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 1.14 2005/01/23 00:52:40 mcnamara Exp $
018: */
019: package org.apache.xml.dtm;
020:
021: import javax.xml.transform.SourceLocator;
022:
023: import org.apache.xml.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: * */
076: public interface DTM {
077:
078: /**
079: * Null node handles are represented by this value.
080: */
081: public static final int NULL = -1;
082:
083: // These nodeType mnemonics and values are deliberately the same as those
084: // used by the DOM, for convenient mapping
085: //
086: // %REVIEW% Should we actually define these as initialized to,
087: // eg. org.w3c.dom.Document.ELEMENT_NODE?
088:
089: /**
090: * The node is a <code>Root</code>.
091: */
092: public static final short ROOT_NODE = 0;
093:
094: /**
095: * The node is an <code>Element</code>.
096: */
097: public static final short ELEMENT_NODE = 1;
098:
099: /**
100: * The node is an <code>Attr</code>.
101: */
102: public static final short ATTRIBUTE_NODE = 2;
103:
104: /**
105: * The node is a <code>Text</code> node.
106: */
107: public static final short TEXT_NODE = 3;
108:
109: /**
110: * The node is a <code>CDATASection</code>.
111: */
112: public static final short CDATA_SECTION_NODE = 4;
113:
114: /**
115: * The node is an <code>EntityReference</code>.
116: */
117: public static final short ENTITY_REFERENCE_NODE = 5;
118:
119: /**
120: * The node is an <code>Entity</code>.
121: */
122: public static final short ENTITY_NODE = 6;
123:
124: /**
125: * The node is a <code>ProcessingInstruction</code>.
126: */
127: public static final short PROCESSING_INSTRUCTION_NODE = 7;
128:
129: /**
130: * The node is a <code>Comment</code>.
131: */
132: public static final short COMMENT_NODE = 8;
133:
134: /**
135: * The node is a <code>Document</code>.
136: */
137: public static final short DOCUMENT_NODE = 9;
138:
139: /**
140: * The node is a <code>DocumentType</code>.
141: */
142: public static final short DOCUMENT_TYPE_NODE = 10;
143:
144: /**
145: * The node is a <code>DocumentFragment</code>.
146: */
147: public static final short DOCUMENT_FRAGMENT_NODE = 11;
148:
149: /**
150: * The node is a <code>Notation</code>.
151: */
152: public static final short NOTATION_NODE = 12;
153:
154: /**
155: * The node is a <code>namespace node</code>. Note that this is not
156: * currently a node type defined by the DOM API.
157: */
158: public static final short NAMESPACE_NODE = 13;
159:
160: /**
161: * The number of valid nodetypes.
162: */
163: public static final short NTYPES = 14;
164:
165: // ========= DTM Implementation Control Functions. ==============
166: // %TBD% RETIRED -- do via setFeature if needed. Remove from impls.
167: // public void setParseBlockSize(int blockSizeSuggestion);
168:
169: /**
170: * Set an implementation dependent feature.
171: * <p>
172: * %REVIEW% Do we really expect to set features on DTMs?
173: *
174: * @param featureId A feature URL.
175: * @param state true if this feature should be on, false otherwise.
176: */
177: public void setFeature(String featureId, boolean state);
178:
179: /**
180: * Set a run time property for this DTM instance.
181: *
182: * @param property a <code>String</code> value
183: * @param value an <code>Object</code> value
184: */
185: public void setProperty(String property, Object value);
186:
187: // ========= Document Navigation Functions =========
188:
189: /**
190: * This returns a stateless "traverser", that can navigate over an
191: * XPath axis, though not in document order.
192: *
193: * @param axis One of Axes.ANCESTORORSELF, etc.
194: *
195: * @return A DTMAxisIterator, or null if the givin axis isn't supported.
196: */
197: public DTMAxisTraverser getAxisTraverser(final int axis);
198:
199: /**
200: * This is a shortcut to the iterators that implement
201: * XPath axes.
202: * Returns a bare-bones iterator that must be initialized
203: * with a start node (using iterator.setStartNode()).
204: *
205: * @param axis One of Axes.ANCESTORORSELF, etc.
206: *
207: * @return A DTMAxisIterator, or null if the givin axis isn't supported.
208: */
209: public DTMAxisIterator getAxisIterator(final int axis);
210:
211: /**
212: * Get an iterator that can navigate over an XPath Axis, predicated by
213: * the extended type ID.
214: *
215: * @param axis
216: * @param type An extended type ID.
217: *
218: * @return A DTMAxisIterator, or null if the givin axis isn't supported.
219: */
220: public DTMAxisIterator getTypedAxisIterator(final int axis,
221: 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,
338: int namespaceHandle, 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,
449: int type);
450:
451: /**
452: * Given an expanded-name ID, return the local name part.
453: *
454: * @param ExpandedNameID an ID that represents an expanded-name.
455: * @return String Local name of this node.
456: */
457: public String getLocalNameFromExpandedNameID(int ExpandedNameID);
458:
459: /**
460: * Given an expanded-name ID, return the namespace URI part.
461: *
462: * @param ExpandedNameID an ID that represents an expanded-name.
463: * @return String URI value of this node's namespace, or null if no
464: * namespace was resolved.
465: */
466: public String getNamespaceFromExpandedNameID(int ExpandedNameID);
467:
468: /**
469: * Given a node handle, return its DOM-style node name. This will
470: * include names such as #text or #document.
471: *
472: * @param nodeHandle the id of the node.
473: * @return String Name of this node, which may be an empty string.
474: * %REVIEW% Document when empty string is possible...
475: */
476: public String getNodeName(int nodeHandle);
477:
478: /**
479: * Given a node handle, return the XPath node name. This should be
480: * the name as described by the XPath data model, NOT the DOM-style
481: * name.
482: *
483: * @param nodeHandle the id of the node.
484: * @return String Name of this node.
485: */
486: public String getNodeNameX(int nodeHandle);
487:
488: /**
489: * Given a node handle, return its DOM-style localname.
490: * (As defined in Namespaces, this is the portion of the name after the
491: * prefix, if present, or the whole node name if no prefix exists)
492: *
493: * @param nodeHandle the id of the node.
494: * @return String Local name of this node.
495: */
496: public String getLocalName(int nodeHandle);
497:
498: /**
499: * Given a namespace handle, return the prefix that the namespace decl is
500: * mapping.
501: * Given a node handle, return the prefix used to map to the namespace.
502: * (As defined in Namespaces, this is the portion of the name before any
503: * colon character).
504: *
505: * <p> %REVIEW% Are you sure you want "" for no prefix? </p>
506: *
507: * @param nodeHandle the id of the node.
508: * @return String prefix of this node's name, or "" if no explicit
509: * namespace prefix was given.
510: */
511: public String getPrefix(int nodeHandle);
512:
513: /**
514: * Given a node handle, return its DOM-style namespace URI
515: * (As defined in Namespaces, this is the declared URI which this node's
516: * prefix -- or default in lieu thereof -- was mapped to.)
517: * @param nodeHandle the id of the node.
518: * @return String URI value of this node's namespace, or null if no
519: * namespace was resolved.
520: */
521: public String getNamespaceURI(int nodeHandle);
522:
523: /**
524: * Given a node handle, return its node value. This is mostly
525: * as defined by the DOM, but may ignore some conveniences.
526: * <p>
527: * @param nodeHandle The node id.
528: * @return String Value of this node, or null if not
529: * meaningful for this node type.
530: */
531: public String getNodeValue(int nodeHandle);
532:
533: /**
534: * Given a node handle, return its DOM-style node type.
535: *
536: * <p>%REVIEW% Generally, returning short is false economy. Return int?</p>
537: *
538: * @param nodeHandle The node id.
539: * @return int Node type, as per the DOM's Node._NODE constants.
540: */
541: public short getNodeType(int nodeHandle);
542:
543: /**
544: * Get the depth level of this node in the tree (equals 1 for
545: * a parentless node).
546: *
547: * @param nodeHandle The node id.
548: * @return the number of ancestors, plus one
549: * @xsl.usage internal
550: */
551: public short getLevel(int nodeHandle);
552:
553: // ============== Document query functions ==============
554:
555: /**
556: * Tests whether DTM DOM implementation implements a specific feature and
557: * that feature is supported by this node.
558: * @param feature The name of the feature to test.
559: * @param version This is the version number of the feature to test.
560: * If the version is not
561: * specified, supporting any version of the feature will cause the
562: * method to return <code>true</code>.
563: * @return Returns <code>true</code> if the specified feature is
564: * supported on this node, <code>false</code> otherwise.
565: */
566: public boolean isSupported(String feature, String version);
567:
568: /**
569: * Return the base URI of the document entity. If it is not known
570: * (because the document was parsed from a socket connection or from
571: * standard input, for example), the value of this property is unknown.
572: *
573: * @return the document base URI String object or null if unknown.
574: */
575: public String getDocumentBaseURI();
576:
577: /**
578: * Set the base URI of the document entity.
579: *
580: * @param baseURI the document base URI String object or null if unknown.
581: */
582: public void setDocumentBaseURI(String baseURI);
583:
584: /**
585: * Return the system identifier of the document entity. If
586: * it is not known, the value of this property is null.
587: *
588: * @param nodeHandle The node id, which can be any valid node handle.
589: * @return the system identifier String object or null if unknown.
590: */
591: public String getDocumentSystemIdentifier(int nodeHandle);
592:
593: /**
594: * Return the name of the character encoding scheme
595: * in which the document entity is expressed.
596: *
597: * @param nodeHandle The node id, which can be any valid node handle.
598: * @return the document encoding String object.
599: */
600: public String getDocumentEncoding(int nodeHandle);
601:
602: /**
603: * Return an indication of the standalone status of the document,
604: * either "yes" or "no". This property is derived from the optional
605: * standalone document declaration in the XML declaration at the
606: * beginning of the document entity, and has no value if there is no
607: * standalone document declaration.
608: *
609: * @param nodeHandle The node id, which can be any valid node handle.
610: * @return the document standalone String object, either "yes", "no", or null.
611: */
612: public String getDocumentStandalone(int nodeHandle);
613:
614: /**
615: * Return a string representing the XML version of the document. This
616: * property is derived from the XML declaration optionally present at the
617: * beginning of the document entity, and has no value if there is no XML
618: * declaration.
619: *
620: * @param documentHandle the document handle
621: * @return the document version String object
622: */
623: public String getDocumentVersion(int documentHandle);
624:
625: /**
626: * Return an indication of
627: * whether the processor has read the complete DTD. Its value is a
628: * boolean. If it is false, then certain properties (indicated in their
629: * descriptions below) may be unknown. If it is true, those properties
630: * are never unknown.
631: *
632: * @return <code>true</code> if all declarations were processed;
633: * <code>false</code> otherwise.
634: */
635: public boolean getDocumentAllDeclarationsProcessed();
636:
637: /**
638: * A document type declaration information item has the following properties:
639: *
640: * 1. [system identifier] The system identifier of the external subset, if
641: * it exists. Otherwise this property has no value.
642: *
643: * @return the system identifier String object, or null if there is none.
644: */
645: public String getDocumentTypeDeclarationSystemIdentifier();
646:
647: /**
648: * Return the public identifier of the external subset,
649: * normalized as described in 4.2.2 External Entities [XML]. If there is
650: * no external subset or if it has no public identifier, this property
651: * has no value.
652: *
653: * @return the public identifier String object, or null if there is none.
654: */
655: public String getDocumentTypeDeclarationPublicIdentifier();
656:
657: /**
658: * Returns the <code>Element</code> whose <code>ID</code> is given by
659: * <code>elementId</code>. If no such element exists, returns
660: * <code>DTM.NULL</code>. Behavior is not defined if more than one element
661: * has this <code>ID</code>. Attributes (including those
662: * with the name "ID") are not of type ID unless so defined by DTD/Schema
663: * information available to the DTM implementation.
664: * Implementations that do not know whether attributes are of type ID or
665: * not are expected to return <code>DTM.NULL</code>.
666: *
667: * <p>%REVIEW% Presumably IDs are still scoped to a single document,
668: * and this operation searches only within a single document, right?
669: * Wouldn't want collisions between DTMs in the same process.</p>
670: *
671: * @param elementId The unique <code>id</code> value for an element.
672: * @return The handle of the matching element.
673: */
674: public int getElementById(String elementId);
675:
676: /**
677: * The getUnparsedEntityURI function returns the URI of the unparsed
678: * entity with the specified name in the same document as the context
679: * node (see [3.3 Unparsed Entities]). It returns the empty string if
680: * there is no such entity.
681: * <p>
682: * XML processors may choose to use the System Identifier (if one
683: * is provided) to resolve the entity, rather than the URI in the
684: * Public Identifier. The details are dependent on the processor, and
685: * we would have to support some form of plug-in resolver to handle
686: * this properly. Currently, we simply return the System Identifier if
687: * present, and hope that it a usable URI or that our caller can
688: * map it to one.
689: * %REVIEW% Resolve Public Identifiers... or consider changing function name.
690: * <p>
691: * If we find a relative URI
692: * reference, XML expects it to be resolved in terms of the base URI
693: * of the document. The DOM doesn't do that for us, and it isn't
694: * entirely clear whether that should be done here; currently that's
695: * pushed up to a higher level of our application. (Note that DOM Level
696: * 1 didn't store the document's base URI.)
697: * %REVIEW% Consider resolving Relative URIs.
698: * <p>
699: * (The DOM's statement that "An XML processor may choose to
700: * completely expand entities before the structure model is passed
701: * to the DOM" refers only to parsed entities, not unparsed, and hence
702: * doesn't affect this function.)
703: *
704: * @param name A string containing the Entity Name of the unparsed
705: * entity.
706: *
707: * @return String containing the URI of the Unparsed Entity, or an
708: * empty string if no such entity exists.
709: */
710: public String getUnparsedEntityURI(String name);
711:
712: // ============== Boolean methods ================
713:
714: /**
715: * Return true if the xsl:strip-space or xsl:preserve-space was processed
716: * during construction of the document contained in this DTM.
717: *
718: * NEEDSDOC ($objectName$) @return
719: */
720: public boolean supportsPreStripping();
721:
722: /**
723: * Figure out whether nodeHandle2 should be considered as being later
724: * in the document than nodeHandle1, in Document Order as defined
725: * by the XPath model. This may not agree with the ordering defined
726: * by other XML applications.
727: * <p>
728: * There are some cases where ordering isn't defined, and neither are
729: * the results of this function -- though we'll generally return true.
730: * <p>
731: * %REVIEW% Make sure this does the right thing with attribute nodes!!!
732: * <p>
733: * %REVIEW% Consider renaming for clarity. Perhaps isDocumentOrder(a,b)?
734: *
735: * @param firstNodeHandle DOM Node to perform position comparison on.
736: * @param secondNodeHandle DOM Node to perform position comparison on.
737: *
738: * @return false if secondNode comes before firstNode, otherwise return true.
739: * You can think of this as
740: * <code>(firstNode.documentOrderPosition <= secondNode.documentOrderPosition)</code>.
741: */
742: public boolean isNodeAfter(int firstNodeHandle, int secondNodeHandle);
743:
744: /**
745: * 2. [element content whitespace] A boolean indicating whether a
746: * text node represents white space appearing within element content
747: * (see [XML], 2.10 "White Space Handling"). Note that validating
748: * XML processors are required by XML 1.0 to provide this
749: * information... but that DOM Level 2 did not support it, since it
750: * depends on knowledge of the DTD which DOM2 could not guarantee
751: * would be available.
752: * <p>
753: * If there is no declaration for the containing element, an XML
754: * processor must assume that the whitespace could be meaningful and
755: * return false. If no declaration has been read, but the [all
756: * declarations processed] property of the document information item
757: * is false (so there may be an unread declaration), then the value
758: * of this property is indeterminate for white space characters and
759: * should probably be reported as false. It is always false for text
760: * nodes that contain anything other than (or in addition to) white
761: * space.
762: * <p>
763: * Note too that it always returns false for non-Text nodes.
764: * <p>
765: * %REVIEW% Joe wants to rename this isWhitespaceInElementContent() for clarity
766: *
767: * @param nodeHandle the node ID.
768: * @return <code>true</code> if the node definitely represents whitespace in
769: * element content; <code>false</code> otherwise.
770: */
771: public boolean isCharacterElementContentWhitespace(int nodeHandle);
772:
773: /**
774: * 10. [all declarations processed] This property is not strictly speaking
775: * part of the infoset of the document. Rather it is an indication of
776: * whether the processor has read the complete DTD. Its value is a
777: * boolean. If it is false, then certain properties (indicated in their
778: * descriptions below) may be unknown. If it is true, those properties
779: * are never unknown.
780: *
781: * @param documentHandle A node handle that must identify a document.
782: * @return <code>true</code> if all declarations were processed;
783: * <code>false</code> otherwise.
784: */
785: public boolean isDocumentAllDeclarationsProcessed(int documentHandle);
786:
787: /**
788: * 5. [specified] A flag indicating whether this attribute was actually
789: * specified in the start-tag of its element, or was defaulted from the
790: * DTD (or schema).
791: *
792: * @param attributeHandle The attribute handle
793: * @return <code>true</code> if the attribute was specified;
794: * <code>false</code> if it was defaulted or the handle doesn't
795: * refer to an attribute node.
796: */
797: public boolean isAttributeSpecified(int attributeHandle);
798:
799: // ========== Direct SAX Dispatch, for optimization purposes ========
800:
801: /**
802: * Directly call the
803: * characters method on the passed ContentHandler for the
804: * string-value of the given node (see http://www.w3.org/TR/xpath#data-model
805: * for the definition of a node's string-value). Multiple calls to the
806: * ContentHandler's characters methods may well occur for a single call to
807: * this method.
808: *
809: * @param nodeHandle The node ID.
810: * @param ch A non-null reference to a ContentHandler.
811: * @param normalize true if the content should be normalized according to
812: * the rules for the XPath
813: * <a href="http://www.w3.org/TR/xpath#function-normalize-space">normalize-space</a>
814: * function.
815: *
816: * @throws org.xml.sax.SAXException
817: */
818: public void dispatchCharactersEvents(int nodeHandle,
819: org.xml.sax.ContentHandler ch, boolean normalize)
820: throws org.xml.sax.SAXException;
821:
822: /**
823: * Directly create SAX parser events representing the XML content of
824: * a DTM subtree. This is a "serialize" operation.
825: *
826: * @param nodeHandle The node ID.
827: * @param ch A non-null reference to a ContentHandler.
828: *
829: * @throws org.xml.sax.SAXException
830: */
831: public void dispatchToEvents(int nodeHandle,
832: org.xml.sax.ContentHandler ch)
833: throws org.xml.sax.SAXException;
834:
835: /**
836: * Return an DOM node for the given node.
837: *
838: * @param nodeHandle The node ID.
839: *
840: * @return A node representation of the DTM node.
841: */
842: public org.w3c.dom.Node getNode(int nodeHandle);
843:
844: // ==== Construction methods (may not be supported by some implementations!) =====
845: // %REVIEW% What response occurs if not supported?
846:
847: /**
848: * @return true iff we're building this model incrementally (eg
849: * we're partnered with a CoroutineParser) and thus require that the
850: * transformation and the parse run simultaneously. Guidance to the
851: * DTMManager.
852: */
853: public boolean needsTwoThreads();
854:
855: // %REVIEW% Do these appends make any sense, should we support a
856: // wider set of methods (like the "append" methods in the
857: // current DTMDocumentImpl draft), or should we just support SAX
858: // listener interfaces? Should it be a separate interface to
859: // make that distinction explicit?
860:
861: /**
862: * Return this DTM's content handler, if it has one.
863: *
864: * @return null if this model doesn't respond to SAX events.
865: */
866: public org.xml.sax.ContentHandler getContentHandler();
867:
868: /**
869: * Return this DTM's lexical handler, if it has one.
870: *
871: * %REVIEW% Should this return null if constrution already done/begun?
872: *
873: * @return null if this model doesn't respond to lexical SAX events.
874: */
875: public org.xml.sax.ext.LexicalHandler getLexicalHandler();
876:
877: /**
878: * Return this DTM's EntityResolver, if it has one.
879: *
880: * @return null if this model doesn't respond to SAX entity ref events.
881: */
882: public org.xml.sax.EntityResolver getEntityResolver();
883:
884: /**
885: * Return this DTM's DTDHandler, if it has one.
886: *
887: * @return null if this model doesn't respond to SAX dtd events.
888: */
889: public org.xml.sax.DTDHandler getDTDHandler();
890:
891: /**
892: * Return this DTM's ErrorHandler, if it has one.
893: *
894: * @return null if this model doesn't respond to SAX error events.
895: */
896: public org.xml.sax.ErrorHandler getErrorHandler();
897:
898: /**
899: * Return this DTM's DeclHandler, if it has one.
900: *
901: * @return null if this model doesn't respond to SAX Decl events.
902: */
903: public org.xml.sax.ext.DeclHandler getDeclHandler();
904:
905: /**
906: * Append a child to "the end of the document". Please note that
907: * the node is always cloned in a base DTM, since our basic behavior
908: * is immutable so nodes can't be removed from their previous
909: * location.
910: *
911: * <p> %REVIEW% DTM maintains an insertion cursor which
912: * performs a depth-first tree walk as nodes come in, and this operation
913: * is really equivalent to:
914: * insertionCursor.appendChild(document.importNode(newChild)))
915: * where the insert point is the last element that was appended (or
916: * the last one popped back to by an end-element operation).</p>
917: *
918: * @param newChild Must be a valid new node handle.
919: * @param clone true if the child should be cloned into the document.
920: * @param cloneDepth if the clone argument is true, specifies that the
921: * clone should include all it's children.
922: */
923: public void appendChild(int newChild, boolean clone,
924: boolean cloneDepth);
925:
926: /**
927: * Append a text node child that will be constructed from a string,
928: * to the end of the document. Behavior is otherwise like appendChild().
929: *
930: * @param str Non-null reference to a string.
931: */
932: public void appendTextChild(String str);
933:
934: /**
935: * Get the location of a node in the source document.
936: *
937: * @param node an <code>int</code> value
938: * @return a <code>SourceLocator</code> value or null if no location
939: * is available
940: */
941: public SourceLocator getSourceLocatorFor(int node);
942:
943: /**
944: * As the DTM is registered with the DTMManager, this method
945: * will be called. This will give the DTM implementation a
946: * chance to initialize any subsystems that are required to
947: * build the DTM
948: */
949: public void documentRegistration();
950:
951: /**
952: * As documents are released from the DTMManager, the DTM implementation
953: * will be notified of the event. This will allow the DTM implementation
954: * to shutdown any subsystem activity that may of been assoiated with
955: * the active DTM Implementation.
956: */
957:
958: public void documentRelease();
959:
960: /**
961: * Migrate a DTM built with an old DTMManager to a new DTMManager.
962: * After the migration, the new DTMManager will treat the DTM as
963: * one that is built by itself.
964: * This is used to support DTM sharing between multiple transformations.
965: * @param manager the DTMManager
966: */
967: public void migrateTo(DTMManager manager);
968: }
|