1 /*
   2  * reserved comment block
   3  * DO NOT REMOVE OR ALTER!
   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.xml.internal.dtm;
  23 
  24 import javax.xml.transform.SourceLocator;
  25 
  26 import com.sun.org.apache.xml.internal.utils.XMLString;
  27 
  28 /**
  29  * <code>DTM</code> is an XML document model expressed as a table
  30  * rather than an object tree. It attempts to provide an interface to
  31  * a parse tree that has very little object creation. (DTM
  32  * implementations may also support incremental construction of the
  33  * model, but that's hidden from the DTM API.)
  34  *
  35  * <p>Nodes in the DTM are identified by integer "handles".  A handle must
  36  * be unique within a process, and carries both node identification and
  37  * document identification.  It must be possible to compare two handles
  38  * (and thus their nodes) for identity with "==".</p>
  39  *
  40  * <p>Namespace URLs, local-names, and expanded-names can all be
  41  * represented by and tested as integer ID values.  An expanded name
  42  * represents (and may or may not directly contain) a combination of
  43  * the URL ID, and the local-name ID.  Note that the namespace URL id
  44  * can be 0, which should have the meaning that the namespace is null.
  45  * For consistancy, zero should not be used for a local-name index. </p>
  46  *
  47  * <p>Text content of a node is represented by an index and length,
  48  * permitting efficient storage such as a shared FastStringBuffer.</p>
  49  *
  50  * <p>The model of the tree, as well as the general navigation model,
  51  * is that of XPath 1.0, for the moment.  The model will eventually be
  52  * adapted to match the XPath 2.0 data model, XML Schema, and
  53  * InfoSet.</p>
  54  *
  55  * <p>DTM does _not_ directly support the W3C's Document Object
  56  * Model. However, it attempts to come close enough that an
  57  * implementation of DTM can be created that wraps a DOM and vice
  58  * versa.</p>
  59  *
  60  * <p><strong>Please Note:</strong> The DTM API is still
  61  * <strong>Subject To Change.</strong> This wouldn't affect most
  62  * users, but might require updating some extensions.</p>
  63  *
  64  * <p> The largest change being contemplated is a reconsideration of
  65  * the Node Handle representation.  We are still not entirely sure
  66  * that an integer packed with two numeric subfields is really the
  67  * best solution. It has been suggested that we move up to a Long, to
  68  * permit more nodes per document without having to reduce the number
  69  * of slots in the DTMManager. There's even been a proposal that we
  70  * replace these integers with "cursor" objects containing the
  71  * internal node id and a pointer to the actual DTM object; this might
  72  * reduce the need to continuously consult the DTMManager to retrieve
  73  * the latter, and might provide a useful "hook" back into normal Java
  74  * heap management.  But changing this datatype would have huge impact
  75  * on Xalan's internals -- especially given Java's lack of C-style
  76  * typedefs -- so we won't cut over unless we're convinced the new
  77  * solution really would be an improvement!</p>
  78  * */
  79 public interface DTM
  80 {
  81 
  82   /**
  83    * Null node handles are represented by this value.
  84    */
  85   public static final int NULL = -1;
  86 
  87   // These nodeType mnemonics and values are deliberately the same as those
  88   // used by the DOM, for convenient mapping
  89   //
  90   // %REVIEW% Should we actually define these as initialized to,
  91   // eg. org.w3c.dom.Document.ELEMENT_NODE?
  92 
  93   /**
  94    * The node is a <code>Root</code>.
  95    */
  96   public static final short ROOT_NODE = 0;
  97 
  98   /**
  99    * The node is an <code>Element</code>.
 100    */
 101   public static final short ELEMENT_NODE = 1;
 102 
 103   /**
 104    * The node is an <code>Attr</code>.
 105    */
 106   public static final short ATTRIBUTE_NODE = 2;
 107 
 108   /**
 109    * The node is a <code>Text</code> node.
 110    */
 111   public static final short TEXT_NODE = 3;
 112 
 113   /**
 114    * The node is a <code>CDATASection</code>.
 115    */
 116   public static final short CDATA_SECTION_NODE = 4;
 117 
 118   /**
 119    * The node is an <code>EntityReference</code>.
 120    */
 121   public static final short ENTITY_REFERENCE_NODE = 5;
 122 
 123   /**
 124    * The node is an <code>Entity</code>.
 125    */
 126   public static final short ENTITY_NODE = 6;
 127 
 128   /**
 129    * The node is a <code>ProcessingInstruction</code>.
 130    */
 131   public static final short PROCESSING_INSTRUCTION_NODE = 7;
 132 
 133   /**
 134    * The node is a <code>Comment</code>.
 135    */
 136   public static final short COMMENT_NODE = 8;
 137 
 138   /**
 139    * The node is a <code>Document</code>.
 140    */
 141   public static final short DOCUMENT_NODE = 9;
 142 
 143   /**
 144    * The node is a <code>DocumentType</code>.
 145    */
 146   public static final short DOCUMENT_TYPE_NODE = 10;
 147 
 148   /**
 149    * The node is a <code>DocumentFragment</code>.
 150    */
 151   public static final short DOCUMENT_FRAGMENT_NODE = 11;
 152 
 153   /**
 154    * The node is a <code>Notation</code>.
 155    */
 156   public static final short NOTATION_NODE = 12;
 157 
 158   /**
 159    * The node is a <code>namespace node</code>. Note that this is not
 160    * currently a node type defined by the DOM API.
 161    */
 162   public static final short NAMESPACE_NODE = 13;
 163 
 164   /**
 165    * The number of valid nodetypes.
 166    */
 167   public static final short  NTYPES = 14;
 168 
 169   // ========= DTM Implementation Control Functions. ==============
 170   // %TBD% RETIRED -- do via setFeature if needed. Remove from impls.
 171   // public void setParseBlockSize(int blockSizeSuggestion);
 172 
 173   /**
 174    * Set an implementation dependent feature.
 175    * <p>
 176    * %REVIEW% Do we really expect to set features on DTMs?
 177    *
 178    * @param featureId A feature URL.
 179    * @param state true if this feature should be on, false otherwise.
 180    */
 181   public void setFeature(String featureId, boolean state);
 182 
 183   /**
 184    * Set a run time property for this DTM instance.
 185    *
 186    * @param property a <code>String</code> value
 187    * @param value an <code>Object</code> value
 188    */
 189   public void setProperty(String property, Object value);
 190 
 191   // ========= Document Navigation Functions =========
 192 
 193   /**
 194    * This returns a stateless "traverser", that can navigate over an
 195    * XPath axis, though not in document order.
 196    *
 197    * @param axis One of Axes.ANCESTORORSELF, etc.
 198    *
 199    * @return A DTMAxisIterator, or null if the givin axis isn't supported.
 200    */
 201   public DTMAxisTraverser getAxisTraverser(final int axis);
 202 
 203   /**
 204    * This is a shortcut to the iterators that implement
 205    * XPath axes.
 206    * Returns a bare-bones iterator that must be initialized
 207    * with a start node (using iterator.setStartNode()).
 208    *
 209    * @param axis One of Axes.ANCESTORORSELF, etc.
 210    *
 211    * @return A DTMAxisIterator, or null if the givin axis isn't supported.
 212    */
 213   public DTMAxisIterator getAxisIterator(final int axis);
 214 
 215   /**
 216    * Get an iterator that can navigate over an XPath Axis, predicated by
 217    * the extended type ID.
 218    *
 219    * @param axis
 220    * @param type An extended type ID.
 221    *
 222    * @return A DTMAxisIterator, or null if the givin axis isn't supported.
 223    */
 224   public DTMAxisIterator getTypedAxisIterator(final int axis, final int type);
 225 
 226   /**
 227    * Given a node handle, test if it has child nodes.
 228    * <p> %REVIEW% This is obviously useful at the DOM layer, where it
 229    * would permit testing this without having to create a proxy
 230    * node. It's less useful in the DTM API, where
 231    * (dtm.getFirstChild(nodeHandle)!=DTM.NULL) is just as fast and
 232    * almost as self-evident. But it's a convenience, and eases porting
 233    * of DOM code to DTM.  </p>
 234    *
 235    * @param nodeHandle int Handle of the node.
 236    * @return int true if the given node has child nodes.
 237    */
 238   public boolean hasChildNodes(int nodeHandle);
 239 
 240   /**
 241    * Given a node handle, get the handle of the node's first child.
 242    *
 243    * @param nodeHandle int Handle of the node.
 244    * @return int DTM node-number of first child,
 245    * or DTM.NULL to indicate none exists.
 246    */
 247   public int getFirstChild(int nodeHandle);
 248 
 249   /**
 250    * Given a node handle, get the handle of the node's last child.
 251    *
 252    * @param nodeHandle int Handle of the node.
 253    * @return int Node-number of last child,
 254    * or DTM.NULL to indicate none exists.
 255    */
 256   public int getLastChild(int nodeHandle);
 257 
 258   /**
 259    * Retrieves an attribute node by local name and namespace URI
 260    *
 261    * %TBD% Note that we currently have no way to support
 262    * the DOM's old getAttribute() call, which accesses only the qname.
 263    *
 264    * @param elementHandle Handle of the node upon which to look up this attribute.
 265    * @param namespaceURI The namespace URI of the attribute to
 266    *   retrieve, or null.
 267    * @param name The local name of the attribute to
 268    *   retrieve.
 269    * @return The attribute node handle with the specified name (
 270    *   <code>nodeName</code>) or <code>DTM.NULL</code> if there is no such
 271    *   attribute.
 272    */
 273   public int getAttributeNode(int elementHandle, String namespaceURI,
 274                               String name);
 275 
 276   /**
 277    * Given a node handle, get the index of the node's first attribute.
 278    *
 279    * @param nodeHandle int Handle of the node.
 280    * @return Handle of first attribute, or DTM.NULL to indicate none exists.
 281    */
 282   public int getFirstAttribute(int nodeHandle);
 283 
 284   /**
 285    * Given a node handle, get the index of the node's first namespace node.
 286    *
 287    * @param nodeHandle handle to node, which should probably be an element
 288    *                   node, but need not be.
 289    *
 290    * @param inScope true if all namespaces in scope should be
 291    *                   returned, false if only the node's own
 292    *                   namespace declarations should be returned.
 293    * @return handle of first namespace,
 294    * or DTM.NULL to indicate none exists.
 295    */
 296   public int getFirstNamespaceNode(int nodeHandle, boolean inScope);
 297 
 298   /**
 299    * Given a node handle, advance to its next sibling.
 300    * @param nodeHandle int Handle of the node.
 301    * @return int Node-number of next sibling,
 302    * or DTM.NULL to indicate none exists.
 303    */
 304   public int getNextSibling(int nodeHandle);
 305 
 306   /**
 307    * Given a node handle, find its preceeding sibling.
 308    * WARNING: DTM implementations may be asymmetric; in some,
 309    * this operation has been resolved by search, and is relatively expensive.
 310    *
 311    * @param nodeHandle the id of the node.
 312    * @return int Node-number of the previous sib,
 313    * or DTM.NULL to indicate none exists.
 314    */
 315   public int getPreviousSibling(int nodeHandle);
 316 
 317   /**
 318    * Given a node handle, advance to the next attribute. If an
 319    * element, we advance to its first attribute; if an attr, we advance to
 320    * the next attr of the same element.
 321    *
 322    * @param nodeHandle int Handle of the node.
 323    * @return int DTM node-number of the resolved attr,
 324    * or DTM.NULL to indicate none exists.
 325    */
 326   public int getNextAttribute(int nodeHandle);
 327 
 328   /**
 329    * Given a namespace handle, advance to the next namespace in the same scope
 330    * (local or local-plus-inherited, as selected by getFirstNamespaceNode)
 331    *
 332    * @param baseHandle handle to original node from where the first child
 333    * was relative to (needed to return nodes in document order).
 334    * @param namespaceHandle handle to node which must be of type
 335    * NAMESPACE_NODE.
 336    * NEEDSDOC @param inScope
 337    * @return handle of next namespace,
 338    * or DTM.NULL to indicate none exists.
 339    */
 340   public int getNextNamespaceNode(int baseHandle, int namespaceHandle,
 341                                   boolean inScope);
 342 
 343   /**
 344    * Given a node handle, find its parent node.
 345    *
 346    * @param nodeHandle the id of the node.
 347    * @return int Node handle of parent,
 348    * or DTM.NULL to indicate none exists.
 349    */
 350   public int getParent(int nodeHandle);
 351 
 352   /**
 353    * Given a DTM which contains only a single document,
 354    * find the Node Handle of the  Document node. Note
 355    * that if the DTM is configured so it can contain multiple
 356    * documents, this call will return the Document currently
 357    * under construction -- but may return null if it's between
 358    * documents. Generally, you should use getOwnerDocument(nodeHandle)
 359    * or getDocumentRoot(nodeHandle) instead.
 360    *
 361    * @return int Node handle of document, or DTM.NULL if a shared DTM
 362    * can not tell us which Document is currently active.
 363    */
 364   public int getDocument();
 365 
 366   /**
 367    * Given a node handle, find the owning document node. This version mimics
 368    * the behavior of the DOM call by the same name.
 369    *
 370    * @param nodeHandle the id of the node.
 371    * @return int Node handle of owning document, or DTM.NULL if the node was
 372    * a Document.
 373    * @see #getDocumentRoot(int nodeHandle)
 374    */
 375   public int getOwnerDocument(int nodeHandle);
 376 
 377   /**
 378    * Given a node handle, find the owning document node.
 379    *
 380    * @param nodeHandle the id of the node.
 381    * @return int Node handle of owning document, or the node itself if it was
 382    * a Document. (Note difference from DOM, where getOwnerDocument returns
 383    * null for the Document node.)
 384    * @see #getOwnerDocument(int nodeHandle)
 385    */
 386   public int getDocumentRoot(int nodeHandle);
 387 
 388   /**
 389    * Get the string-value of a node as a String object
 390    * (see http://www.w3.org/TR/xpath#data-model
 391    * for the definition of a node's string-value).
 392    *
 393    * @param nodeHandle The node ID.
 394    *
 395    * @return A string object that represents the string-value of the given node.
 396    */
 397   public XMLString getStringValue(int nodeHandle);
 398 
 399   /**
 400    * Get number of character array chunks in
 401    * the string-value of a node.
 402    * (see http://www.w3.org/TR/xpath#data-model
 403    * for the definition of a node's string-value).
 404    * Note that a single text node may have multiple text chunks.
 405    *
 406    * @param nodeHandle The node ID.
 407    *
 408    * @return number of character array chunks in
 409    *         the string-value of a node.
 410    */
 411   public int getStringValueChunkCount(int nodeHandle);
 412 
 413   /**
 414    * Get a character array chunk in the string-value of a node.
 415    * (see http://www.w3.org/TR/xpath#data-model
 416    * for the definition of a node's string-value).
 417    * Note that a single text node may have multiple text chunks.
 418    *
 419    * @param nodeHandle The node ID.
 420    * @param chunkIndex Which chunk to get.
 421    * @param startAndLen  A two-integer array which, upon return, WILL
 422    * BE FILLED with values representing the chunk's start position
 423    * within the returned character buffer and the length of the chunk.
 424    * @return The character array buffer within which the chunk occurs,
 425    * setting startAndLen's contents as a side-effect.
 426    */
 427   public char[] getStringValueChunk(int nodeHandle, int chunkIndex,
 428                                     int[] startAndLen);
 429 
 430   /**
 431    * Given a node handle, return an ID that represents the node's expanded name.
 432    *
 433    * @param nodeHandle The handle to the node in question.
 434    *
 435    * @return the expanded-name id of the node.
 436    */
 437   public int getExpandedTypeID(int nodeHandle);
 438 
 439   /**
 440    * Given an expanded name, return an ID.  If the expanded-name does not
 441    * exist in the internal tables, the entry will be created, and the ID will
 442    * be returned.  Any additional nodes that are created that have this
 443    * expanded name will use this ID.
 444    *
 445    * NEEDSDOC @param namespace
 446    * NEEDSDOC @param localName
 447    * NEEDSDOC @param type
 448    *
 449    * @return the expanded-name id of the node.
 450    */
 451   public int getExpandedTypeID(String namespace, String localName, int type);
 452 
 453   /**
 454    * Given an expanded-name ID, return the local name part.
 455    *
 456    * @param ExpandedNameID an ID that represents an expanded-name.
 457    * @return String Local name of this node.
 458    */
 459   public String getLocalNameFromExpandedNameID(int ExpandedNameID);
 460 
 461   /**
 462    * Given an expanded-name ID, return the namespace URI part.
 463    *
 464    * @param ExpandedNameID an ID that represents an expanded-name.
 465    * @return String URI value of this node's namespace, or null if no
 466    * namespace was resolved.
 467    */
 468   public String getNamespaceFromExpandedNameID(int ExpandedNameID);
 469 
 470   /**
 471    * Given a node handle, return its DOM-style node name. This will
 472    * include names such as #text or #document.
 473    *
 474    * @param nodeHandle the id of the node.
 475    * @return String Name of this node, which may be an empty string.
 476    * %REVIEW% Document when empty string is possible...
 477    */
 478   public String getNodeName(int nodeHandle);
 479 
 480   /**
 481    * Given a node handle, return the XPath node name.  This should be
 482    * the name as described by the XPath data model, NOT the DOM-style
 483    * name.
 484    *
 485    * @param nodeHandle the id of the node.
 486    * @return String Name of this node.
 487    */
 488   public String getNodeNameX(int nodeHandle);
 489 
 490   /**
 491    * Given a node handle, return its DOM-style localname.
 492    * (As defined in Namespaces, this is the portion of the name after the
 493    * prefix, if present, or the whole node name if no prefix exists)
 494    *
 495    * @param nodeHandle the id of the node.
 496    * @return String Local name of this node.
 497    */
 498   public String getLocalName(int nodeHandle);
 499 
 500   /**
 501    * Given a namespace handle, return the prefix that the namespace decl is
 502    * mapping.
 503    * Given a node handle, return the prefix used to map to the namespace.
 504    * (As defined in Namespaces, this is the portion of the name before any
 505    * colon character).
 506    *
 507    * <p> %REVIEW% Are you sure you want "" for no prefix?  </p>
 508    *
 509    * @param nodeHandle the id of the node.
 510    * @return String prefix of this node's name, or "" if no explicit
 511    * namespace prefix was given.
 512    */
 513   public String getPrefix(int nodeHandle);
 514 
 515   /**
 516    * Given a node handle, return its DOM-style namespace URI
 517    * (As defined in Namespaces, this is the declared URI which this node's
 518    * prefix -- or default in lieu thereof -- was mapped to.)
 519    * @param nodeHandle the id of the node.
 520    * @return String URI value of this node's namespace, or null if no
 521    * namespace was resolved.
 522    */
 523   public String getNamespaceURI(int nodeHandle);
 524 
 525   /**
 526    * Given a node handle, return its node value. This is mostly
 527    * as defined by the DOM, but may ignore some conveniences.
 528    * <p>
 529    * @param nodeHandle The node id.
 530    * @return String Value of this node, or null if not
 531    * meaningful for this node type.
 532    */
 533   public String getNodeValue(int nodeHandle);
 534 
 535   /**
 536    * Given a node handle, return its DOM-style node type.
 537    *
 538    * <p>%REVIEW% Generally, returning short is false economy. Return int?</p>
 539    *
 540    * @param nodeHandle The node id.
 541    * @return int Node type, as per the DOM's Node._NODE constants.
 542    */
 543   public short getNodeType(int nodeHandle);
 544 
 545   /**
 546    * Get the depth level of this node in the tree (equals 1 for
 547    * a parentless node).
 548    *
 549    * @param nodeHandle The node id.
 550    * @return the number of ancestors, plus one
 551    * @xsl.usage internal
 552    */
 553   public short getLevel(int nodeHandle);
 554 
 555   // ============== Document query functions ==============
 556 
 557   /**
 558    * Tests whether DTM DOM implementation implements a specific feature and
 559    * that feature is supported by this node.
 560    * @param feature The name of the feature to test.
 561    * @param version This is the version number of the feature to test.
 562    *   If the version is not
 563    *   specified, supporting any version of the feature will cause the
 564    *   method to return <code>true</code>.
 565    * @return Returns <code>true</code> if the specified feature is
 566    *   supported on this node, <code>false</code> otherwise.
 567    */
 568   public boolean isSupported(String feature, String version);
 569 
 570   /**
 571    * Return the base URI of the document entity. If it is not known
 572    * (because the document was parsed from a socket connection or from
 573    * standard input, for example), the value of this property is unknown.
 574    *
 575    * @return the document base URI String object or null if unknown.
 576    */
 577   public String getDocumentBaseURI();
 578 
 579   /**
 580    * Set the base URI of the document entity.
 581    *
 582    * @param baseURI the document base URI String object or null if unknown.
 583    */
 584   public void setDocumentBaseURI(String baseURI);
 585 
 586   /**
 587    * Return the system identifier of the document entity. If
 588    * it is not known, the value of this property is null.
 589    *
 590    * @param nodeHandle The node id, which can be any valid node handle.
 591    * @return the system identifier String object or null if unknown.
 592    */
 593   public String getDocumentSystemIdentifier(int nodeHandle);
 594 
 595   /**
 596    * Return the name of the character encoding scheme
 597    *        in which the document entity is expressed.
 598    *
 599    * @param nodeHandle The node id, which can be any valid node handle.
 600    * @return the document encoding String object.
 601    */
 602   public String getDocumentEncoding(int nodeHandle);
 603 
 604   /**
 605    * Return an indication of the standalone status of the document,
 606    *        either "yes" or "no". This property is derived from the optional
 607    *        standalone document declaration in the XML declaration at the
 608    *        beginning of the document entity, and has no value if there is no
 609    *        standalone document declaration.
 610    *
 611    * @param nodeHandle The node id, which can be any valid node handle.
 612    * @return the document standalone String object, either "yes", "no", or null.
 613    */
 614   public String getDocumentStandalone(int nodeHandle);
 615 
 616   /**
 617    * Return a string representing the XML version of the document. This
 618    * property is derived from the XML declaration optionally present at the
 619    * beginning of the document entity, and has no value if there is no XML
 620    * declaration.
 621    *
 622    * @param documentHandle the document handle
 623    * @return the document version String object
 624    */
 625   public String getDocumentVersion(int documentHandle);
 626 
 627   /**
 628    * Return an indication of
 629    * whether the processor has read the complete DTD. Its value is a
 630    * boolean. If it is false, then certain properties (indicated in their
 631    * descriptions below) may be unknown. If it is true, those properties
 632    * are never unknown.
 633    *
 634    * @return <code>true</code> if all declarations were processed;
 635    *         <code>false</code> otherwise.
 636    */
 637   public boolean getDocumentAllDeclarationsProcessed();
 638 
 639   /**
 640    *   A document type declaration information item has the following properties:
 641    *
 642    *     1. [system identifier] The system identifier of the external subset, if
 643    *        it exists. Otherwise this property has no value.
 644    *
 645    * @return the system identifier String object, or null if there is none.
 646    */
 647   public String getDocumentTypeDeclarationSystemIdentifier();
 648 
 649   /**
 650    * Return the public identifier of the external subset,
 651    * normalized as described in 4.2.2 External Entities [XML]. If there is
 652    * no external subset or if it has no public identifier, this property
 653    * has no value.
 654    *
 655    * @return the public identifier String object, or null if there is none.
 656    */
 657   public String getDocumentTypeDeclarationPublicIdentifier();
 658 
 659   /**
 660    * Returns the <code>Element</code> whose <code>ID</code> is given by
 661    * <code>elementId</code>. If no such element exists, returns
 662    * <code>DTM.NULL</code>. Behavior is not defined if more than one element
 663    * has this <code>ID</code>. Attributes (including those
 664    * with the name "ID") are not of type ID unless so defined by DTD/Schema
 665    * information available to the DTM implementation.
 666    * Implementations that do not know whether attributes are of type ID or
 667    * not are expected to return <code>DTM.NULL</code>.
 668    *
 669    * <p>%REVIEW% Presumably IDs are still scoped to a single document,
 670    * and this operation searches only within a single document, right?
 671    * Wouldn't want collisions between DTMs in the same process.</p>
 672    *
 673    * @param elementId The unique <code>id</code> value for an element.
 674    * @return The handle of the matching element.
 675    */
 676   public int getElementById(String elementId);
 677 
 678   /**
 679    * The getUnparsedEntityURI function returns the URI of the unparsed
 680    * entity with the specified name in the same document as the context
 681    * node (see [3.3 Unparsed Entities]). It returns the empty string if
 682    * there is no such entity.
 683    * <p>
 684    * XML processors may choose to use the System Identifier (if one
 685    * is provided) to resolve the entity, rather than the URI in the
 686    * Public Identifier. The details are dependent on the processor, and
 687    * we would have to support some form of plug-in resolver to handle
 688    * this properly. Currently, we simply return the System Identifier if
 689    * present, and hope that it a usable URI or that our caller can
 690    * map it to one.
 691    * %REVIEW% Resolve Public Identifiers... or consider changing function name.
 692    * <p>
 693    * If we find a relative URI
 694    * reference, XML expects it to be resolved in terms of the base URI
 695    * of the document. The DOM doesn't do that for us, and it isn't
 696    * entirely clear whether that should be done here; currently that's
 697    * pushed up to a higher level of our application. (Note that DOM Level
 698    * 1 didn't store the document's base URI.)
 699    * %REVIEW% Consider resolving Relative URIs.
 700    * <p>
 701    * (The DOM's statement that "An XML processor may choose to
 702    * completely expand entities before the structure model is passed
 703    * to the DOM" refers only to parsed entities, not unparsed, and hence
 704    * doesn't affect this function.)
 705    *
 706    * @param name A string containing the Entity Name of the unparsed
 707    * entity.
 708    *
 709    * @return String containing the URI of the Unparsed Entity, or an
 710    * empty string if no such entity exists.
 711    */
 712   public String getUnparsedEntityURI(String name);
 713 
 714   // ============== Boolean methods ================
 715 
 716   /**
 717    * Return true if the xsl:strip-space or xsl:preserve-space was processed
 718    * during construction of the document contained in this DTM.
 719    *
 720    * NEEDSDOC ($objectName$) @return
 721    */
 722   public boolean supportsPreStripping();
 723 
 724   /**
 725    * Figure out whether nodeHandle2 should be considered as being later
 726    * in the document than nodeHandle1, in Document Order as defined
 727    * by the XPath model. This may not agree with the ordering defined
 728    * by other XML applications.
 729    * <p>
 730    * There are some cases where ordering isn't defined, and neither are
 731    * the results of this function -- though we'll generally return true.
 732    * <p>
 733    * %REVIEW% Make sure this does the right thing with attribute nodes!!!
 734    * <p>
 735    * %REVIEW% Consider renaming for clarity. Perhaps isDocumentOrder(a,b)?
 736    *
 737    * @param firstNodeHandle DOM Node to perform position comparison on.
 738    * @param secondNodeHandle DOM Node to perform position comparison on.
 739    *
 740    * @return false if secondNode comes before firstNode, otherwise return true.
 741    * You can think of this as
 742    * <code>(firstNode.documentOrderPosition &lt;= secondNode.documentOrderPosition)</code>.
 743    */
 744   public boolean isNodeAfter(int firstNodeHandle, int secondNodeHandle);
 745 
 746   /**
 747    * 2. [element content whitespace] A boolean indicating whether a
 748    * text node represents white space appearing within element content
 749    * (see [XML], 2.10 "White Space Handling").  Note that validating
 750    * XML processors are required by XML 1.0 to provide this
 751    * information... but that DOM Level 2 did not support it, since it
 752    * depends on knowledge of the DTD which DOM2 could not guarantee
 753    * would be available.
 754    * <p>
 755    * If there is no declaration for the containing element, an XML
 756    * processor must assume that the whitespace could be meaningful and
 757    * return false. If no declaration has been read, but the [all
 758    * declarations processed] property of the document information item
 759    * is false (so there may be an unread declaration), then the value
 760    * of this property is indeterminate for white space characters and
 761    * should probably be reported as false. It is always false for text
 762    * nodes that contain anything other than (or in addition to) white
 763    * space.
 764    * <p>
 765    * Note too that it always returns false for non-Text nodes.
 766    * <p>
 767    * %REVIEW% Joe wants to rename this isWhitespaceInElementContent() for clarity
 768    *
 769    * @param nodeHandle the node ID.
 770    * @return <code>true</code> if the node definitely represents whitespace in
 771    * element content; <code>false</code> otherwise.
 772    */
 773   public boolean isCharacterElementContentWhitespace(int nodeHandle);
 774 
 775   /**
 776    *    10. [all declarations processed] This property is not strictly speaking
 777    *        part of the infoset of the document. Rather it is an indication of
 778    *        whether the processor has read the complete DTD. Its value is a
 779    *        boolean. If it is false, then certain properties (indicated in their
 780    *        descriptions below) may be unknown. If it is true, those properties
 781    *        are never unknown.
 782    *
 783    * @param documentHandle A node handle that must identify a document.
 784    * @return <code>true</code> if all declarations were processed;
 785    *         <code>false</code> otherwise.
 786    */
 787   public boolean isDocumentAllDeclarationsProcessed(int documentHandle);
 788 
 789   /**
 790    *     5. [specified] A flag indicating whether this attribute was actually
 791    *        specified in the start-tag of its element, or was defaulted from the
 792    *        DTD (or schema).
 793    *
 794    * @param attributeHandle The attribute handle
 795    * @return <code>true</code> if the attribute was specified;
 796    *         <code>false</code> if it was defaulted or the handle doesn't
 797    *            refer to an attribute node.
 798    */
 799   public boolean isAttributeSpecified(int attributeHandle);
 800 
 801   // ========== Direct SAX Dispatch, for optimization purposes ========
 802 
 803   /**
 804    * Directly call the
 805    * characters method on the passed ContentHandler for the
 806    * string-value of the given node (see http://www.w3.org/TR/xpath#data-model
 807    * for the definition of a node's string-value). Multiple calls to the
 808    * ContentHandler's characters methods may well occur for a single call to
 809    * this method.
 810    *
 811    * @param nodeHandle The node ID.
 812    * @param ch A non-null reference to a ContentHandler.
 813    * @param normalize true if the content should be normalized according to
 814    * the rules for the XPath
 815    * <a href="http://www.w3.org/TR/xpath#function-normalize-space">normalize-space</a>
 816    * function.
 817    *
 818    * @throws org.xml.sax.SAXException
 819    */
 820   public void dispatchCharactersEvents(
 821     int nodeHandle, org.xml.sax.ContentHandler ch, boolean normalize)
 822       throws org.xml.sax.SAXException;
 823 
 824   /**
 825    * Directly create SAX parser events representing the XML content of
 826    * a DTM subtree. This is a "serialize" operation.
 827    *
 828    * @param nodeHandle The node ID.
 829    * @param ch A non-null reference to a ContentHandler.
 830    *
 831    * @throws org.xml.sax.SAXException
 832    */
 833   public void dispatchToEvents(int nodeHandle, org.xml.sax.ContentHandler ch)
 834     throws org.xml.sax.SAXException;
 835 
 836   /**
 837    * Return an DOM node for the given node.
 838    *
 839    * @param nodeHandle The node ID.
 840    *
 841    * @return A node representation of the DTM node.
 842    */
 843   public org.w3c.dom.Node getNode(int nodeHandle);
 844 
 845   // ==== Construction methods (may not be supported by some implementations!) =====
 846   // %REVIEW% What response occurs if not supported?
 847 
 848   /**
 849    * @return true iff we're building this model incrementally (eg
 850    * we're partnered with a CoroutineParser) and thus require that the
 851    * transformation and the parse run simultaneously. Guidance to the
 852    * DTMManager.
 853    */
 854   public boolean needsTwoThreads();
 855 
 856   // %REVIEW% Do these appends make any sense, should we support a
 857   // wider set of methods (like the "append" methods in the
 858   // current DTMDocumentImpl draft), or should we just support SAX
 859   // listener interfaces?  Should it be a separate interface to
 860   // make that distinction explicit?
 861 
 862   /**
 863    * Return this DTM's content handler, if it has one.
 864    *
 865    * @return null if this model doesn't respond to SAX events.
 866    */
 867   public org.xml.sax.ContentHandler getContentHandler();
 868 
 869   /**
 870    * Return this DTM's lexical handler, if it has one.
 871    *
 872    * %REVIEW% Should this return null if constrution already done/begun?
 873    *
 874    * @return null if this model doesn't respond to lexical SAX events.
 875    */
 876   public org.xml.sax.ext.LexicalHandler getLexicalHandler();
 877 
 878   /**
 879    * Return this DTM's EntityResolver, if it has one.
 880    *
 881    * @return null if this model doesn't respond to SAX entity ref events.
 882    */
 883   public org.xml.sax.EntityResolver getEntityResolver();
 884 
 885   /**
 886    * Return this DTM's DTDHandler, if it has one.
 887    *
 888    * @return null if this model doesn't respond to SAX dtd events.
 889    */
 890   public org.xml.sax.DTDHandler getDTDHandler();
 891 
 892   /**
 893    * Return this DTM's ErrorHandler, if it has one.
 894    *
 895    * @return null if this model doesn't respond to SAX error events.
 896    */
 897   public org.xml.sax.ErrorHandler getErrorHandler();
 898 
 899   /**
 900    * Return this DTM's DeclHandler, if it has one.
 901    *
 902    * @return null if this model doesn't respond to SAX Decl events.
 903    */
 904   public org.xml.sax.ext.DeclHandler getDeclHandler();
 905 
 906   /**
 907    * Append a child to "the end of the document". Please note that
 908    * the node is always cloned in a base DTM, since our basic behavior
 909    * is immutable so nodes can't be removed from their previous
 910    * location.
 911    *
 912    * <p> %REVIEW%  DTM maintains an insertion cursor which
 913    * performs a depth-first tree walk as nodes come in, and this operation
 914    * is really equivalent to:
 915    *    insertionCursor.appendChild(document.importNode(newChild)))
 916    * where the insert point is the last element that was appended (or
 917    * the last one popped back to by an end-element operation).</p>
 918    *
 919    * @param newChild Must be a valid new node handle.
 920    * @param clone true if the child should be cloned into the document.
 921    * @param cloneDepth if the clone argument is true, specifies that the
 922    *                   clone should include all it's children.
 923    */
 924   public void appendChild(int newChild, boolean clone, 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 }