Source Home >> Java Source 1.6.0 >> com.sun.org.apache.xml.internal.dtm.DTM V 0.09

001/*
002 * Copyright 1999-2004 The Apache Software Foundation.
003 *
004 * Licensed under the Apache License, Version 2.0 (the "License");
005 * you may not use this file except in compliance with the License.
006 * You may obtain a copy of the License at
007 *
008 *     http://www.apache.org/licenses/LICENSE-2.0
009 *
010 * Unless required by applicable law or agreed to in writing, software
011 * distributed under the License is distributed on an "AS IS" BASIS,
012 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
013 * See the License for the specific language governing permissions and
014 * limitations under the License.
015 */
016/*
017 * $Id: DTM.java,v 1.2.4.1 2005/09/15 08:14:51 suresh_emailid Exp $
018 */
019package com.sun.org.apache.xml.internal.dtm;
020
021import javax.xml.transform.SourceLocator;
022
023import com.sun.org.apache.xml.internal.utils.XMLString;
024
025/**
026 * <code>DTM</code> is an XML document model expressed as a table
027 * rather than an object tree. It attempts to provide an interface to
028 * a parse tree that has very little object creation. (DTM
029 * implementations may also support incremental construction of the
030 * model, but that's hidden from the DTM API.)
031 *
032 * <p>Nodes in the DTM are identified by integer "handles".  A handle must
033 * be unique within a process, and carries both node identification and
034 * document identification.  It must be possible to compare two handles
035 * (and thus their nodes) for identity with "==".</p>
036 *
037 * <p>Namespace URLs, local-names, and expanded-names can all be
038 * represented by and tested as integer ID values.  An expanded name
039 * represents (and may or may not directly contain) a combination of
040 * the URL ID, and the local-name ID.  Note that the namespace URL id
041 * can be 0, which should have the meaning that the namespace is null.
042 * For consistancy, zero should not be used for a local-name index. </p>
043 *
044 * <p>Text content of a node is represented by an index and length,
045 * permitting efficient storage such as a shared FastStringBuffer.</p>
046 *
047 * <p>The model of the tree, as well as the general navigation model,
048 * is that of XPath 1.0, for the moment.  The model will eventually be
049 * adapted to match the XPath 2.0 data model, XML Schema, and
050 * InfoSet.</p>
051 *
052 * <p>DTM does _not_ directly support the W3C's Document Object
053 * Model. However, it attempts to come close enough that an
054 * implementation of DTM can be created that wraps a DOM and vice
055 * versa.</p>
056 *
057 * <p><strong>Please Note:</strong> The DTM API is still
058 * <strong>Subject To Change.</strong> This wouldn't affect most
059 * users, but might require updating some extensions.</p>
060 *
061 * <p> The largest change being contemplated is a reconsideration of
062 * the Node Handle representation.  We are still not entirely sure
063 * that an integer packed with two numeric subfields is really the
064 * best solution. It has been suggested that we move up to a Long, to
065 * permit more nodes per document without having to reduce the number
066 * of slots in the DTMManager. There's even been a proposal that we
067 * replace these integers with "cursor" objects containing the
068 * internal node id and a pointer to the actual DTM object; this might
069 * reduce the need to continuously consult the DTMManager to retrieve
070 * the latter, and might provide a useful "hook" back into normal Java
071 * heap management.  But changing this datatype would have huge impact
072 * on Xalan's internals -- especially given Java's lack of C-style
073 * typedefs -- so we won't cut over unless we're convinced the new
074 * solution really would be an improvement!</p>
075 * */
076public interface DTM
077{
078
079  /**
080   * Null node handles are represented by this value.
081   */
082  public static final int NULL = -1;
083
084  // These nodeType mnemonics and values are deliberately the same as those
085  // used by the DOM, for convenient mapping
086  //
087  // %REVIEW% Should we actually define these as initialized to,
088  // eg. org.w3c.dom.Document.ELEMENT_NODE?
089
090  /**
091   * The node is a <code>Root</code>.
092   */
093  public static final short ROOT_NODE = 0;
094  
095  /**
096   * The node is an <code>Element</code>.
097   */
098  public static final short ELEMENT_NODE = 1;
099
100  /**
101   * The node is an <code>Attr</code>.
102   */
103  public static final short ATTRIBUTE_NODE = 2;
104
105  /**
106   * The node is a <code>Text</code> node.
107   */
108  public static final short TEXT_NODE = 3;
109
110  /**
111   * The node is a <code>CDATASection</code>.
112   */
113  public static final short CDATA_SECTION_NODE = 4;
114
115  /**
116   * The node is an <code>EntityReference</code>.
117   */
118  public static final short ENTITY_REFERENCE_NODE = 5;
119
120  /**
121   * The node is an <code>Entity</code>.
122   */
123  public static final short ENTITY_NODE = 6;
124
125  /**
126   * The node is a <code>ProcessingInstruction</code>.
127   */
128  public static final short PROCESSING_INSTRUCTION_NODE = 7;
129
130  /**
131   * The node is a <code>Comment</code>.
132   */
133  public static final short COMMENT_NODE = 8;
134
135  /**
136   * The node is a <code>Document</code>.
137   */
138  public static final short DOCUMENT_NODE = 9;
139
140  /**
141   * The node is a <code>DocumentType</code>.
142   */
143  public static final short DOCUMENT_TYPE_NODE = 10;
144
145  /**
146   * The node is a <code>DocumentFragment</code>.
147   */
148  public static final short DOCUMENT_FRAGMENT_NODE = 11;
149
150  /**
151   * The node is a <code>Notation</code>.
152   */
153  public static final short NOTATION_NODE = 12;
154
155  /**
156   * The node is a <code>namespace node</code>. Note that this is not
157   * currently a node type defined by the DOM API.
158   */
159  public static final short NAMESPACE_NODE = 13;
160  
161  /**
162   * The number of valid nodetypes.
163   */
164  public static final short  NTYPES = 14;
165
166  // ========= DTM Implementation Control Functions. ==============
167  // %TBD% RETIRED -- do via setFeature if needed. Remove from impls.
168  // public void setParseBlockSize(int blockSizeSuggestion);
169
170  /**
171   * Set an implementation dependent feature.
172   * <p>
173   * %REVIEW% Do we really expect to set features on DTMs?
174   *
175   * @param featureId A feature URL.
176   * @param state true if this feature should be on, false otherwise.
177   */
178  public void setFeature(String featureId, boolean state);
179
180  /**
181   * Set a run time property for this DTM instance.
182   *
183   * @param property a <code>String</code> value
184   * @param value an <code>Object</code> value
185   */
186  public void setProperty(String property, Object value);
187
188  // ========= Document Navigation Functions =========
189
190  /**
191   * This returns a stateless "traverser", that can navigate over an
192   * XPath axis, though not in document order.
193   *
194   * @param axis One of Axes.ANCESTORORSELF, etc.
195   *
196   * @return A DTMAxisIterator, or null if the givin axis isn't supported.
197   */
198  public DTMAxisTraverser getAxisTraverser(final int axis);
199
200  /**
201   * This is a shortcut to the iterators that implement
202   * XPath axes.
203   * Returns a bare-bones iterator that must be initialized
204   * with a start node (using iterator.setStartNode()).
205   *
206   * @param axis One of Axes.ANCESTORORSELF, etc.
207   *
208   * @return A DTMAxisIterator, or null if the givin axis isn't supported.
209   */
210  public DTMAxisIterator getAxisIterator(final int axis);
211
212  /**
213   * Get an iterator that can navigate over an XPath Axis, predicated by
214   * the extended type ID.
215   *
216   * @param axis
217   * @param type An extended type ID.
218   *
219   * @return A DTMAxisIterator, or null if the givin axis isn't supported.
220   */
221  public DTMAxisIterator getTypedAxisIterator(final int axis, final int type);
222
223  /**
224   * Given a node handle, test if it has child nodes.
225   * <p> %REVIEW% This is obviously useful at the DOM layer, where it
226   * would permit testing this without having to create a proxy
227   * node. It's less useful in the DTM API, where
228   * (dtm.getFirstChild(nodeHandle)!=DTM.NULL) is just as fast and
229   * almost as self-evident. But it's a convenience, and eases porting
230   * of DOM code to DTM.  </p>
231   *
232   * @param nodeHandle int Handle of the node.
233   * @return int true if the given node has child nodes.
234   */
235  public boolean hasChildNodes(int nodeHandle);
236
237  /**
238   * Given a node handle, get the handle of the node's first child.
239   *
240   * @param nodeHandle int Handle of the node.
241   * @return int DTM node-number of first child,
242   * or DTM.NULL to indicate none exists.
243   */
244  public int getFirstChild(int nodeHandle);
245
246  /**
247   * Given a node handle, get the handle of the node's last child.
248   *
249   * @param nodeHandle int Handle of the node.
250   * @return int Node-number of last child,
251   * or DTM.NULL to indicate none exists.
252   */
253  public int getLastChild(int nodeHandle);
254
255  /**
256   * Retrieves an attribute node by local name and namespace URI
257   *
258   * %TBD% Note that we currently have no way to support
259   * the DOM's old getAttribute() call, which accesses only the qname.
260   *
261   * @param elementHandle Handle of the node upon which to look up this attribute.
262   * @param namespaceURI The namespace URI of the attribute to
263   *   retrieve, or null.
264   * @param name The local name of the attribute to
265   *   retrieve.
266   * @return The attribute node handle with the specified name (
267   *   <code>nodeName</code>) or <code>DTM.NULL</code> if there is no such
268   *   attribute.
269   */
270  public int getAttributeNode(int elementHandle, String namespaceURI,
271                              String name);
272
273  /**
274   * Given a node handle, get the index of the node's first attribute.
275   *
276   * @param nodeHandle int Handle of the node.
277   * @return Handle of first attribute, or DTM.NULL to indicate none exists.
278   */
279  public int getFirstAttribute(int nodeHandle);
280
281  /**
282   * Given a node handle, get the index of the node's first namespace node.
283   *
284   * @param nodeHandle handle to node, which should probably be an element
285   *                   node, but need not be.
286   *
287   * @param inScope true if all namespaces in scope should be
288   *                   returned, false if only the node's own
289   *                   namespace declarations should be returned.
290   * @return handle of first namespace,
291   * or DTM.NULL to indicate none exists.
292   */
293  public int getFirstNamespaceNode(int nodeHandle, boolean inScope);
294
295  /**
296   * Given a node handle, advance to its next sibling.
297   * @param nodeHandle int Handle of the node.
298   * @return int Node-number of next sibling,
299   * or DTM.NULL to indicate none exists.
300   */
301  public int getNextSibling(int nodeHandle);
302
303  /**
304   * Given a node handle, find its preceeding sibling.
305   * WARNING: DTM implementations may be asymmetric; in some,
306   * this operation has been resolved by search, and is relatively expensive.
307   *
308   * @param nodeHandle the id of the node.
309   * @return int Node-number of the previous sib,
310   * or DTM.NULL to indicate none exists.
311   */
312  public int getPreviousSibling(int nodeHandle);
313
314  /**
315   * Given a node handle, advance to the next attribute. If an
316   * element, we advance to its first attribute; if an attr, we advance to
317   * the next attr of the same element.
318   *
319   * @param nodeHandle int Handle of the node.
320   * @return int DTM node-number of the resolved attr,
321   * or DTM.NULL to indicate none exists.
322   */
323  public int getNextAttribute(int nodeHandle);
324
325  /**
326   * Given a namespace handle, advance to the next namespace in the same scope
327   * (local or local-plus-inherited, as selected by getFirstNamespaceNode)
328   *
329   * @param baseHandle handle to original node from where the first child
330   * was relative to (needed to return nodes in document order).
331   * @param namespaceHandle handle to node which must be of type
332   * NAMESPACE_NODE.
333   * NEEDSDOC @param inScope
334   * @return handle of next namespace,
335   * or DTM.NULL to indicate none exists.
336   */
337  public int getNextNamespaceNode(int baseHandle, int namespaceHandle,
338                                  boolean inScope);
339
340  /**
341   * Given a node handle, find its parent node.
342   *
343   * @param nodeHandle the id of the node.
344   * @return int Node handle of parent,
345   * or DTM.NULL to indicate none exists.
346   */
347  public int getParent(int nodeHandle);
348
349  /**
350   * Given a DTM which contains only a single document, 
351   * find the Node Handle of the  Document node. Note 
352   * that if the DTM is configured so it can contain multiple
353   * documents, this call will return the Document currently
354   * under construction -- but may return null if it's between
355   * documents. Generally, you should use getOwnerDocument(nodeHandle)
356   * or getDocumentRoot(nodeHandle) instead.
357   *
358   * @return int Node handle of document, or DTM.NULL if a shared DTM
359   * can not tell us which Document is currently active.
360   */
361  public int getDocument();
362
363  /**
364   * Given a node handle, find the owning document node. This version mimics
365   * the behavior of the DOM call by the same name.
366   *
367   * @param nodeHandle the id of the node.
368   * @return int Node handle of owning document, or DTM.NULL if the node was
369   * a Document.
370   * @see #getDocumentRoot(int nodeHandle)
371   */
372  public int getOwnerDocument(int nodeHandle);
373
374  /**
375   * Given a node handle, find the owning document node.
376   *
377   * @param nodeHandle the id of the node.
378   * @return int Node handle of owning document, or the node itself if it was
379   * a Document. (Note difference from DOM, where getOwnerDocument returns
380   * null for the Document node.)
381   * @see #getOwnerDocument(int nodeHandle)
382   */
383  public int getDocumentRoot(int nodeHandle);
384
385  /**
386   * Get the string-value of a node as a String object
387   * (see http://www.w3.org/TR/xpath#data-model
388   * for the definition of a node's string-value).
389   *
390   * @param nodeHandle The node ID.
391   *
392   * @return A string object that represents the string-value of the given node.
393   */
394  public XMLString getStringValue(int nodeHandle);
395
396  /**
397   * Get number of character array chunks in
398   * the string-value of a node.
399   * (see http://www.w3.org/TR/xpath#data-model
400   * for the definition of a node's string-value).
401   * Note that a single text node may have multiple text chunks.
402   *
403   * @param nodeHandle The node ID.
404   *
405   * @return number of character array chunks in
406   *         the string-value of a node.
407   */
408  public int getStringValueChunkCount(int nodeHandle);
409
410  /**
411   * Get a character array chunk in the string-value of a node.
412   * (see http://www.w3.org/TR/xpath#data-model
413   * for the definition of a node's string-value).
414   * Note that a single text node may have multiple text chunks.
415   *
416   * @param nodeHandle The node ID.
417   * @param chunkIndex Which chunk to get.
418   * @param startAndLen  A two-integer array which, upon return, WILL
419   * BE FILLED with values representing the chunk's start position
420   * within the returned character buffer and the length of the chunk.
421   * @return The character array buffer within which the chunk occurs,
422   * setting startAndLen's contents as a side-effect.
423   */
424  public char[] getStringValueChunk(int nodeHandle, int chunkIndex,
425                                    int[] startAndLen);
426
427  /**
428   * Given a node handle, return an ID that represents the node's expanded name.
429   *
430   * @param nodeHandle The handle to the node in question.
431   *
432   * @return the expanded-name id of the node.
433   */
434  public int getExpandedTypeID(int nodeHandle);
435
436  /**
437   * Given an expanded name, return an ID.  If the expanded-name does not
438   * exist in the internal tables, the entry will be created, and the ID will
439   * be returned.  Any additional nodes that are created that have this
440   * expanded name will use this ID.
441   *
442   * NEEDSDOC @param namespace
443   * NEEDSDOC @param localName
444   * NEEDSDOC @param type
445   *
446   * @return the expanded-name id of the node.
447   */
448  public int getExpandedTypeID(String namespace, String localName, int type);
449
450  /**
451   * Given an expanded-name ID, return the local name part.
452   *
453   * @param ExpandedNameID an ID that represents an expanded-name.
454   * @return String Local name of this node.
455   */
456  public String getLocalNameFromExpandedNameID(int ExpandedNameID);
457
458  /**
459   * Given an expanded-name ID, return the namespace URI part.
460   *
461   * @param ExpandedNameID an ID that represents an expanded-name.
462   * @return String URI value of this node's namespace, or null if no
463   * namespace was resolved.
464   */
465  public String getNamespaceFromExpandedNameID(int ExpandedNameID);
466
467  /**
468   * Given a node handle, return its DOM-style node name. This will
469   * include names such as #text or #document.
470   *
471   * @param nodeHandle the id of the node.
472   * @return String Name of this node, which may be an empty string.
473   * %REVIEW% Document when empty string is possible...
474   */
475  public String getNodeName(int nodeHandle);
476
477  /**
478   * Given a node handle, return the XPath node name.  This should be
479   * the name as described by the XPath data model, NOT the DOM-style
480   * name.
481   *
482   * @param nodeHandle the id of the node.
483   * @return String Name of this node.
484   */
485  public String getNodeNameX(int nodeHandle);
486
487  /**
488   * Given a node handle, return its DOM-style localname.
489   * (As defined in Namespaces, this is the portion of the name after the
490   * prefix, if present, or the whole node name if no prefix exists)
491   *
492   * @param nodeHandle the id of the node.
493   * @return String Local name of this node.
494   */
495  public String getLocalName(int nodeHandle);
496
497  /**
498   * Given a namespace handle, return the prefix that the namespace decl is
499   * mapping.
500   * Given a node handle, return the prefix used to map to the namespace.
501   * (As defined in Namespaces, this is the portion of the name before any
502   * colon character).
503   *
504   * <p> %REVIEW% Are you sure you want "" for no prefix?  </p>
505   *
506   * @param nodeHandle the id of the node.
507   * @return String prefix of this node's name, or "" if no explicit
508   * namespace prefix was given.
509   */
510  public String getPrefix(int nodeHandle);
511
512  /**
513   * Given a node handle, return its DOM-style namespace URI
514   * (As defined in Namespaces, this is the declared URI which this node's
515   * prefix -- or default in lieu thereof -- was mapped to.)
516   * @param nodeHandle the id of the node.
517   * @return String URI value of this node's namespace, or null if no
518   * namespace was resolved.
519   */
520  public String getNamespaceURI(int nodeHandle);
521
522  /**
523   * Given a node handle, return its node value. This is mostly
524   * as defined by the DOM, but may ignore some conveniences.
525   * <p>
526   * @param nodeHandle The node id.
527   * @return String Value of this node, or null if not
528   * meaningful for this node type.
529   */
530  public String getNodeValue(int nodeHandle);
531
532  /**
533   * Given a node handle, return its DOM-style node type.
534   *
535   * <p>%REVIEW% Generally, returning short is false economy. Return int?</p>
536   *
537   * @param nodeHandle The node id.
538   * @return int Node type, as per the DOM's Node._NODE constants.
539   */
540  public short getNodeType(int nodeHandle);
541
542  /**
543   * Get the depth level of this node in the tree (equals 1 for
544   * a parentless node).
545   *
546   * @param nodeHandle The node id.
547   * @return the number of ancestors, plus one
548   * @xsl.usage internal
549   */
550  public short getLevel(int nodeHandle);
551
552  // ============== Document query functions ==============
553
554  /**
555   * Tests whether DTM DOM implementation implements a specific feature and
556   * that feature is supported by this node.
557   * @param feature The name of the feature to test.
558   * @param version This is the version number of the feature to test.
559   *   If the version is not
560   *   specified, supporting any version of the feature will cause the
561   *   method to return <code>true</code>.
562   * @return Returns <code>true</code> if the specified feature is
563   *   supported on this node, <code>false</code> otherwise.
564   */
565  public boolean isSupported(String feature, String version);
566
567  /**
568   * Return the base URI of the document entity. If it is not known
569   * (because the document was parsed from a socket connection or from
570   * standard input, for example), the value of this property is unknown.
571   *
572   * @return the document base URI String object or null if unknown.
573   */
574  public String getDocumentBaseURI();
575
576  /**
577   * Set the base URI of the document entity.
578   *
579   * @param baseURI the document base URI String object or null if unknown.
580   */
581  public void setDocumentBaseURI(String baseURI);
582
583  /**
584   * Return the system identifier of the document entity. If
585   * it is not known, the value of this property is null.
586   *
587   * @param nodeHandle The node id, which can be any valid node handle.
588   * @return the system identifier String object or null if unknown.
589   */
590  public String getDocumentSystemIdentifier(int nodeHandle);
591
592  /**
593   * Return the name of the character encoding scheme
594   *        in which the document entity is expressed.
595   *
596   * @param nodeHandle The node id, which can be any valid node handle.
597   * @return the document encoding String object.
598   */
599  public String getDocumentEncoding(int nodeHandle);
600
601  /**
602   * Return an indication of the standalone status of the document,
603   *        either "yes" or "no". This property is derived from the optional
604   *        standalone document declaration in the XML declaration at the
605   *        beginning of the document entity, and has no value if there is no
606   *        standalone document declaration.
607   *
608   * @param nodeHandle The node id, which can be any valid node handle.
609   * @return the document standalone String object, either "yes", "no", or null.
610   */
611  public String getDocumentStandalone(int nodeHandle);
612
613  /**
614   * Return a string representing the XML version of the document. This
615   * property is derived from the XML declaration optionally present at the
616   * beginning of the document entity, and has no value if there is no XML
617   * declaration.
618   *
619   * @param documentHandle the document handle
620   * @return the document version String object
621   */
622  public String getDocumentVersion(int documentHandle);
623
624  /**
625   * Return an indication of
626   * whether the processor has read the complete DTD. Its value is a
627   * boolean. If it is false, then certain properties (indicated in their
628   * descriptions below) may be unknown. If it is true, those properties
629   * are never unknown.
630   *
631   * @return <code>true</code> if all declarations were processed;
632   *         <code>false</code> otherwise.
633   */
634  public boolean getDocumentAllDeclarationsProcessed();
635
636  /**
637   *   A document type declaration information item has the following properties:
638   *
639   *     1. [system identifier] The system identifier of the external subset, if
640   *        it exists. Otherwise this property has no value.
641   *
642   * @return the system identifier String object, or null if there is none.
643   */
644  public String getDocumentTypeDeclarationSystemIdentifier();
645
646  /**
647   * Return the public identifier of the external subset,
648   * normalized as described in 4.2.2 External Entities [XML]. If there is
649   * no external subset or if it has no public identifier, this property
650   * has no value.
651   *
652   * @return the public identifier String object, or null if there is none.
653   */
654  public String getDocumentTypeDeclarationPublicIdentifier();
655
656  /**
657   * Returns the <code>Element</code> whose <code>ID</code> is given by
658   * <code>elementId</code>. If no such element exists, returns
659   * <code>DTM.NULL</code>. Behavior is not defined if more than one element
660   * has this <code>ID</code>. Attributes (including those
661   * with the name "ID") are not of type ID unless so defined by DTD/Schema
662   * information available to the DTM implementation.
663   * Implementations that do not know whether attributes are of type ID or
664   * not are expected to return <code>DTM.NULL</code>.
665   *
666   * <p>%REVIEW% Presumably IDs are still scoped to a single document,
667   * and this operation searches only within a single document, right?
668   * Wouldn't want collisions between DTMs in the same process.</p>
669   *
670   * @param elementId The unique <code>id</code> value for an element.
671   * @return The handle of the matching element.
672   */
673  public int getElementById(String elementId);
674
675  /**
676   * The getUnparsedEntityURI function returns the URI of the unparsed
677   * entity with the specified name in the same document as the context
678   * node (see [3.3 Unparsed Entities]). It returns the empty string if
679   * there is no such entity.
680   * <p>
681   * XML processors may choose to use the System Identifier (if one
682   * is provided) to resolve the entity, rather than the URI in the
683   * Public Identifier. The details are dependent on the processor, and
684   * we would have to support some form of plug-in resolver to handle
685   * this properly. Currently, we simply return the System Identifier if
686   * present, and hope that it a usable URI or that our caller can
687   * map it to one.
688   * %REVIEW% Resolve Public Identifiers... or consider changing function name.
689   * <p>
690   * If we find a relative URI
691   * reference, XML expects it to be resolved in terms of the base URI
692   * of the document. The DOM doesn't do that for us, and it isn't
693   * entirely clear whether that should be done here; currently that's
694   * pushed up to a higher level of our application. (Note that DOM Level
695   * 1 didn't store the document's base URI.)
696   * %REVIEW% Consider resolving Relative URIs.
697   * <p>
698   * (The DOM's statement that "An XML processor may choose to
699   * completely expand entities before the structure model is passed
700   * to the DOM" refers only to parsed entities, not unparsed, and hence
701   * doesn't affect this function.)
702   *
703   * @param name A string containing the Entity Name of the unparsed
704   * entity.
705   *
706   * @return String containing the URI of the Unparsed Entity, or an
707   * empty string if no such entity exists.
708   */
709  public String getUnparsedEntityURI(String name);
710
711  // ============== Boolean methods ================
712
713  /**
714   * Return true if the xsl:strip-space or xsl:preserve-space was processed
715   * during construction of the document contained in this DTM.
716   *
717   * NEEDSDOC ($objectName$) @return
718   */
719  public boolean supportsPreStripping();
720
721  /**
722   * Figure out whether nodeHandle2 should be considered as being later
723   * in the document than nodeHandle1, in Document Order as defined
724   * by the XPath model. This may not agree with the ordering defined
725   * by other XML applications.
726   * <p>
727   * There are some cases where ordering isn't defined, and neither are
728   * the results of this function -- though we'll generally return true.
729   * <p>
730   * %REVIEW% Make sure this does the right thing with attribute nodes!!!
731   * <p>
732   * %REVIEW% Consider renaming for clarity. Perhaps isDocumentOrder(a,b)?
733   *
734   * @param firstNodeHandle DOM Node to perform position comparison on.
735   * @param secondNodeHandle DOM Node to perform position comparison on.
736   *
737   * @return false if secondNode comes before firstNode, otherwise return true.
738   * You can think of this as
739   * <code>(firstNode.documentOrderPosition <= secondNode.documentOrderPosition)</code>.
740   */
741  public boolean isNodeAfter(int firstNodeHandle, int secondNodeHandle);
742
743  /**
744   * 2. [element content whitespace] A boolean indicating whether a
745   * text node represents white space appearing within element content
746   * (see [XML], 2.10 "White Space Handling").  Note that validating
747   * XML processors are required by XML 1.0 to provide this
748   * information... but that DOM Level 2 did not support it, since it
749   * depends on knowledge of the DTD which DOM2 could not guarantee
750   * would be available.
751   * <p>
752   * If there is no declaration for the containing element, an XML
753   * processor must assume that the whitespace could be meaningful and
754   * return false. If no declaration has been read, but the [all
755   * declarations processed] property of the document information item
756   * is false (so there may be an unread declaration), then the value
757   * of this property is indeterminate for white space characters and
758   * should probably be reported as false. It is always false for text
759   * nodes that contain anything other than (or in addition to) white
760   * space.
761   * <p>
762   * Note too that it always returns false for non-Text nodes.
763   * <p>
764   * %REVIEW% Joe wants to rename this isWhitespaceInElementContent() for clarity
765   *
766   * @param nodeHandle the node ID.
767   * @return <code>true</code> if the node definitely represents whitespace in
768   * element content; <code>false</code> otherwise.
769   */
770  public boolean isCharacterElementContentWhitespace(int nodeHandle);
771
772  /**
773   *    10. [all declarations processed] This property is not strictly speaking
774   *        part of the infoset of the document. Rather it is an indication of
775   *        whether the processor has read the complete DTD. Its value is a
776   *        boolean. If it is false, then certain properties (indicated in their
777   *        descriptions below) may be unknown. If it is true, those properties
778   *        are never unknown.
779   *
780   * @param documentHandle A node handle that must identify a document.
781   * @return <code>true</code> if all declarations were processed;
782   *         <code>false</code> otherwise.
783   */
784  public boolean isDocumentAllDeclarationsProcessed(int documentHandle);
785
786  /**
787   *     5. [specified] A flag indicating whether this attribute was actually
788   *        specified in the start-tag of its element, or was defaulted from the
789   *        DTD (or schema).
790   *
791   * @param attributeHandle The attribute handle
792   * @return <code>true</code> if the attribute was specified;
793   *         <code>false</code> if it was defaulted or the handle doesn't
794   *            refer to an attribute node.
795   */
796  public boolean isAttributeSpecified(int attributeHandle);
797
798  // ========== Direct SAX Dispatch, for optimization purposes ========
799
800  /**
801   * Directly call the
802   * characters method on the passed ContentHandler for the
803   * string-value of the given node (see http://www.w3.org/TR/xpath#data-model
804   * for the definition of a node's string-value). Multiple calls to the
805   * ContentHandler's characters methods may well occur for a single call to
806   * this method.
807   *
808   * @param nodeHandle The node ID.
809   * @param ch A non-null reference to a ContentHandler.
810   * @param normalize true if the content should be normalized according to
811   * the rules for the XPath
812   * <a href="http://www.w3.org/TR/xpath#function-normalize-space">normalize-space</a>
813   * function.
814   *
815   * @throws org.xml.sax.SAXException
816   */
817  public void dispatchCharactersEvents(
818    int nodeHandle, org.xml.sax.ContentHandler ch, boolean normalize)
819      throws org.xml.sax.SAXException;
820
821  /**
822   * Directly create SAX parser events representing the XML content of
823   * a DTM subtree. This is a "serialize" operation.
824   *
825   * @param nodeHandle The node ID.
826   * @param ch A non-null reference to a ContentHandler.
827   *
828   * @throws org.xml.sax.SAXException
829   */
830  public void dispatchToEvents(int nodeHandle, org.xml.sax.ContentHandler ch)
831    throws org.xml.sax.SAXException;
832
833  /**
834   * Return an DOM node for the given node.
835   *
836   * @param nodeHandle The node ID.
837   *
838   * @return A node representation of the DTM node.
839   */
840  public org.w3c.dom.Node getNode(int nodeHandle);
841
842  // ==== Construction methods (may not be supported by some implementations!) =====
843  // %REVIEW% What response occurs if not supported?
844
845  /**
846   * @return true iff we're building this model incrementally (eg
847   * we're partnered with a CoroutineParser) and thus require that the
848   * transformation and the parse run simultaneously. Guidance to the
849   * DTMManager.
850   */
851  public boolean needsTwoThreads();
852
853  // %REVIEW% Do these appends make any sense, should we support a
854  // wider set of methods (like the "append" methods in the
855  // current DTMDocumentImpl draft), or should we just support SAX
856  // listener interfaces?  Should it be a separate interface to
857  // make that distinction explicit?
858
859  /**
860   * Return this DTM's content handler, if it has one.
861   *
862   * @return null if this model doesn't respond to SAX events.
863   */
864  public org.xml.sax.ContentHandler getContentHandler();
865
866  /**
867   * Return this DTM's lexical handler, if it has one.
868   *
869   * %REVIEW% Should this return null if constrution already done/begun?
870   *
871   * @return null if this model doesn't respond to lexical SAX events.
872   */
873  public org.xml.sax.ext.LexicalHandler getLexicalHandler();
874
875  /**
876   * Return this DTM's EntityResolver, if it has one.
877   *
878   * @return null if this model doesn't respond to SAX entity ref events.
879   */
880  public org.xml.sax.EntityResolver getEntityResolver();
881
882  /**
883   * Return this DTM's DTDHandler, if it has one.
884   *
885   * @return null if this model doesn't respond to SAX dtd events.
886   */
887  public org.xml.sax.DTDHandler getDTDHandler();
888
889  /**
890   * Return this DTM's ErrorHandler, if it has one.
891   *
892   * @return null if this model doesn't respond to SAX error events.
893   */
894  public org.xml.sax.ErrorHandler getErrorHandler();
895
896  /**
897   * Return this DTM's DeclHandler, if it has one.
898   *
899   * @return null if this model doesn't respond to SAX Decl events.
900   */
901  public org.xml.sax.ext.DeclHandler getDeclHandler();
902
903  /**
904   * Append a child to "the end of the document". Please note that
905   * the node is always cloned in a base DTM, since our basic behavior
906   * is immutable so nodes can't be removed from their previous
907   * location.
908   *
909   * <p> %REVIEW%  DTM maintains an insertion cursor which
910   * performs a depth-first tree walk as nodes come in, and this operation
911   * is really equivalent to:
912   *    insertionCursor.appendChild(document.importNode(newChild)))
913   * where the insert point is the last element that was appended (or
914   * the last one popped back to by an end-element operation).</p>
915   *
916   * @param newChild Must be a valid new node handle.
917   * @param clone true if the child should be cloned into the document.
918   * @param cloneDepth if the clone argument is true, specifies that the
919   *                   clone should include all it's children.
920   */
921  public void appendChild(int newChild, boolean clone, boolean cloneDepth);
922
923  /**
924   * Append a text node child that will be constructed from a string,
925   * to the end of the document. Behavior is otherwise like appendChild().
926   *
927   * @param str Non-null reference to a string.
928   */
929  public void appendTextChild(String str);
930
931  /**
932   * Get the location of a node in the source document.
933   *
934   * @param node an <code>int</code> value
935   * @return a <code>SourceLocator</code> value or null if no location
936   * is available
937   */
938  public SourceLocator getSourceLocatorFor(int node);
939
940  /**
941   * As the DTM is registered with the DTMManager, this method
942   * will be called. This will give the DTM implementation a
943   * chance to initialize any subsystems that are required to
944   * build the DTM
945   */
946  public void documentRegistration();
947
948  /**
949   * As documents are released from the DTMManager, the DTM implementation
950   * will be notified of the event. This will allow the DTM implementation
951   * to shutdown any subsystem activity that may of been assoiated with
952   * the active DTM Implementation.
953   */
954
955   public void documentRelease();
956
957   /**
958    * Migrate a DTM built with an old DTMManager to a new DTMManager.
959    * After the migration, the new DTMManager will treat the DTM as
960    * one that is built by itself.
961    * This is used to support DTM sharing between multiple transformations.
962    * @param manager the DTMManager
963    */
964   public void migrateTo(DTMManager manager);
965}