that wraps the resolved entity.
1100 * This method will never return null.
1101 *
1102 * @throws IOException Thrown on i/o error.
1103 * @throws XNIException Thrown by entity resolver to signal an error.
1104 */
1105 public XMLInputSource resolveEntity(XMLResourceIdentifier resourceIdentifier) throws IOException, XNIException {
1106 if(resourceIdentifier == null ) return null;
1107 String publicId = resourceIdentifier.getPublicId();
1108 String literalSystemId = resourceIdentifier.getLiteralSystemId();
1109 String baseSystemId = resourceIdentifier.getBaseSystemId();
1110 String expandedSystemId = resourceIdentifier.getExpandedSystemId();
1111
1112 // if no base systemId given, assume that it's relative
1113 // to the systemId of the current scanned entity
1114 // Sometimes the system id is not (properly) expanded.
1115 // We need to expand the system id if:
1116 // a. the expanded one was null; or
1117 // b. the base system id was null, but becomes non-null from the current entity.
1118 boolean needExpand = (expandedSystemId == null);
1119 // REVISIT: why would the baseSystemId ever be null? if we
1120 // didn't have to make this check we wouldn't have to reuse the
1121 // fXMLResourceIdentifier object...
1122 if (baseSystemId == null && fCurrentEntity != null && fCurrentEntity.entityLocation != null) {
1123 baseSystemId = fCurrentEntity.entityLocation.getExpandedSystemId();
1124 if (baseSystemId != null)
1125 needExpand = true;
1126 }
1127 if (needExpand)
1128 expandedSystemId = expandSystemId(literalSystemId, baseSystemId,false);
1129
1130 // give the entity resolver a chance
1131 XMLInputSource xmlInputSource = null;
1132
1133 if (fEntityResolver != null) {
1134 resourceIdentifier.setBaseSystemId(baseSystemId);
1135 resourceIdentifier.setExpandedSystemId(expandedSystemId);
1136 xmlInputSource = fEntityResolver.resolveEntity(resourceIdentifier);
1137 }
1138
1139 if (xmlInputSource == null && fUseCatalog) {
1140 if (fCatalogFeatures == null) {
1141 fCatalogFeatures = JdkXmlUtils.getCatalogFeatures(fDefer, fCatalogFile, fPrefer, fResolve);
1142 }
1143 fCatalogFile = fCatalogFeatures.get(Feature.FILES);
1144 if (fCatalogFile != null) {
1145 /*
1146 since the method can be called from various processors, both
1147 EntityResolver and URIResolver are used to attempt to find
1148 a match
1149 */
1150 InputSource is = null;
1151 try {
1152 if (fCatalogResolver == null) {
1153 fCatalogResolver = CatalogManager.catalogResolver(fCatalogFeatures);
1154 }
1155 String pid = (publicId != null? publicId : resourceIdentifier.getNamespace());
1156 if (pid != null || literalSystemId != null) {
1157 is = fCatalogResolver.resolveEntity(pid, literalSystemId);
1158 }
1159 } catch (CatalogException e) {}
1160
1161 if (is != null && !is.isEmpty()) {
1162 xmlInputSource = new XMLInputSource(is, true);
1163 } else if (literalSystemId != null) {
1164 if (fCatalogResolver == null) {
1165 fCatalogResolver = CatalogManager.catalogResolver(fCatalogFeatures);
1166 }
1167
1168 Source source = null;
1169 try {
1170 source = fCatalogResolver.resolve(literalSystemId, baseSystemId);
1171 } catch (CatalogException e) {
1172 throw new XNIException(e);
1173 }
1174 if (source != null && !source.isEmpty()) {
1175 xmlInputSource = new XMLInputSource(publicId, source.getSystemId(), baseSystemId, true);
1176 }
1177 }
1178 }
1179 }
1180
1181 // do default resolution
1182 // REVISIT: what's the correct behavior if the user provided an entity
1183 // resolver (fEntityResolver != null), but resolveEntity doesn't return
1184 // an input source (xmlInputSource == null)?
1185 // do we do default resolution, or do we just return null? -SG
1186 if (xmlInputSource == null) {
1187 // REVISIT: when systemId is null, I think we should return null.
1188 // is this the right solution? -SG
1189 //if (systemId != null)
1190 xmlInputSource = new XMLInputSource(publicId, literalSystemId, baseSystemId, false);
1191 }
1192
1193 if (DEBUG_RESOLVER) {
1194 System.err.println("XMLEntityManager.resolveEntity(" + publicId + ")");
1195 System.err.println(" = " + xmlInputSource);
1196 }
1197
1198 return xmlInputSource;
1199
1200 } // resolveEntity(XMLResourceIdentifier):XMLInputSource
1201
1202 /**
1203 * Starts a named entity.
1204 *
1205 * @param isGE flag to indicate whether the entity is a General Entity
1206 * @param entityName The name of the entity to start.
1207 * @param literal True if this entity is started within a literal
1208 * value.
1209 *
1210 * @throws IOException Thrown on i/o error.
1211 * @throws XNIException Thrown by entity handler to signal an error.
1212 */
1213 public void startEntity(boolean isGE, String entityName, boolean literal)
1214 throws IOException, XNIException {
1215
1216 // was entity declared?
1217 Entity entity = fEntityStorage.getEntity(entityName);
1218 if (entity == null) {
1219 if (fEntityHandler != null) {
1220 String encoding = null;
1221 fResourceIdentifier.clear();
1222 fEntityAugs.removeAllItems();
1223 fEntityAugs.putItem(Constants.ENTITY_SKIPPED, Boolean.TRUE);
1224 fEntityHandler.startEntity(entityName, fResourceIdentifier, encoding, fEntityAugs);
1225 fEntityAugs.removeAllItems();
1226 fEntityAugs.putItem(Constants.ENTITY_SKIPPED, Boolean.TRUE);
1227 fEntityHandler.endEntity(entityName, fEntityAugs);
1228 }
1229 return;
1230 }
1231
1232 // should we skip external entities?
1233 boolean external = entity.isExternal();
1234 Entity.ExternalEntity externalEntity = null;
1235 String extLitSysId = null, extBaseSysId = null, expandedSystemId = null;
1236 if (external) {
1237 externalEntity = (Entity.ExternalEntity)entity;
1238 extLitSysId = (externalEntity.entityLocation != null ? externalEntity.entityLocation.getLiteralSystemId() : null);
1239 extBaseSysId = (externalEntity.entityLocation != null ? externalEntity.entityLocation.getBaseSystemId() : null);
1240 expandedSystemId = expandSystemId(extLitSysId, extBaseSysId);
1241 boolean unparsed = entity.isUnparsed();
1242 boolean parameter = entityName.startsWith("%");
1243 boolean general = !parameter;
1244 if (unparsed || (general && !fExternalGeneralEntities) ||
1245 (parameter && !fExternalParameterEntities) ||
1246 !fSupportDTD || !fSupportExternalEntities) {
1247
1248 if (fEntityHandler != null) {
1249 fResourceIdentifier.clear();
1250 final String encoding = null;
1251 fResourceIdentifier.setValues(
1252 (externalEntity.entityLocation != null ? externalEntity.entityLocation.getPublicId() : null),
1253 extLitSysId, extBaseSysId, expandedSystemId);
1254 fEntityAugs.removeAllItems();
1255 fEntityAugs.putItem(Constants.ENTITY_SKIPPED, Boolean.TRUE);
1256 fEntityHandler.startEntity(entityName, fResourceIdentifier, encoding, fEntityAugs);
1257 fEntityAugs.removeAllItems();
1258 fEntityAugs.putItem(Constants.ENTITY_SKIPPED, Boolean.TRUE);
1259 fEntityHandler.endEntity(entityName, fEntityAugs);
1260 }
1261 return;
1262 }
1263 }
1264
1265 // is entity recursive?
1266 int size = fEntityStack.size();
1267 for (int i = size; i >= 0; i--) {
1268 Entity activeEntity = i == size
1269 ? fCurrentEntity
1270 : fEntityStack.get(i);
1271 if (activeEntity.name == entityName) {
1272 String path = entityName;
1273 for (int j = i + 1; j < size; j++) {
1274 activeEntity = fEntityStack.get(j);
1275 path = path + " -> " + activeEntity.name;
1276 }
1277 path = path + " -> " + fCurrentEntity.name;
1278 path = path + " -> " + entityName;
1279 fErrorReporter.reportError(this.getEntityScanner(),XMLMessageFormatter.XML_DOMAIN,
1280 "RecursiveReference",
1281 new Object[] { entityName, path },
1282 XMLErrorReporter.SEVERITY_FATAL_ERROR);
1283
1284 if (fEntityHandler != null) {
1285 fResourceIdentifier.clear();
1286 final String encoding = null;
1287 if (external) {
1288 fResourceIdentifier.setValues(
1289 (externalEntity.entityLocation != null ? externalEntity.entityLocation.getPublicId() : null),
1290 extLitSysId, extBaseSysId, expandedSystemId);
1291 }
1292 fEntityAugs.removeAllItems();
1293 fEntityAugs.putItem(Constants.ENTITY_SKIPPED, Boolean.TRUE);
1294 fEntityHandler.startEntity(entityName, fResourceIdentifier, encoding, fEntityAugs);
1295 fEntityAugs.removeAllItems();
1296 fEntityAugs.putItem(Constants.ENTITY_SKIPPED, Boolean.TRUE);
1297 fEntityHandler.endEntity(entityName, fEntityAugs);
1298 }
1299
1300 return;
1301 }
1302 }
1303
1304 // resolve external entity
1305 StaxXMLInputSource staxInputSource = null;
1306 XMLInputSource xmlInputSource = null ;
1307
1308 if (external) {
1309 staxInputSource = resolveEntityAsPerStax(externalEntity.entityLocation);
1310 /** xxx: Waiting from the EG
1311 * //simply return if there was entity resolver registered and application
1312 * //returns either XMLStreamReader or XMLEventReader.
1313 * if(staxInputSource.hasXMLStreamOrXMLEventReader()) return ;
1314 */
1315 xmlInputSource = staxInputSource.getXMLInputSource() ;
1316 if (!fISCreatedByResolver) {
1317 //let the not-LoadExternalDTD or not-SupportDTD process to handle the situation
1318 if (fLoadExternalDTD) {
1319 String accessError = SecuritySupport.checkAccess(expandedSystemId, fAccessExternalDTD, Constants.ACCESS_EXTERNAL_ALL);
1320 if (accessError != null) {
1321 fErrorReporter.reportError(this.getEntityScanner(),XMLMessageFormatter.XML_DOMAIN,
1322 "AccessExternalEntity",
1323 new Object[] { SecuritySupport.sanitizePath(expandedSystemId), accessError },
1324 XMLErrorReporter.SEVERITY_FATAL_ERROR);
1325 }
1326 }
1327 }
1328 }
1329 // wrap internal entity
1330 else {
1331 Entity.InternalEntity internalEntity = (Entity.InternalEntity)entity;
1332 Reader reader = new StringReader(internalEntity.text);
1333 xmlInputSource = new XMLInputSource(null, null, null, reader, null);
1334 }
1335
1336 // start the entity
1337 startEntity(isGE, entityName, xmlInputSource, literal, external);
1338
1339 } // startEntity(String,boolean)
1340
1341 /**
1342 * Starts the document entity. The document entity has the "[xml]"
1343 * pseudo-name.
1344 *
1345 * @param xmlInputSource The input source of the document entity.
1346 *
1347 * @throws IOException Thrown on i/o error.
1348 * @throws XNIException Thrown by entity handler to signal an error.
1349 */
1350 public void startDocumentEntity(XMLInputSource xmlInputSource)
1351 throws IOException, XNIException {
1352 startEntity(false, XMLEntity, xmlInputSource, false, true);
1353 } // startDocumentEntity(XMLInputSource)
1354
1355 //xxx these methods are not required.
1356 /**
1357 * Starts the DTD entity. The DTD entity has the "[dtd]"
1358 * pseudo-name.
1359 *
1360 * @param xmlInputSource The input source of the DTD entity.
1361 *
1362 * @throws IOException Thrown on i/o error.
1363 * @throws XNIException Thrown by entity handler to signal an error.
1364 */
1365 public void startDTDEntity(XMLInputSource xmlInputSource)
1366 throws IOException, XNIException {
1367 startEntity(false, DTDEntity, xmlInputSource, false, true);
1368 } // startDTDEntity(XMLInputSource)
1369
1370 // indicate start of external subset so that
1371 // location of entity decls can be tracked
1372 public void startExternalSubset() {
1373 fInExternalSubset = true;
1374 }
1375
1376 public void endExternalSubset() {
1377 fInExternalSubset = false;
1378 }
1379
1380 /**
1381 * Starts an entity.
1382 * <p>
1383 * This method can be used to insert an application defined XML
1384 * entity stream into the parsing stream.
1385 *
1386 * @param isGE flag to indicate whether the entity is a General Entity
1387 * @param name The name of the entity.
1388 * @param xmlInputSource The input source of the entity.
1389 * @param literal True if this entity is started within a
1390 * literal value.
1391 * @param isExternal whether this entity should be treated as an internal or external entity.
1392 *
1393 * @throws IOException Thrown on i/o error.
1394 * @throws XNIException Thrown by entity handler to signal an error.
1395 */
1396 public void startEntity(boolean isGE, String name,
1397 XMLInputSource xmlInputSource,
1398 boolean literal, boolean isExternal)
1399 throws IOException, XNIException {
1400
1401 String encoding = setupCurrentEntity(isGE, name, xmlInputSource, literal, isExternal);
1402
1403 //when entity expansion limit is set by the Application, we need to
1404 //check for the entity expansion limit set by the parser, if number of entity
1405 //expansions exceeds the entity expansion limit, parser will throw fatal error.
1406 // Note that this represents the nesting level of open entities.
1407 fEntityExpansionCount++;
1408 if(fLimitAnalyzer != null) {
1409 fLimitAnalyzer.addValue(entityExpansionIndex, name, 1);
1410 }
1411 if( fSecurityManager != null && fSecurityManager.isOverLimit(entityExpansionIndex, fLimitAnalyzer)){
1412 fSecurityManager.debugPrint(fLimitAnalyzer);
1413 fErrorReporter.reportError(XMLMessageFormatter.XML_DOMAIN,"EntityExpansionLimit",
1414 new Object[]{fSecurityManager.getLimitValueByIndex(entityExpansionIndex)},
1415 XMLErrorReporter.SEVERITY_FATAL_ERROR );
1416 // is there anything better to do than reset the counter?
1417 // at least one can envision debugging applications where this might
1418 // be useful...
1419 fEntityExpansionCount = 0;
1420 }
1421
1422 // call handler
1423 if (fEntityHandler != null) {
1424 fEntityHandler.startEntity(name, fResourceIdentifier, encoding, null);
1425 }
1426
1427 } // startEntity(String,XMLInputSource)
1428
1429 /**
1430 * Return the current entity being scanned. Current entity is SET using startEntity function.
1431 * @return Entity.ScannedEntity
1432 */
1433
1434 public Entity.ScannedEntity getCurrentEntity(){
1435 return fCurrentEntity ;
1436 }
1437
1438 /**
1439 * Return the top level entity handled by this manager, or null
1440 * if no entity was added.
1441 */
1442 public Entity.ScannedEntity getTopLevelEntity() {
1443 return (Entity.ScannedEntity)
1444 (fEntityStack.empty() ? null : fEntityStack.get(0));
1445 }
1446
1447
1448 /**
1449 * Close all opened InputStreams and Readers opened by this parser.
1450 */
1451 public void closeReaders() {
1452 /** this call actually does nothing, readers are closed in the endEntity method
1453 * through the current entity.
1454 * The change seems to have happened during the jdk6 development with the
1455 * addition of StAX
1456 **/
1457 }
1458
1459 public void endEntity() throws IOException, XNIException {
1460
1461 // call handler
1462 if (DEBUG_BUFFER) {
1463 System.out.print("(endEntity: ");
1464 print();
1465 System.out.println();
1466 }
1467 //pop the entity from the stack
1468 Entity.ScannedEntity entity = fEntityStack.size() > 0 ? (Entity.ScannedEntity)fEntityStack.pop() : null ;
1469
1470 /** need to close the reader first since the program can end
1471 * prematurely (e.g. fEntityHandler.endEntity may throw exception)
1472 * leaving the reader open
1473 */
1474 //close the reader
1475 if(fCurrentEntity != null){
1476 //close the reader
1477 try{
1478 if (fLimitAnalyzer != null) {
1479 fLimitAnalyzer.endEntity(XMLSecurityManager.Limit.GENERAL_ENTITY_SIZE_LIMIT, fCurrentEntity.name);
1480 if (fCurrentEntity.name.equals("[xml]")) {
1481 fSecurityManager.debugPrint(fLimitAnalyzer);
1482 }
1483 }
1484 fCurrentEntity.close();
1485 }catch(IOException ex){
1486 throw new XNIException(ex);
1487 }
1488 }
1489
1490 if (fEntityHandler != null) {
1491 //so this is the last opened entity, signal it to current fEntityHandler using Augmentation
1492 if(entity == null){
1493 fEntityAugs.removeAllItems();
1494 fEntityAugs.putItem(Constants.LAST_ENTITY, Boolean.TRUE);
1495 fEntityHandler.endEntity(fCurrentEntity.name, fEntityAugs);
1496 fEntityAugs.removeAllItems();
1497 }else{
1498 fEntityHandler.endEntity(fCurrentEntity.name, null);
1499 }
1500 }
1501 //check if it is a document entity
1502 boolean documentEntity = fCurrentEntity.name == XMLEntity;
1503
1504 //set popped entity as current entity
1505 fCurrentEntity = entity;
1506 fEntityScanner.setCurrentEntity(fCurrentEntity);
1507
1508 //check if there are any entity left in the stack -- if there are
1509 //no entries EOF has been reached.
1510 // throw exception when it is the last entity but it is not a document entity
1511
1512 if(fCurrentEntity == null & !documentEntity){
1513 throw new EOFException() ;
1514 }
1515
1516 if (DEBUG_BUFFER) {
1517 System.out.print(")endEntity: ");
1518 print();
1519 System.out.println();
1520 }
1521
1522 } // endEntity()
1523
1524
1525 //
1526 // XMLComponent methods
1527 //
1528 public void reset(PropertyManager propertyManager){
1529 // xerces properties
1530 fSymbolTable = (SymbolTable)propertyManager.getProperty(Constants.XERCES_PROPERTY_PREFIX + Constants.SYMBOL_TABLE_PROPERTY);
1531 fErrorReporter = (XMLErrorReporter)propertyManager.getProperty(Constants.XERCES_PROPERTY_PREFIX + Constants.ERROR_REPORTER_PROPERTY);
1532 try {
1533 fStaxEntityResolver = (StaxEntityResolverWrapper)propertyManager.getProperty(STAX_ENTITY_RESOLVER);
1534 } catch (XMLConfigurationException e) {
1535 fStaxEntityResolver = null;
1536 }
1537
1538 fSupportDTD = ((Boolean)propertyManager.getProperty(XMLInputFactory.SUPPORT_DTD));
1539 fReplaceEntityReferences = ((Boolean)propertyManager.getProperty(XMLInputFactory.IS_REPLACING_ENTITY_REFERENCES));
1540 fSupportExternalEntities = ((Boolean)propertyManager.getProperty(XMLInputFactory.IS_SUPPORTING_EXTERNAL_ENTITIES));
1541
1542 // Zephyr feature ignore-external-dtd is the opposite of Xerces' load-external-dtd
1543 fLoadExternalDTD = !((Boolean)propertyManager.getProperty(Constants.ZEPHYR_PROPERTY_PREFIX + Constants.IGNORE_EXTERNAL_DTD));
1544
1545 //Use Catalog
1546 fUseCatalog = (Boolean)propertyManager.getProperty(XMLConstants.USE_CATALOG);
1547 fCatalogFile = (String)propertyManager.getProperty(JdkXmlUtils.CATALOG_FILES);
1548 fDefer = (String)propertyManager.getProperty(JdkXmlUtils.CATALOG_DEFER);
1549 fPrefer = (String)propertyManager.getProperty(JdkXmlUtils.CATALOG_PREFER);
1550 fResolve = (String)propertyManager.getProperty(JdkXmlUtils.CATALOG_RESOLVE);
1551
1552 // JAXP 1.5 feature
1553 XMLSecurityPropertyManager spm = (XMLSecurityPropertyManager) propertyManager.getProperty(XML_SECURITY_PROPERTY_MANAGER);
1554 fAccessExternalDTD = spm.getValue(XMLSecurityPropertyManager.Property.ACCESS_EXTERNAL_DTD);
1555
1556 fSecurityManager = (XMLSecurityManager)propertyManager.getProperty(SECURITY_MANAGER);
1557
1558 fLimitAnalyzer = new XMLLimitAnalyzer();
1559 //reset fEntityStorage
1560 fEntityStorage.reset(propertyManager);
1561 //reset XMLEntityReaderImpl
1562 fEntityScanner.reset(propertyManager);
1563
1564 // initialize state
1565 //fStandalone = false;
1566 fEntities.clear();
1567 fEntityStack.removeAllElements();
1568 fCurrentEntity = null;
1569 fValidation = false;
1570 fExternalGeneralEntities = true;
1571 fExternalParameterEntities = true;
1572 fAllowJavaEncodings = true ;
1573 }
1574
1575 /**
1576 * Resets the component. The component can query the component manager
1577 * about any features and properties that affect the operation of the
1578 * component.
1579 *
1580 * @param componentManager The component manager.
1581 *
1582 * @throws SAXException Thrown by component on initialization error.
1583 * For example, if a feature or property is
1584 * required for the operation of the component, the
1585 * component manager may throw a
1586 * SAXNotRecognizedException or a
1587 * SAXNotSupportedException.
1588 */
1589 public void reset(XMLComponentManager componentManager)
1590 throws XMLConfigurationException {
1591
1592 boolean parser_settings = componentManager.getFeature(PARSER_SETTINGS, true);
1593
1594 if (!parser_settings) {
1595 // parser settings have not been changed
1596 reset();
1597 if(fEntityScanner != null){
1598 fEntityScanner.reset(componentManager);
1599 }
1600 if(fEntityStorage != null){
1601 fEntityStorage.reset(componentManager);
1602 }
1603 return;
1604 }
1605
1606 // sax features
1607 fValidation = componentManager.getFeature(VALIDATION, false);
1608 fExternalGeneralEntities = componentManager.getFeature(EXTERNAL_GENERAL_ENTITIES, true);
1609 fExternalParameterEntities = componentManager.getFeature(EXTERNAL_PARAMETER_ENTITIES, true);
1610
1611 // xerces features
1612 fAllowJavaEncodings = componentManager.getFeature(ALLOW_JAVA_ENCODINGS, false);
1613 fWarnDuplicateEntityDef = componentManager.getFeature(WARN_ON_DUPLICATE_ENTITYDEF, false);
1614 fStrictURI = componentManager.getFeature(STANDARD_URI_CONFORMANT, false);
1615 fLoadExternalDTD = componentManager.getFeature(LOAD_EXTERNAL_DTD, true);
1616
1617 // xerces properties
1618 fSymbolTable = (SymbolTable)componentManager.getProperty(SYMBOL_TABLE);
1619 fErrorReporter = (XMLErrorReporter)componentManager.getProperty(ERROR_REPORTER);
1620 fEntityResolver = (XMLEntityResolver)componentManager.getProperty(ENTITY_RESOLVER, null);
1621 fStaxEntityResolver = (StaxEntityResolverWrapper)componentManager.getProperty(STAX_ENTITY_RESOLVER, null);
1622 fValidationManager = (ValidationManager)componentManager.getProperty(VALIDATION_MANAGER, null);
1623 fSecurityManager = (XMLSecurityManager)componentManager.getProperty(SECURITY_MANAGER, null);
1624 entityExpansionIndex = fSecurityManager.getIndex(Constants.JDK_ENTITY_EXPANSION_LIMIT);
1625
1626 //StAX Property
1627 fSupportDTD = true;
1628 fReplaceEntityReferences = true;
1629 fSupportExternalEntities = true;
1630
1631 // JAXP 1.5 feature
1632 XMLSecurityPropertyManager spm = (XMLSecurityPropertyManager) componentManager.getProperty(XML_SECURITY_PROPERTY_MANAGER, null);
1633 if (spm == null) {
1634 spm = new XMLSecurityPropertyManager();
1635 }
1636 fAccessExternalDTD = spm.getValue(XMLSecurityPropertyManager.Property.ACCESS_EXTERNAL_DTD);
1637
1638 //Use Catalog
1639 fUseCatalog = componentManager.getFeature(XMLConstants.USE_CATALOG, true);
1640 fCatalogFile = (String)componentManager.getProperty(JdkXmlUtils.CATALOG_FILES);
1641 fDefer = (String)componentManager.getProperty(JdkXmlUtils.CATALOG_DEFER);
1642 fPrefer = (String)componentManager.getProperty(JdkXmlUtils.CATALOG_PREFER);
1643 fResolve = (String)componentManager.getProperty(JdkXmlUtils.CATALOG_RESOLVE);
1644
1645 //reset general state
1646 reset();
1647
1648 fEntityScanner.reset(componentManager);
1649 fEntityStorage.reset(componentManager);
1650
1651 } // reset(XMLComponentManager)
1652
1653 // reset general state. Should not be called other than by
1654 // a class acting as a component manager but not
1655 // implementing that interface for whatever reason.
1656 public void reset() {
1657 fLimitAnalyzer = new XMLLimitAnalyzer();
1658 // initialize state
1659 fStandalone = false;
1660 fEntities.clear();
1661 fEntityStack.removeAllElements();
1662 fEntityExpansionCount = 0;
1663
1664 fCurrentEntity = null;
1665 // reset scanner
1666 if(fXML10EntityScanner != null){
1667 fXML10EntityScanner.reset(fSymbolTable, this, fErrorReporter);
1668 }
1669 if(fXML11EntityScanner != null) {
1670 fXML11EntityScanner.reset(fSymbolTable, this, fErrorReporter);
1671 }
1672
1673 // DEBUG
1674 if (DEBUG_ENTITIES) {
1675 addInternalEntity("text", "Hello, World.");
1676 addInternalEntity("empty-element", "<foo/>");
1677 addInternalEntity("balanced-element", "<foo></foo>");
1678 addInternalEntity("balanced-element-with-text", "<foo>Hello, World</foo>");
1679 addInternalEntity("balanced-element-with-entity", "<foo>&text;</foo>");
1680 addInternalEntity("unbalanced-entity", "<foo>");
1681 addInternalEntity("recursive-entity", "<foo>&recursive-entity2;</foo>");
1682 addInternalEntity("recursive-entity2", "<bar>&recursive-entity3;</bar>");
1683 addInternalEntity("recursive-entity3", "<baz>&recursive-entity;</baz>");
1684 try {
1685 addExternalEntity("external-text", null, "external-text.ent", "test/external-text.xml");
1686 addExternalEntity("external-balanced-element", null, "external-balanced-element.ent", "test/external-balanced-element.xml");
1687 addExternalEntity("one", null, "ent/one.ent", "test/external-entity.xml");
1688 addExternalEntity("two", null, "ent/two.ent", "test/ent/one.xml");
1689 }
1690 catch (IOException ex) {
1691 // should never happen
1692 }
1693 }
1694
1695 fEntityHandler = null;
1696
1697 // reset scanner
1698 //if(fEntityScanner!=null)
1699 // fEntityScanner.reset(fSymbolTable, this,fErrorReporter);
1700
1701 }
1702 /**
1703 * Returns a list of feature identifiers that are recognized by
1704 * this component. This method may return null if no features
1705 * are recognized by this component.
1706 */
1707 public String[] getRecognizedFeatures() {
1708 return RECOGNIZED_FEATURES.clone();
1709 } // getRecognizedFeatures():String[]
1710
1711 /**
1712 * Sets the state of a feature. This method is called by the component
1713 * manager any time after reset when a feature changes state.
1714 * <p>
1715 * <strong>Note:</strong> Components should silently ignore features
1716 * that do not affect the operation of the component.
1717 *
1718 * @param featureId The feature identifier.
1719 * @param state The state of the feature.
1720 *
1721 * @throws SAXNotRecognizedException The component should not throw
1722 * this exception.
1723 * @throws SAXNotSupportedException The component should not throw
1724 * this exception.
1725 */
1726 public void setFeature(String featureId, boolean state)
1727 throws XMLConfigurationException {
1728
1729 // xerces features
1730 if (featureId.startsWith(Constants.XERCES_FEATURE_PREFIX)) {
1731 final int suffixLength = featureId.length() - Constants.XERCES_FEATURE_PREFIX.length();
1732 if (suffixLength == Constants.ALLOW_JAVA_ENCODINGS_FEATURE.length() &&
1733 featureId.endsWith(Constants.ALLOW_JAVA_ENCODINGS_FEATURE)) {
1734 fAllowJavaEncodings = state;
1735 }
1736 if (suffixLength == Constants.LOAD_EXTERNAL_DTD_FEATURE.length() &&
1737 featureId.endsWith(Constants.LOAD_EXTERNAL_DTD_FEATURE)) {
1738 fLoadExternalDTD = state;
1739 return;
1740 }
1741 } else if (featureId.equals(XMLConstants.USE_CATALOG)) {
1742 fUseCatalog = state;
1743 }
1744
1745 } // setFeature(String,boolean)
1746
1747 /**
1748 * Sets the value of a property. This method is called by the component
1749 * manager any time after reset when a property changes value.
1750 * <p>
1751 * <strong>Note:</strong> Components should silently ignore properties
1752 * that do not affect the operation of the component.
1753 *
1754 * @param propertyId The property identifier.
1755 * @param value The value of the property.
1756 *
1757 * @throws SAXNotRecognizedException The component should not throw
1758 * this exception.
1759 * @throws SAXNotSupportedException The component should not throw
1760 * this exception.
1761 */
1762 public void setProperty(String propertyId, Object value){
1763 // Xerces properties
1764 if (propertyId.startsWith(Constants.XERCES_PROPERTY_PREFIX)) {
1765 final int suffixLength = propertyId.length() - Constants.XERCES_PROPERTY_PREFIX.length();
1766
1767 if (suffixLength == Constants.SYMBOL_TABLE_PROPERTY.length() &&
1768 propertyId.endsWith(Constants.SYMBOL_TABLE_PROPERTY)) {
1769 fSymbolTable = (SymbolTable)value;
1770 return;
1771 }
1772 if (suffixLength == Constants.ERROR_REPORTER_PROPERTY.length() &&
1773 propertyId.endsWith(Constants.ERROR_REPORTER_PROPERTY)) {
1774 fErrorReporter = (XMLErrorReporter)value;
1775 return;
1776 }
1777 if (suffixLength == Constants.ENTITY_RESOLVER_PROPERTY.length() &&
1778 propertyId.endsWith(Constants.ENTITY_RESOLVER_PROPERTY)) {
1779 fEntityResolver = (XMLEntityResolver)value;
1780 return;
1781 }
1782 if (suffixLength == Constants.BUFFER_SIZE_PROPERTY.length() &&
1783 propertyId.endsWith(Constants.BUFFER_SIZE_PROPERTY)) {
1784 Integer bufferSize = (Integer)value;
1785 if (bufferSize != null &&
1786 bufferSize.intValue() > DEFAULT_XMLDECL_BUFFER_SIZE) {
1787 fBufferSize = bufferSize.intValue();
1788 fEntityScanner.setBufferSize(fBufferSize);
1789 fBufferPool.setExternalBufferSize(fBufferSize);
1790 }
1791 }
1792 if (suffixLength == Constants.SECURITY_MANAGER_PROPERTY.length() &&
1793 propertyId.endsWith(Constants.SECURITY_MANAGER_PROPERTY)) {
1794 fSecurityManager = (XMLSecurityManager)value;
1795 }
1796 }
1797
1798 //JAXP 1.5 properties
1799 if (propertyId.equals(XML_SECURITY_PROPERTY_MANAGER))
1800 {
1801 XMLSecurityPropertyManager spm = (XMLSecurityPropertyManager)value;
1802 fAccessExternalDTD = spm.getValue(XMLSecurityPropertyManager.Property.ACCESS_EXTERNAL_DTD);
1803 return;
1804 }
1805
1806 //Catalog properties
1807 if (propertyId.equals(JdkXmlUtils.CATALOG_FILES)) {
1808 fCatalogFile = (String)value;
1809 } else if (propertyId.equals(JdkXmlUtils.CATALOG_DEFER)) {
1810 fDefer = (String)value;
1811 } else if (propertyId.equals(JdkXmlUtils.CATALOG_PREFER)) {
1812 fPrefer = (String)value;
1813 } else if (propertyId.equals(JdkXmlUtils.CATALOG_RESOLVE)) {
1814 fResolve = (String)value;
1815 }
1816 }
1817
1818 public void setLimitAnalyzer(XMLLimitAnalyzer fLimitAnalyzer) {
1819 this.fLimitAnalyzer = fLimitAnalyzer;
1820 }
1821
1822 /**
1823 * Returns a list of property identifiers that are recognized by
1824 * this component. This method may return null if no properties
1825 * are recognized by this component.
1826 */
1827 public String[] getRecognizedProperties() {
1828 return RECOGNIZED_PROPERTIES.clone();
1829 } // getRecognizedProperties():String[]
1830 /**
1831 * Returns the default state for a feature, or null if this
1832 * component does not want to report a default value for this
1833 * feature.
1834 *
1835 * @param featureId The feature identifier.
1836 *
1837 * @since Xerces 2.2.0
1838 */
1839 public Boolean getFeatureDefault(String featureId) {
1840 for (int i = 0; i < RECOGNIZED_FEATURES.length; i++) {
1841 if (RECOGNIZED_FEATURES[i].equals(featureId)) {
1842 return FEATURE_DEFAULTS[i];
1843 }
1844 }
1845 return null;
1846 } // getFeatureDefault(String):Boolean
1847
1848 /**
1849 * Returns the default state for a property, or null if this
1850 * component does not want to report a default value for this
1851 * property.
1852 *
1853 * @param propertyId The property identifier.
1854 *
1855 * @since Xerces 2.2.0
1856 */
1857 public Object getPropertyDefault(String propertyId) {
1858 for (int i = 0; i < RECOGNIZED_PROPERTIES.length; i++) {
1859 if (RECOGNIZED_PROPERTIES[i].equals(propertyId)) {
1860 return PROPERTY_DEFAULTS[i];
1861 }
1862 }
1863 return null;
1864 } // getPropertyDefault(String):Object
1865
1866 //
1867 // Public static methods
1868 //
1869
1870 /**
1871 * Expands a system id and returns the system id as a URI, if
1872 * it can be expanded. A return value of null means that the
1873 * identifier is already expanded. An exception thrown
1874 * indicates a failure to expand the id.
1875 *
1876 * @param systemId The systemId to be expanded.
1877 *
1878 * @return Returns the URI string representing the expanded system
1879 * identifier. A null value indicates that the given
1880 * system identifier is already expanded.
1881 *
1882 */
1883 public static String expandSystemId(String systemId) {
1884 return expandSystemId(systemId, null);
1885 } // expandSystemId(String):String
1886
1887 //
1888 // Public static methods
1889 //
1890
1891 // current value of the "user.dir" property
1892 private static String gUserDir;
1893 // cached URI object for the current value of the escaped "user.dir" property stored as a URI
1894 private static URI gUserDirURI;
1895 // which ASCII characters need to be escaped
1896 private static boolean gNeedEscaping[] = new boolean[128];
1897 // the first hex character if a character needs to be escaped
1898 private static char gAfterEscaping1[] = new char[128];
1899 // the second hex character if a character needs to be escaped
1900 private static char gAfterEscaping2[] = new char[128];
1901 private static char[] gHexChs = {'0', '1', '2', '3', '4', '5', '6', '7',
1902 '8', '9', 'A', 'B', 'C', 'D', 'E', 'F'};
1903 // initialize the above 3 arrays
1904 static {
1905 for (int i = 0; i <= 0x1f; i++) {
1906 gNeedEscaping[i] = true;
1907 gAfterEscaping1[i] = gHexChs[i >> 4];
1908 gAfterEscaping2[i] = gHexChs[i & 0xf];
1909 }
1910 gNeedEscaping[0x7f] = true;
1911 gAfterEscaping1[0x7f] = '7';
1912 gAfterEscaping2[0x7f] = 'F';
1913 char[] escChs = {' ', '<', '>', '#', '%', '"', '{', '}',
1914 '|', '\\', '^', '~', '[', ']', '`'};
1915 int len = escChs.length;
1916 char ch;
1917 for (int i = 0; i < len; i++) {
1918 ch = escChs[i];
1919 gNeedEscaping[ch] = true;
1920 gAfterEscaping1[ch] = gHexChs[ch >> 4];
1921 gAfterEscaping2[ch] = gHexChs[ch & 0xf];
1922 }
1923 }
1924
1925 // To escape the "user.dir" system property, by using %HH to represent
1926 // special ASCII characters: 0x00~0x1F, 0x7F, ' ', '<', '>', '#', '%'
1927 // and '"'. It's a static method, so needs to be synchronized.
1928 // this method looks heavy, but since the system property isn't expected
1929 // to change often, so in most cases, we only need to return the URI
1930 // that was escaped before.
1931 // According to the URI spec, non-ASCII characters (whose value >= 128)
1932 // need to be escaped too.
1933 // REVISIT: don't know how to escape non-ASCII characters, especially
1934 // which encoding to use. Leave them for now.
1935 private static synchronized URI getUserDir() throws URI.MalformedURIException {
1936 // get the user.dir property
1937 String userDir = "";
1938 try {
1939 userDir = SecuritySupport.getSystemProperty("user.dir");
1940 }
1941 catch (SecurityException se) {
1942 }
1943
1944 // return empty string if property value is empty string.
1945 if (userDir.length() == 0)
1946 return new URI("file", "", "", null, null);
1947 // compute the new escaped value if the new property value doesn't
1948 // match the previous one
1949 if (gUserDirURI != null && userDir.equals(gUserDir)) {
1950 return gUserDirURI;
1951 }
1952
1953 // record the new value as the global property value
1954 gUserDir = userDir;
1955
1956 char separator = java.io.File.separatorChar;
1957 userDir = userDir.replace(separator, '/');
1958
1959 int len = userDir.length(), ch;
1960 StringBuilder buffer = new StringBuilder(len*3);
1961 // change C:/blah to /C:/blah
1962 if (len >= 2 && userDir.charAt(1) == ':') {
1963 ch = Character.toUpperCase(userDir.charAt(0));
1964 if (ch >= 'A' && ch <= 'Z') {
1965 buffer.append('/');
1966 }
1967 }
1968
1969 // for each character in the path
1970 int i = 0;
1971 for (; i < len; i++) {
1972 ch = userDir.charAt(i);
1973 // if it's not an ASCII character, break here, and use UTF-8 encoding
1974 if (ch >= 128)
1975 break;
1976 if (gNeedEscaping[ch]) {
1977 buffer.append('%');
1978 buffer.append(gAfterEscaping1[ch]);
1979 buffer.append(gAfterEscaping2[ch]);
1980 // record the fact that it's escaped
1981 }
1982 else {
1983 buffer.append((char)ch);
1984 }
1985 }
1986
1987 // we saw some non-ascii character
1988 if (i < len) {
1989 // get UTF-8 bytes for the remaining sub-string
1990 byte[] bytes = null;
1991 byte b;
1992 try {
1993 bytes = userDir.substring(i).getBytes("UTF-8");
1994 } catch (java.io.UnsupportedEncodingException e) {
1995 // should never happen
1996 return new URI("file", "", userDir, null, null);
1997 }
1998 len = bytes.length;
1999
2000 // for each byte
2001 for (i = 0; i < len; i++) {
2002 b = bytes[i];
2003 // for non-ascii character: make it positive, then escape
2004 if (b < 0) {
2005 ch = b + 256;
2006 buffer.append('%');
2007 buffer.append(gHexChs[ch >> 4]);
2008 buffer.append(gHexChs[ch & 0xf]);
2009 }
2010 else if (gNeedEscaping[b]) {
2011 buffer.append('%');
2012 buffer.append(gAfterEscaping1[b]);
2013 buffer.append(gAfterEscaping2[b]);
2014 }
2015 else {
2016 buffer.append((char)b);
2017 }
2018 }
2019 }
2020
2021 // change blah/blah to blah/blah/
2022 if (!userDir.endsWith("/"))
2023 buffer.append('/');
2024
2025 gUserDirURI = new URI("file", "", buffer.toString(), null, null);
2026
2027 return gUserDirURI;
2028 }
2029
2030 public static OutputStream createOutputStream(String uri) throws IOException {
2031 // URI was specified. Handle relative URIs.
2032 final String expanded = XMLEntityManager.expandSystemId(uri, null, true);
2033 final URL url = new URL(expanded != null ? expanded : uri);
2034 OutputStream out = null;
2035 String protocol = url.getProtocol();
2036 String host = url.getHost();
2037 // Use FileOutputStream if this URI is for a local file.
2038 if (protocol.equals("file")
2039 && (host == null || host.length() == 0 || host.equals("localhost"))) {
2040 File file = new File(getPathWithoutEscapes(url.getPath()));
2041 if (!file.exists()) {
2042 File parent = file.getParentFile();
2043 if (parent != null && !parent.exists()) {
2044 parent.mkdirs();
2045 }
2046 }
2047 out = new FileOutputStream(file);
2048 }
2049 // Try to write to some other kind of URI. Some protocols
2050 // won't support this, though HTTP should work.
2051 else {
2052 URLConnection urlCon = url.openConnection();
2053 urlCon.setDoInput(false);
2054 urlCon.setDoOutput(true);
2055 urlCon.setUseCaches(false); // Enable tunneling.
2056 if (urlCon instanceof HttpURLConnection) {
2057 // The DOM L3 REC says if we are writing to an HTTP URI
2058 // it is to be done with an HTTP PUT.
2059 HttpURLConnection httpCon = (HttpURLConnection) urlCon;
2060 httpCon.setRequestMethod("PUT");
2061 }
2062 out = urlCon.getOutputStream();
2063 }
2064 return out;
2065 }
2066
2067 private static String getPathWithoutEscapes(String origPath) {
2068 if (origPath != null && origPath.length() != 0 && origPath.indexOf('%') != -1) {
2069 // Locate the escape characters
2070 StringTokenizer tokenizer = new StringTokenizer(origPath, "%");
2071 StringBuilder result = new StringBuilder(origPath.length());
2072 int size = tokenizer.countTokens();
2073 result.append(tokenizer.nextToken());
2074 for(int i = 1; i < size; ++i) {
2075 String token = tokenizer.nextToken();
2076 // Decode the 2 digit hexadecimal number following % in '%nn'
2077 result.append((char)Integer.valueOf(token.substring(0, 2), 16).intValue());
2078 result.append(token.substring(2));
2079 }
2080 return result.toString();
2081 }
2082 return origPath;
2083 }
2084
2085 /**
2086 * Absolutizes a URI using the current value
2087 * of the "user.dir" property as the base URI. If
2088 * the URI is already absolute, this is a no-op.
2089 *
2090 * @param uri the URI to absolutize
2091 */
2092 public static void absolutizeAgainstUserDir(URI uri)
2093 throws URI.MalformedURIException {
2094 uri.absolutize(getUserDir());
2095 }
2096
2097 /**
2098 * Expands a system id and returns the system id as a URI, if
2099 * it can be expanded. A return value of null means that the
2100 * identifier is already expanded. An exception thrown
2101 * indicates a failure to expand the id.
2102 *
2103 * @param systemId The systemId to be expanded.
2104 *
2105 * @return Returns the URI string representing the expanded system
2106 * identifier. A null value indicates that the given
2107 * system identifier is already expanded.
2108 *
2109 */
2110 public static String expandSystemId(String systemId, String baseSystemId) {
2111
2112 // check for bad parameters id
2113 if (systemId == null || systemId.length() == 0) {
2114 return systemId;
2115 }
2116 // if id already expanded, return
2117 try {
2118 URI uri = new URI(systemId);
2119 if (uri != null) {
2120 return systemId;
2121 }
2122 } catch (URI.MalformedURIException e) {
2123 // continue on...
2124 }
2125 // normalize id
2126 String id = fixURI(systemId);
2127
2128 // normalize base
2129 URI base = null;
2130 URI uri = null;
2131 try {
2132 if (baseSystemId == null || baseSystemId.length() == 0 ||
2133 baseSystemId.equals(systemId)) {
2134 String dir = getUserDir().toString();
2135 base = new URI("file", "", dir, null, null);
2136 } else {
2137 try {
2138 base = new URI(fixURI(baseSystemId));
2139 } catch (URI.MalformedURIException e) {
2140 if (baseSystemId.indexOf(':') != -1) {
2141 // for xml schemas we might have baseURI with
2142 // a specified drive
2143 base = new URI("file", "", fixURI(baseSystemId), null, null);
2144 } else {
2145 String dir = getUserDir().toString();
2146 dir = dir + fixURI(baseSystemId);
2147 base = new URI("file", "", dir, null, null);
2148 }
2149 }
2150 }
2151 // expand id
2152 uri = new URI(base, id);
2153 } catch (Exception e) {
2154 // let it go through
2155
2156 }
2157
2158 if (uri == null) {
2159 return systemId;
2160 }
2161 return uri.toString();
2162
2163 } // expandSystemId(String,String):String
2164
2165 /**
2166 * Expands a system id and returns the system id as a URI, if
2167 * it can be expanded. A return value of null means that the
2168 * identifier is already expanded. An exception thrown
2169 * indicates a failure to expand the id.
2170 *
2171 * @param systemId The systemId to be expanded.
2172 *
2173 * @return Returns the URI string representing the expanded system
2174 * identifier. A null value indicates that the given
2175 * system identifier is already expanded.
2176 *
2177 */
2178 public static String expandSystemId(String systemId, String baseSystemId,
2179 boolean strict)
2180 throws URI.MalformedURIException {
2181
2182 // check if there is a system id before
2183 // trying to expand it.
2184 if (systemId == null) {
2185 return null;
2186 }
2187
2188 // system id has to be a valid URI
2189 if (strict) {
2190 try {
2191 // if it's already an absolute one, return it
2192 new URI(systemId);
2193 return systemId;
2194 }
2195 catch (URI.MalformedURIException ex) {
2196 }
2197 URI base = null;
2198 // if there isn't a base uri, use the working directory
2199 if (baseSystemId == null || baseSystemId.length() == 0) {
2200 base = new URI("file", "", getUserDir().toString(), null, null);
2201 }
2202 // otherwise, use the base uri
2203 else {
2204 try {
2205 base = new URI(baseSystemId);
2206 }
2207 catch (URI.MalformedURIException e) {
2208 // assume "base" is also a relative uri
2209 String dir = getUserDir().toString();
2210 dir = dir + baseSystemId;
2211 base = new URI("file", "", dir, null, null);
2212 }
2213 }
2214 // absolutize the system id using the base
2215 URI uri = new URI(base, systemId);
2216 // return the string rep of the new uri (an absolute one)
2217 return uri.toString();
2218
2219 // if any exception is thrown, it'll get thrown to the caller.
2220 }
2221
2222 // Assume the URIs are well-formed. If it turns out they're not, try fixing them up.
2223 try {
2224 return expandSystemIdStrictOff(systemId, baseSystemId);
2225 }
2226 catch (URI.MalformedURIException e) {
2227 /** Xerces URI rejects unicode, try java.net.URI
2228 * this is not ideal solution, but it covers known cases which either
2229 * Xerces URI or java.net.URI can handle alone
2230 * will file bug against java.net.URI
2231 */
2232 try {
2233 return expandSystemIdStrictOff1(systemId, baseSystemId);
2234 } catch (URISyntaxException ex) {
2235 // continue on...
2236 }
2237 }
2238 // check for bad parameters id
2239 if (systemId.length() == 0) {
2240 return systemId;
2241 }
2242
2243 // normalize id
2244 String id = fixURI(systemId);
2245
2246 // normalize base
2247 URI base = null;
2248 URI uri = null;
2249 try {
2250 if (baseSystemId == null || baseSystemId.length() == 0 ||
2251 baseSystemId.equals(systemId)) {
2252 base = getUserDir();
2253 }
2254 else {
2255 try {
2256 base = new URI(fixURI(baseSystemId).trim());
2257 }
2258 catch (URI.MalformedURIException e) {
2259 if (baseSystemId.indexOf(':') != -1) {
2260 // for xml schemas we might have baseURI with
2261 // a specified drive
2262 base = new URI("file", "", fixURI(baseSystemId).trim(), null, null);
2263 }
2264 else {
2265 base = new URI(getUserDir(), fixURI(baseSystemId));
2266 }
2267 }
2268 }
2269 // expand id
2270 uri = new URI(base, id.trim());
2271 }
2272 catch (Exception e) {
2273 // let it go through
2274
2275 }
2276
2277 if (uri == null) {
2278 return systemId;
2279 }
2280 return uri.toString();
2281
2282 } // expandSystemId(String,String,boolean):String
2283
2284 /**
2285 * Helper method for expandSystemId(String,String,boolean):String
2286 */
2287 private static String expandSystemIdStrictOn(String systemId, String baseSystemId)
2288 throws URI.MalformedURIException {
2289
2290 URI systemURI = new URI(systemId, true);
2291 // If it's already an absolute one, return it
2292 if (systemURI.isAbsoluteURI()) {
2293 return systemId;
2294 }
2295
2296 // If there isn't a base URI, use the working directory
2297 URI baseURI = null;
2298 if (baseSystemId == null || baseSystemId.length() == 0) {
2299 baseURI = getUserDir();
2300 }
2301 else {
2302 baseURI = new URI(baseSystemId, true);
2303 if (!baseURI.isAbsoluteURI()) {
2304 // assume "base" is also a relative uri
2305 baseURI.absolutize(getUserDir());
2306 }
2307 }
2308
2309 // absolutize the system identifier using the base URI
2310 systemURI.absolutize(baseURI);
2311
2312 // return the string rep of the new uri (an absolute one)
2313 return systemURI.toString();
2314
2315 // if any exception is thrown, it'll get thrown to the caller.
2316
2317 } // expandSystemIdStrictOn(String,String):String
2318
2319 /**
2320 * Helper method for expandSystemId(String,String,boolean):String
2321 */
2322 private static String expandSystemIdStrictOff(String systemId, String baseSystemId)
2323 throws URI.MalformedURIException {
2324
2325 URI systemURI = new URI(systemId, true);
2326 // If it's already an absolute one, return it
2327 if (systemURI.isAbsoluteURI()) {
2328 if (systemURI.getScheme().length() > 1) {
2329 return systemId;
2330 }
2331 /**
2332 * If the scheme's length is only one character,
2333 * it's likely that this was intended as a file
2334 * path. Fixing this up in expandSystemId to
2335 * maintain backwards compatibility.
2336 */
2337 throw new URI.MalformedURIException();
2338 }
2339
2340 // If there isn't a base URI, use the working directory
2341 URI baseURI = null;
2342 if (baseSystemId == null || baseSystemId.length() == 0) {
2343 baseURI = getUserDir();
2344 }
2345 else {
2346 baseURI = new URI(baseSystemId, true);
2347 if (!baseURI.isAbsoluteURI()) {
2348 // assume "base" is also a relative uri
2349 baseURI.absolutize(getUserDir());
2350 }
2351 }
2352
2353 // absolutize the system identifier using the base URI
2354 systemURI.absolutize(baseURI);
2355
2356 // return the string rep of the new uri (an absolute one)
2357 return systemURI.toString();
2358
2359 // if any exception is thrown, it'll get thrown to the caller.
2360
2361 } // expandSystemIdStrictOff(String,String):String
2362
2363 private static String expandSystemIdStrictOff1(String systemId, String baseSystemId)
2364 throws URISyntaxException, URI.MalformedURIException {
2365
2366 java.net.URI systemURI = new java.net.URI(systemId);
2367 // If it's already an absolute one, return it
2368 if (systemURI.isAbsolute()) {
2369 if (systemURI.getScheme().length() > 1) {
2370 return systemId;
2371 }
2372 /**
2373 * If the scheme's length is only one character,
2374 * it's likely that this was intended as a file
2375 * path. Fixing this up in expandSystemId to
2376 * maintain backwards compatibility.
2377 */
2378 throw new URISyntaxException(systemId, "the scheme's length is only one character");
2379 }
2380
2381 // If there isn't a base URI, use the working directory
2382 URI baseURI = null;
2383 if (baseSystemId == null || baseSystemId.length() == 0) {
2384 baseURI = getUserDir();
2385 }
2386 else {
2387 baseURI = new URI(baseSystemId, true);
2388 if (!baseURI.isAbsoluteURI()) {
2389 // assume "base" is also a relative uri
2390 baseURI.absolutize(getUserDir());
2391 }
2392 }
2393
2394 // absolutize the system identifier using the base URI
2395 // systemURI.absolutize(baseURI);
2396 systemURI = (new java.net.URI(baseURI.toString())).resolve(systemURI);
2397
2398 // return the string rep of the new uri (an absolute one)
2399 return systemURI.toString();
2400
2401 // if any exception is thrown, it'll get thrown to the caller.
2402
2403 } // expandSystemIdStrictOff(String,String):String
2404
2405 //
2406 // Protected methods
2407 //
2408
2409
2410 /**
2411 * Returns the IANA encoding name that is auto-detected from
2412 * the bytes specified, with the endian-ness of that encoding where appropriate.
2413 *
2414 * @param b4 The first four bytes of the input.
2415 * @param count The number of bytes actually read.
2416 * @return a 2-element array: the first element, an IANA-encoding string,
2417 * the second element a Boolean which is true iff the document is big endian, false
2418 * if it's little-endian, and null if the distinction isn't relevant.
2419 */
2420 protected Object[] getEncodingName(byte[] b4, int count) {
2421
2422 if (count < 2) {
2423 return defaultEncoding;
2424 }
2425
2426 // UTF-16, with BOM
2427 int b0 = b4[0] & 0xFF;
2428 int b1 = b4[1] & 0xFF;
2429 if (b0 == 0xFE && b1 == 0xFF) {
2430 // UTF-16, big-endian
2431 return new Object [] {"UTF-16BE", true};
2432 }
2433 if (b0 == 0xFF && b1 == 0xFE) {
2434 // UTF-16, little-endian
2435 return new Object [] {"UTF-16LE", false};
2436 }
2437
2438 // default to UTF-8 if we don't have enough bytes to make a
2439 // good determination of the encoding
2440 if (count < 3) {
2441 return defaultEncoding;
2442 }
2443
2444 // UTF-8 with a BOM
2445 int b2 = b4[2] & 0xFF;
2446 if (b0 == 0xEF && b1 == 0xBB && b2 == 0xBF) {
2447 return defaultEncoding;
2448 }
2449
2450 // default to UTF-8 if we don't have enough bytes to make a
2451 // good determination of the encoding
2452 if (count < 4) {
2453 return defaultEncoding;
2454 }
2455
2456 // other encodings
2457 int b3 = b4[3] & 0xFF;
2458 if (b0 == 0x00 && b1 == 0x00 && b2 == 0x00 && b3 == 0x3C) {
2459 // UCS-4, big endian (1234)
2460 return new Object [] {"ISO-10646-UCS-4", true};
2461 }
2462 if (b0 == 0x3C && b1 == 0x00 && b2 == 0x00 && b3 == 0x00) {
2463 // UCS-4, little endian (4321)
2464 return new Object [] {"ISO-10646-UCS-4", false};
2465 }
2466 if (b0 == 0x00 && b1 == 0x00 && b2 == 0x3C && b3 == 0x00) {
2467 // UCS-4, unusual octet order (2143)
2468 // REVISIT: What should this be?
2469 return new Object [] {"ISO-10646-UCS-4", null};
2470 }
2471 if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x00) {
2472 // UCS-4, unusual octect order (3412)
2473 // REVISIT: What should this be?
2474 return new Object [] {"ISO-10646-UCS-4", null};
2475 }
2476 if (b0 == 0x00 && b1 == 0x3C && b2 == 0x00 && b3 == 0x3F) {
2477 // UTF-16, big-endian, no BOM
2478 // (or could turn out to be UCS-2...
2479 // REVISIT: What should this be?
2480 return new Object [] {"UTF-16BE", true};
2481 }
2482 if (b0 == 0x3C && b1 == 0x00 && b2 == 0x3F && b3 == 0x00) {
2483 // UTF-16, little-endian, no BOM
2484 // (or could turn out to be UCS-2...
2485 return new Object [] {"UTF-16LE", false};
2486 }
2487 if (b0 == 0x4C && b1 == 0x6F && b2 == 0xA7 && b3 == 0x94) {
2488 // EBCDIC
2489 // a la xerces1, return CP037 instead of EBCDIC here
2490 return new Object [] {"CP037", null};
2491 }
2492
2493 return defaultEncoding;
2494
2495 } // getEncodingName(byte[],int):Object[]
2496
2497 /**
2498 * Creates a reader capable of reading the given input stream in
2499 * the specified encoding.
2500 *
2501 * @param inputStream The input stream.
2502 * @param encoding The encoding name that the input stream is
2503 * encoded using. If the user has specified that
2504 * Java encoding names are allowed, then the
2505 * encoding name may be a Java encoding name;
2506 * otherwise, it is an ianaEncoding name.
2507 * @param isBigEndian For encodings (like uCS-4), whose names cannot
2508 * specify a byte order, this tells whether the order is bigEndian. null menas
2509 * unknown or not relevant.
2510 *
2511 * @return Returns a reader.
2512 */
2513 protected Reader createReader(InputStream inputStream, String encoding, Boolean isBigEndian)
2514 throws IOException {
2515
2516 // normalize encoding name
2517 if (encoding == null) {
2518 encoding = "UTF-8";
2519 }
2520
2521 // try to use an optimized reader
2522 String ENCODING = encoding.toUpperCase(Locale.ENGLISH);
2523 if (ENCODING.equals("UTF-8")) {
2524 if (DEBUG_ENCODINGS) {
2525 System.out.println("$$$ creating UTF8Reader");
2526 }
2527 return new UTF8Reader(inputStream, fBufferSize, fErrorReporter.getMessageFormatter(XMLMessageFormatter.XML_DOMAIN), fErrorReporter.getLocale() );
2528 }
2529 if (ENCODING.equals("US-ASCII")) {
2530 if (DEBUG_ENCODINGS) {
2531 System.out.println("$$$ creating ASCIIReader");
2532 }
2533 return new ASCIIReader(inputStream, fBufferSize, fErrorReporter.getMessageFormatter(XMLMessageFormatter.XML_DOMAIN), fErrorReporter.getLocale());
2534 }
2535 if(ENCODING.equals("ISO-10646-UCS-4")) {
2536 if(isBigEndian != null) {
2537 boolean isBE = isBigEndian.booleanValue();
2538 if(isBE) {
2539 return new UCSReader(inputStream, UCSReader.UCS4BE);
2540 } else {
2541 return new UCSReader(inputStream, UCSReader.UCS4LE);
2542 }
2543 } else {
2544 fErrorReporter.reportError(this.getEntityScanner(),XMLMessageFormatter.XML_DOMAIN,
2545 "EncodingByteOrderUnsupported",
2546 new Object[] { encoding },
2547 XMLErrorReporter.SEVERITY_FATAL_ERROR);
2548 }
2549 }
2550 if(ENCODING.equals("ISO-10646-UCS-2")) {
2551 if(isBigEndian != null) { // sould never happen with this encoding...
2552 boolean isBE = isBigEndian.booleanValue();
2553 if(isBE) {
2554 return new UCSReader(inputStream, UCSReader.UCS2BE);
2555 } else {
2556 return new UCSReader(inputStream, UCSReader.UCS2LE);
2557 }
2558 } else {
2559 fErrorReporter.reportError(this.getEntityScanner(),XMLMessageFormatter.XML_DOMAIN,
2560 "EncodingByteOrderUnsupported",
2561 new Object[] { encoding },
2562 XMLErrorReporter.SEVERITY_FATAL_ERROR);
2563 }
2564 }
2565
2566 // check for valid name
2567 boolean validIANA = XMLChar.isValidIANAEncoding(encoding);
2568 boolean validJava = XMLChar.isValidJavaEncoding(encoding);
2569 if (!validIANA || (fAllowJavaEncodings && !validJava)) {
2570 fErrorReporter.reportError(this.getEntityScanner(),XMLMessageFormatter.XML_DOMAIN,
2571 "EncodingDeclInvalid",
2572 new Object[] { encoding },
2573 XMLErrorReporter.SEVERITY_FATAL_ERROR);
2574 // NOTE: AndyH suggested that, on failure, we use ISO Latin 1
2575 // because every byte is a valid ISO Latin 1 character.
2576 // It may not translate correctly but if we failed on
2577 // the encoding anyway, then we're expecting the content
2578 // of the document to be bad. This will just prevent an
2579 // invalid UTF-8 sequence to be detected. This is only
2580 // important when continue-after-fatal-error is turned
2581 // on. -Ac
2582 encoding = "ISO-8859-1";
2583 }
2584
2585 // try to use a Java reader
2586 String javaEncoding = EncodingMap.getIANA2JavaMapping(ENCODING);
2587 if (javaEncoding == null) {
2588 if(fAllowJavaEncodings) {
2589 javaEncoding = encoding;
2590 } else {
2591 fErrorReporter.reportError(this.getEntityScanner(),XMLMessageFormatter.XML_DOMAIN,
2592 "EncodingDeclInvalid",
2593 new Object[] { encoding },
2594 XMLErrorReporter.SEVERITY_FATAL_ERROR);
2595 // see comment above.
2596 javaEncoding = "ISO8859_1";
2597 }
2598 }
2599 if (DEBUG_ENCODINGS) {
2600 System.out.print("$$$ creating Java InputStreamReader: encoding="+javaEncoding);
2601 if (javaEncoding == encoding) {
2602 System.out.print(" (IANA encoding)");
2603 }
2604 System.out.println();
2605 }
2606 return new BufferedReader( new InputStreamReader(inputStream, javaEncoding));
2607
2608 } // createReader(InputStream,String, Boolean): Reader
2609
2610
2611 /**
2612 * Return the public identifier for the current document event.
2613 * <p>
2614 * The return value is the public identifier of the document
2615 * entity or of the external parsed entity in which the markup
2616 * triggering the event appears.
2617 *
2618 * @return A string containing the public identifier, or
2619 * null if none is available.
2620 */
2621 public String getPublicId() {
2622 return (fCurrentEntity != null && fCurrentEntity.entityLocation != null) ? fCurrentEntity.entityLocation.getPublicId() : null;
2623 } // getPublicId():String
2624
2625 /**
2626 * Return the expanded system identifier for the current document event.
2627 * <p>
2628 * The return value is the expanded system identifier of the document
2629 * entity or of the external parsed entity in which the markup
2630 * triggering the event appears.
2631 * <p>
2632 * If the system identifier is a URL, the parser must resolve it
2633 * fully before passing it to the application.
2634 *
2635 * @return A string containing the expanded system identifier, or null
2636 * if none is available.
2637 */
2638 public String getExpandedSystemId() {
2639 if (fCurrentEntity != null) {
2640 if (fCurrentEntity.entityLocation != null &&
2641 fCurrentEntity.entityLocation.getExpandedSystemId() != null ) {
2642 return fCurrentEntity.entityLocation.getExpandedSystemId();
2643 } else {
2644 // search for the first external entity on the stack
2645 int size = fEntityStack.size();
2646 for (int i = size - 1; i >= 0 ; i--) {
2647 Entity.ScannedEntity externalEntity =
2648 (Entity.ScannedEntity)fEntityStack.get(i);
2649
2650 if (externalEntity.entityLocation != null &&
2651 externalEntity.entityLocation.getExpandedSystemId() != null) {
2652 return externalEntity.entityLocation.getExpandedSystemId();
2653 }
2654 }
2655 }
2656 }
2657 return null;
2658 } // getExpandedSystemId():String
2659
2660 /**
2661 * Return the literal system identifier for the current document event.
2662 * <p>
2663 * The return value is the literal system identifier of the document
2664 * entity or of the external parsed entity in which the markup
2665 * triggering the event appears.
2666 * <p>
2667 * @return A string containing the literal system identifier, or null
2668 * if none is available.
2669 */
2670 public String getLiteralSystemId() {
2671 if (fCurrentEntity != null) {
2672 if (fCurrentEntity.entityLocation != null &&
2673 fCurrentEntity.entityLocation.getLiteralSystemId() != null ) {
2674 return fCurrentEntity.entityLocation.getLiteralSystemId();
2675 } else {
2676 // search for the first external entity on the stack
2677 int size = fEntityStack.size();
2678 for (int i = size - 1; i >= 0 ; i--) {
2679 Entity.ScannedEntity externalEntity =
2680 (Entity.ScannedEntity)fEntityStack.get(i);
2681
2682 if (externalEntity.entityLocation != null &&
2683 externalEntity.entityLocation.getLiteralSystemId() != null) {
2684 return externalEntity.entityLocation.getLiteralSystemId();
2685 }
2686 }
2687 }
2688 }
2689 return null;
2690 } // getLiteralSystemId():String
2691
2692 /**
2693 * Return the line number where the current document event ends.
2694 * <p>
2695 * <strong>Warning:</strong> The return value from the method
2696 * is intended only as an approximation for the sake of error
2697 * reporting; it is not intended to provide sufficient information
2698 * to edit the character content of the original XML document.
2699 * <p>
2700 * The return value is an approximation of the line number
2701 * in the document entity or external parsed entity where the
2702 * markup triggering the event appears.
2703 * <p>
2704 * If possible, the SAX driver should provide the line position
2705 * of the first character after the text associated with the document
2706 * event. The first line in the document is line 1.
2707 *
2708 * @return The line number, or -1 if none is available.
2709 */
2710 public int getLineNumber() {
2711 if (fCurrentEntity != null) {
2712 if (fCurrentEntity.isExternal()) {
2713 return fCurrentEntity.lineNumber;
2714 } else {
2715 // search for the first external entity on the stack
2716 int size = fEntityStack.size();
2717 for (int i=size-1; i>0 ; i--) {
2718 Entity.ScannedEntity firstExternalEntity = (Entity.ScannedEntity)fEntityStack.get(i);
2719 if (firstExternalEntity.isExternal()) {
2720 return firstExternalEntity.lineNumber;
2721 }
2722 }
2723 }
2724 }
2725
2726 return -1;
2727
2728 } // getLineNumber():int
2729
2730 /**
2731 * Return the column number where the current document event ends.
2732 * <p>
2733 * <strong>Warning:</strong> The return value from the method
2734 * is intended only as an approximation for the sake of error
2735 * reporting; it is not intended to provide sufficient information
2736 * to edit the character content of the original XML document.
2737 * <p>
2738 * The return value is an approximation of the column number
2739 * in the document entity or external parsed entity where the
2740 * markup triggering the event appears.
2741 * <p>
2742 * If possible, the SAX driver should provide the line position
2743 * of the first character after the text associated with the document
2744 * event.
2745 * <p>
2746 * If possible, the SAX driver should provide the line position
2747 * of the first character after the text associated with the document
2748 * event. The first column in each line is column 1.
2749 *
2750 * @return The column number, or -1 if none is available.
2751 */
2752 public int getColumnNumber() {
2753 if (fCurrentEntity != null) {
2754 if (fCurrentEntity.isExternal()) {
2755 return fCurrentEntity.columnNumber;
2756 } else {
2757 // search for the first external entity on the stack
2758 int size = fEntityStack.size();
2759 for (int i=size-1; i>0 ; i--) {
2760 Entity.ScannedEntity firstExternalEntity = (Entity.ScannedEntity)fEntityStack.get(i);
2761 if (firstExternalEntity.isExternal()) {
2762 return firstExternalEntity.columnNumber;
2763 }
2764 }
2765 }
2766 }
2767
2768 return -1;
2769 } // getColumnNumber():int
2770
2771
2772 //
2773 // Protected static methods
2774 //
2775
2776 /**
2777 * Fixes a platform dependent filename to standard URI form.
2778 *
2779 * @param str The string to fix.
2780 *
2781 * @return Returns the fixed URI string.
2782 */
2783 protected static String fixURI(String str) {
2784
2785 // handle platform dependent strings
2786 str = str.replace(java.io.File.separatorChar, '/');
2787
2788 // Windows fix
2789 if (str.length() >= 2) {
2790 char ch1 = str.charAt(1);
2791 // change "C:blah" to "/C:blah"
2792 if (ch1 == ':') {
2793 char ch0 = Character.toUpperCase(str.charAt(0));
2794 if (ch0 >= 'A' && ch0 <= 'Z') {
2795 str = "/" + str;
2796 }
2797 }
2798 // change "//blah" to "file://blah"
2799 else if (ch1 == '/' && str.charAt(0) == '/') {
2800 str = "file:" + str;
2801 }
2802 }
2803
2804 // replace spaces in file names with %20.
2805 // Original comment from JDK5: the following algorithm might not be
2806 // very performant, but people who want to use invalid URI's have to
2807 // pay the price.
2808 int pos = str.indexOf(' ');
2809 if (pos >= 0) {
2810 StringBuilder sb = new StringBuilder(str.length());
2811 // put characters before ' ' into the string builder
2812 for (int i = 0; i < pos; i++)
2813 sb.append(str.charAt(i));
2814 // and %20 for the space
2815 sb.append("%20");
2816 // for the remamining part, also convert ' ' to "%20".
2817 for (int i = pos+1; i < str.length(); i++) {
2818 if (str.charAt(i) == ' ')
2819 sb.append("%20");
2820 else
2821 sb.append(str.charAt(i));
2822 }
2823 str = sb.toString();
2824 }
2825
2826 // done
2827 return str;
2828
2829 } // fixURI(String):String
2830
2831
2832 //
2833 // Package visible methods
2834 //
2835 /** Prints the contents of the buffer. */
2836 final void print() {
2837 if (DEBUG_BUFFER) {
2838 if (fCurrentEntity != null) {
2839 System.out.print('[');
2840 System.out.print(fCurrentEntity.count);
2841 System.out.print(' ');
2842 System.out.print(fCurrentEntity.position);
2843 if (fCurrentEntity.count > 0) {
2844 System.out.print(" \"");
2845 for (int i = 0; i < fCurrentEntity.count; i++) {
2846 if (i == fCurrentEntity.position) {
2847 System.out.print('^');
2848 }
2849 char c = fCurrentEntity.ch[i];
2850 switch (c) {
2851 case '\n': {
2852 System.out.print("\\n");
2853 break;
2854 }
2855 case '\r': {
2856 System.out.print("\\r");
2857 break;
2858 }
2859 case '\t': {
2860 System.out.print("\\t");
2861 break;
2862 }
2863 case '\\': {
2864 System.out.print("\\\\");
2865 break;
2866 }
2867 default: {
2868 System.out.print(c);
2869 }
2870 }
2871 }
2872 if (fCurrentEntity.position == fCurrentEntity.count) {
2873 System.out.print('^');
2874 }
2875 System.out.print('"');
2876 }
2877 System.out.print(']');
2878 System.out.print(" @ ");
2879 System.out.print(fCurrentEntity.lineNumber);
2880 System.out.print(',');
2881 System.out.print(fCurrentEntity.columnNumber);
2882 } else {
2883 System.out.print("*NO CURRENT ENTITY*");
2884 }
2885 }
2886 } // print()
2887
2888 /**
2889 * Buffer used in entity manager to reuse character arrays instead
2890 * of creating new ones every time.
2891 *
2892 * @xerces.internal
2893 *
2894 * @author Ankit Pasricha, IBM
2895 */
2896 private static class CharacterBuffer {
2897
2898 /** character buffer */
2899 private char[] ch;
2900
2901 /** whether the buffer is for an external or internal scanned entity */
2902 private boolean isExternal;
2903
2904 public CharacterBuffer(boolean isExternal, int size) {
2905 this.isExternal = isExternal;
2906 ch = new char[size];
2907 }
2908 }
2909
2910
2911 /**
2912 * Stores a number of character buffers and provides it to the entity
2913 * manager to use when an entity is seen.
2914 *
2915 * @xerces.internal
2916 *
2917 * @author Ankit Pasricha, IBM
2918 */
2919 private static class CharacterBufferPool {
2920
2921 private static final int DEFAULT_POOL_SIZE = 3;
2922
2923 private CharacterBuffer[] fInternalBufferPool;
2924 private CharacterBuffer[] fExternalBufferPool;
2925
2926 private int fExternalBufferSize;
2927 private int fInternalBufferSize;
2928 private int poolSize;
2929
2930 private int fInternalTop;
2931 private int fExternalTop;
2932
2933 public CharacterBufferPool(int externalBufferSize, int internalBufferSize) {
2934 this(DEFAULT_POOL_SIZE, externalBufferSize, internalBufferSize);
2935 }
2936
2937 public CharacterBufferPool(int poolSize, int externalBufferSize, int internalBufferSize) {
2938 fExternalBufferSize = externalBufferSize;
2939 fInternalBufferSize = internalBufferSize;
2940 this.poolSize = poolSize;
2941 init();
2942 }
2943
2944 /** Initializes buffer pool. **/
2945 private void init() {
2946 fInternalBufferPool = new CharacterBuffer[poolSize];
2947 fExternalBufferPool = new CharacterBuffer[poolSize];
2948 fInternalTop = -1;
2949 fExternalTop = -1;
2950 }
2951
2952 /** Retrieves buffer from pool. **/
2953 public CharacterBuffer getBuffer(boolean external) {
2954 if (external) {
2955 if (fExternalTop > -1) {
2956 return fExternalBufferPool[fExternalTop--];
2957 }
2958 else {
2959 return new CharacterBuffer(true, fExternalBufferSize);
2960 }
2961 }
2962 else {
2963 if (fInternalTop > -1) {
2964 return fInternalBufferPool[fInternalTop--];
2965 }
2966 else {
2967 return new CharacterBuffer(false, fInternalBufferSize);
2968 }
2969 }
2970 }
2971
2972 /** Returns buffer to pool. **/
2973 public void returnToPool(CharacterBuffer buffer) {
2974 if (buffer.isExternal) {
2975 if (fExternalTop < fExternalBufferPool.length - 1) {
2976 fExternalBufferPool[++fExternalTop] = buffer;
2977 }
2978 }
2979 else if (fInternalTop < fInternalBufferPool.length - 1) {
2980 fInternalBufferPool[++fInternalTop] = buffer;
2981 }
2982 }
2983
2984 /** Sets the size of external buffers and dumps the old pool. **/
2985 public void setExternalBufferSize(int bufferSize) {
2986 fExternalBufferSize = bufferSize;
2987 fExternalBufferPool = new CharacterBuffer[poolSize];
2988 fExternalTop = -1;
2989 }
2990 }
2991
2992 /**
2993 * This class wraps the byte inputstreams we're presented with.
2994 * We need it because java.io.InputStreams don't provide
2995 * functionality to reread processed bytes, and they have a habit
2996 * of reading more than one character when you call their read()
2997 * methods. This means that, once we discover the true (declared)
2998 * encoding of a document, we can neither backtrack to read the
2999 * whole doc again nor start reading where we are with a new
3000 * reader.
3001 *
3002 * This class allows rewinding an inputStream by allowing a mark
3003 * to be set, and the stream reset to that position. <strong>The
3004 * class assumes that it needs to read one character per
3005 * invocation when it's read() method is inovked, but uses the
3006 * underlying InputStream's read(char[], offset length) method--it
3007 * won't buffer data read this way!</strong>
3008 *
3009 * @xerces.internal
3010 *
3011 * @author Neil Graham, IBM
3012 * @author Glenn Marcy, IBM
3013 */
3014
3015 protected final class RewindableInputStream extends InputStream {
3016
3017 private InputStream fInputStream;
3018 private byte[] fData;
3019 private int fStartOffset;
3020 private int fEndOffset;
3021 private int fOffset;
3022 private int fLength;
3023 private int fMark;
3024
3025 public RewindableInputStream(InputStream is) {
3026 fData = new byte[DEFAULT_XMLDECL_BUFFER_SIZE];
3027 fInputStream = is;
3028 fStartOffset = 0;
3029 fEndOffset = -1;
3030 fOffset = 0;
3031 fLength = 0;
3032 fMark = 0;
3033 }
3034
3035 public void setStartOffset(int offset) {
3036 fStartOffset = offset;
3037 }
3038
3039 public void rewind() {
3040 fOffset = fStartOffset;
3041 }
3042
3043 public int read() throws IOException {
3044 int b = 0;
3045 if (fOffset < fLength) {
3046 return fData[fOffset++] & 0xff;
3047 }
3048 if (fOffset == fEndOffset) {
3049 return -1;
3050 }
3051 if (fOffset == fData.length) {
3052 byte[] newData = new byte[fOffset << 1];
3053 System.arraycopy(fData, 0, newData, 0, fOffset);
3054 fData = newData;
3055 }
3056 b = fInputStream.read();
3057 if (b == -1) {
3058 fEndOffset = fOffset;
3059 return -1;
3060 }
3061 fData[fLength++] = (byte)b;
3062 fOffset++;
3063 return b & 0xff;
3064 }
3065
3066 public int read(byte[] b, int off, int len) throws IOException {
3067 int bytesLeft = fLength - fOffset;
3068 if (bytesLeft == 0) {
3069 if (fOffset == fEndOffset) {
3070 return -1;
3071 }
3072
3073 /**
3074 * //System.out.println("fCurrentEntitty = " + fCurrentEntity );
3075 * //System.out.println("fInputStream = " + fInputStream );
3076 * // better get some more for the voracious reader... */
3077
3078 if(fCurrentEntity.mayReadChunks || !fCurrentEntity.xmlDeclChunkRead) {
3079
3080 if (!fCurrentEntity.xmlDeclChunkRead)
3081 {
3082 fCurrentEntity.xmlDeclChunkRead = true;
3083 len = Entity.ScannedEntity.DEFAULT_XMLDECL_BUFFER_SIZE;
3084 }
3085 return fInputStream.read(b, off, len);
3086 }
3087
3088 int returnedVal = read();
3089 if(returnedVal == -1) {
3090 fEndOffset = fOffset;
3091 return -1;
3092 }
3093 b[off] = (byte)returnedVal;
3094 return 1;
3095
3096 }
3097 if (len < bytesLeft) {
3098 if (len <= 0) {
3099 return 0;
3100 }
3101 } else {
3102 len = bytesLeft;
3103 }
3104 if (b != null) {
3105 System.arraycopy(fData, fOffset, b, off, len);
3106 }
3107 fOffset += len;
3108 return len;
3109 }
3110
3111 public long skip(long n)
3112 throws IOException {
3113 int bytesLeft;
3114 if (n <= 0) {
3115 return 0;
3116 }
3117 bytesLeft = fLength - fOffset;
3118 if (bytesLeft == 0) {
3119 if (fOffset == fEndOffset) {
3120 return 0;
3121 }
3122 return fInputStream.skip(n);
3123 }
3124 if (n <= bytesLeft) {
3125 fOffset += n;
3126 return n;
3127 }
3128 fOffset += bytesLeft;
3129 if (fOffset == fEndOffset) {
3130 return bytesLeft;
3131 }
3132 n -= bytesLeft;
3133 /*
3134 * In a manner of speaking, when this class isn't permitting more
3135 * than one byte at a time to be read, it is "blocking". The
3136 * available() method should indicate how much can be read without
3137 * blocking, so while we're in this mode, it should only indicate
3138 * that bytes in its buffer are available; otherwise, the result of
3139 * available() on the underlying InputStream is appropriate.
3140 */
3141 return fInputStream.skip(n) + bytesLeft;
3142 }
3143
3144 public int available() throws IOException {
3145 int bytesLeft = fLength - fOffset;
3146 if (bytesLeft == 0) {
3147 if (fOffset == fEndOffset) {
3148 return -1;
3149 }
3150 return fCurrentEntity.mayReadChunks ? fInputStream.available()
3151 : 0;
3152 }
3153 return bytesLeft;
3154 }
3155
3156 public void mark(int howMuch) {
3157 fMark = fOffset;
3158 }
3159
3160 public void reset() {
3161 fOffset = fMark;
3162 //test();
3163 }
3164
3165 public boolean markSupported() {
3166 return true;
3167 }
3168
3169 public void close() throws IOException {
3170 if (fInputStream != null) {
3171 fInputStream.close();
3172 fInputStream = null;
3173 }
3174 }
3175 } // end of RewindableInputStream class
3176
3177 public void test(){
3178 //System.out.println("TESTING: Added familytree to entityManager");
3179 //Usecase1
3180 fEntityStorage.addExternalEntity("entityUsecase1",null,
3181 "/space/home/stax/sun/6thJan2004/zephyr/data/test.txt",
3182 "/space/home/stax/sun/6thJan2004/zephyr/data/entity.xml");
3183
3184 //Usecase2
3185 fEntityStorage.addInternalEntity("entityUsecase2","<Test>value</Test>");
3186 fEntityStorage.addInternalEntity("entityUsecase3","value3");
3187 fEntityStorage.addInternalEntity("text", "Hello World.");
3188 fEntityStorage.addInternalEntity("empty-element", "<foo/>");
3189 fEntityStorage.addInternalEntity("balanced-element", "<foo></foo>");
3190 fEntityStorage.addInternalEntity("balanced-element-with-text", "<foo>Hello, World</foo>");
3191 fEntityStorage.addInternalEntity("balanced-element-with-entity", "<foo>&text;</foo>");
3192 fEntityStorage.addInternalEntity("unbalanced-entity", "<foo>");
3193 fEntityStorage.addInternalEntity("recursive-entity", "<foo>&recursive-entity2;</foo>");
3194 fEntityStorage.addInternalEntity("recursive-entity2", "<bar>&recursive-entity3;</bar>");
3195 fEntityStorage.addInternalEntity("recursive-entity3", "<baz>&recursive-entity;</baz>");
3196 fEntityStorage.addInternalEntity("ch","©");
3197 fEntityStorage.addInternalEntity("ch1","T");
3198 fEntityStorage.addInternalEntity("% ch2","param");
3199 }
3200
3201 } // class XMLEntityManager