org.apache.xerces.dom
Class DocumentImpl

java.lang.Object
  |
  +--org.apache.xerces.dom.NodeImpl
        |
        +--org.apache.xerces.dom.ChildNode
              |
              +--org.apache.xerces.dom.ParentNode
                    |
                    +--org.apache.xerces.dom.CoreDocumentImpl
                          |
                          +--org.apache.xerces.dom.DocumentImpl

All Implemented Interfaces:

java.lang.Cloneable, Document, DocumentEvent, DocumentRange, DocumentTraversal, EventTarget, Node, NodeList, java.io.Serializable

Direct Known Subclasses:

DeferredDocumentImpl, HTMLDocumentImpl, WMLDocumentImpl

public class DocumentImpl

extends CoreDocumentImpl

implements DocumentTraversal, DocumentEvent, DocumentRange

The Document interface represents the entire HTML or XML document. Conceptually, it is the root of the document tree, and provides the primary access to the document's data.

Since elements, text nodes, comments, processing instructions, etc. cannot exist outside the context of a Document, the Document interface also contains the factory methods needed to create these objects. The Node objects created have a ownerDocument attribute which associates them with the Document within whose context they were created.

The DocumentImpl class also implements the DOM Level 2 DocumentTraversal interface. This interface is comprised of factory methods needed to create NodeIterators and TreeWalkers. The process of creating NodeIterator objects also adds these references to this document. After finishing with an iterator it is important to remove the object using the remove methods in this implementation. This allows the release of the references from the iterator objects to the DOM Nodes.

Note: When any node in the document is serialized, the entire document is serialized along with it.

Since:

PR-DOM-Level-1-19980818.

Version:

Author:

Arnaud Le Hors, IBM, Joe Kesselman, IBM, Andy Clark, IBM, Ralf Pfeiffer, IBM

See Also:

Serialized Form

Field Summary

protected java.util.Hashtable	eventListeners Table for event listeners registered to this document nodes.
protected java.util.Vector	iterators Iterators
protected boolean	mutationEvents Bypass mutation events firing.
protected java.util.Vector	ranges Ranges
protected java.util.Hashtable	userData Table for user data attached to this document nodes.

Fields inherited from class org.apache.xerces.dom.CoreDocumentImpl

allowGrammarAccess, changes, docElement, docType, encoding, errorChecking, identifiers, kidOK, standalone, version

Fields inherited from class org.apache.xerces.dom.ParentNode

fCachedChild, fCachedChildIndex, fCachedLength, firstChild, ownerDocument

Fields inherited from class org.apache.xerces.dom.ChildNode

nextSibling, previousSibling

Fields inherited from class org.apache.xerces.dom.NodeImpl

ELEMENT_DEFINITION_NODE, FIRSTCHILD, flags, HASSTRING, IGNORABLEWS, OWNED, ownerNode, READONLY, SPECIFIED, SYNCCHILDREN, SYNCDATA, UNNORMALIZED

Fields inherited from interface org.w3c.dom.Node

ATTRIBUTE_NODE, CDATA_SECTION_NODE, COMMENT_NODE, DOCUMENT_FRAGMENT_NODE, DOCUMENT_NODE, DOCUMENT_TYPE_NODE, ELEMENT_NODE, ENTITY_NODE, ENTITY_REFERENCE_NODE, NOTATION_NODE, PROCESSING_INSTRUCTION_NODE, TEXT_NODE

Constructor Summary

DocumentImpl()
NON-DOM: Actually creating a Document is outside the DOM's spec, since it has to operate in terms of a particular implementation.

DocumentImpl(boolean grammarAccess)
Constructor.

DocumentImpl(DocumentType doctype)
For DOM2 support.

DocumentImpl(DocumentType doctype, boolean grammarAccess)
For DOM2 support.

Method Summary

protected void	addEventListener(NodeImpl node, java.lang.String type, EventListener listener, boolean useCapture) Introduced in DOM Level 2.
Node	cloneNode(boolean deep) Deep-clone a document, including fixing ownerDoc for the cloned children.
Event	createEvent(java.lang.String type) Introduced in DOM Level 2.
NodeIterator	createNodeIterator(Node root, int whatToShow, NodeFilter filter, boolean entityReferenceExpansion) Create and return a NodeIterator.
NodeIterator	createNodeIterator(Node root, short whatToShow, NodeFilter filter) NON-DOM extension: Create and return a NodeIterator.
Range	createRange() This interface can be obtained from the object implementing the Document interface using binding-specific casting methods.
TreeWalker	createTreeWalker(Node root, int whatToShow, NodeFilter filter, boolean entityReferenceExpansion) Create and return a TreeWalker.
TreeWalker	createTreeWalker(Node root, short whatToShow, NodeFilter filter) NON-DOM extension: Create and return a TreeWalker.
protected void	dispatchAggregateEvents(NodeImpl node, AttrImpl enclosingAttr, java.lang.String oldvalue, short change) NON-DOM INTERNAL: Generate the "aggregated" post-mutation events DOMAttrModified and DOMSubtreeModified.
protected void	dispatchAggregateEvents(NodeImpl node, org.apache.xerces.dom.DocumentImpl.EnclosingAttr ea) NON-DOM INTERNAL: Convenience wrapper for calling dispatchAggregateEvents when the context was established by savedEnclosingAttr.
protected boolean	dispatchEvent(NodeImpl node, Event event) Introduced in DOM Level 2.
protected void	dispatchEventToSubtree(NodeImpl node, Node n, Event e) NON-DOM INTERNAL: DOMNodeInsertedIntoDocument and ...RemovedFrom...
protected java.util.Vector	getEventListeners(NodeImpl n) Retreive event listener registered on a given node
DOMImplementation	getImplementation() Retrieve information describing the abilities of this particular DOM implementation.
protected java.lang.Object	getUserData(NodeImpl n) Retreive user data related to a given node
protected void	removeEventListener(NodeImpl node, java.lang.String type, EventListener listener, boolean useCapture) Introduced in DOM Level 2.
protected void	saveEnclosingAttr(NodeImpl node) NON-DOM INTERNAL: Pre-mutation context check, in preparation for later generating DOMAttrModified events.
protected void	setEventListeners(NodeImpl n, java.util.Vector listeners) Store event listener registered on a given node This is another place where we could use weak references! Indeed, the node here won't be GC'ed as long as some listener is registered on it, since the eventsListeners table will have a reference to the node.
protected void	setUserData(NodeImpl n, java.lang.Object data) Store user data related to a given node This is a place where we could use weak references! Indeed, the node here won't be GC'ed as long as some user data is attached to it, since the userData table will have a reference to the node.

Methods inherited from class org.apache.xerces.dom.CoreDocumentImpl

adoptNode, changed, changes, clone, cloneNode, createAttribute, createAttributeNS, createCDATASection, createComment, createDocumentFragment, createDocumentType, createElement, createElementDefinition, createElementNS, createEntity, createEntityReference, createNotation, createProcessingInstruction, createTextNode, getDoctype, getDocumentElement, getElementById, getElementsByTagName, getElementsByTagNameNS, getEncoding, getErrorChecking, getIdentifier, getIdentifiers, getNodeName, getNodeType, getOwnerDocument, getStandalone, getStrictErrorChecking, getVersion, importNode, insertBefore, isKidOK, isXMLName, putIdentifier, removeChild, removeIdentifier, replaceChild, setEncoding, setErrorChecking, setStandalone, setStrictErrorChecking, setVersion

Methods inherited from class org.apache.xerces.dom.ParentNode

getChildNodes, getChildNodesUnoptimized, getFirstChild, getLastChild, getLength, hasChildNodes, item, normalize, setReadOnly, synchronizeChildren

Methods inherited from class org.apache.xerces.dom.ChildNode

getNextSibling, getParentNode, getPreviousSibling

Methods inherited from class org.apache.xerces.dom.NodeImpl

addEventListener, appendChild, dispatchEvent, getAttributes, getLocalName, getNamespaceURI, getNodeValue, getPrefix, getReadOnly, getUserData, hasAttributes, isSupported, removeEventListener, setNodeValue, setPrefix, setUserData, synchronizeData, toString

Methods inherited from class java.lang.Object

equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface org.w3c.dom.Node

appendChild, getAttributes, getChildNodes, getFirstChild, getLastChild, getLocalName, getNamespaceURI, getNextSibling, getNodeValue, getParentNode, getPrefix, getPreviousSibling, hasAttributes, hasChildNodes, isSupported, normalize, setNodeValue, setPrefix

Field Detail

iterators

protected java.util.Vector iterators

Iterators

ranges

protected java.util.Vector ranges

Ranges

userData

protected java.util.Hashtable userData

Table for user data attached to this document nodes.

eventListeners

protected java.util.Hashtable eventListeners

Table for event listeners registered to this document nodes.

mutationEvents

protected boolean mutationEvents

Bypass mutation events firing.

Constructor Detail

DocumentImpl

public DocumentImpl()

NON-DOM: Actually creating a Document is outside the DOM's spec, since it has to operate in terms of a particular implementation.

DocumentImpl

public DocumentImpl(boolean grammarAccess)

Constructor.

DocumentImpl

public DocumentImpl(DocumentType doctype)

For DOM2 support. The createDocument factory method is in DOMImplementation.

DocumentImpl

public DocumentImpl(DocumentType doctype,
                    boolean grammarAccess)

For DOM2 support.

Method Detail

cloneNode

public Node cloneNode(boolean deep)

Deep-clone a document, including fixing ownerDoc for the cloned children. Note that this requires bypassing the WRONG_DOCUMENT_ERR protection. I've chosen to implement it by calling importNode which is DOM Level 2.

Overrides:

cloneNode in class CoreDocumentImpl

Parameters:

deep - boolean, iff true replicate children

Returns:

org.w3c.dom.Node

getImplementation

public DOMImplementation getImplementation()

Retrieve information describing the abilities of this particular DOM implementation. Intended to support applications that may be using DOMs retrieved from several different sources, potentially with different underlying representations.

Overrides:

getImplementation in class CoreDocumentImpl

setUserData

protected void setUserData(NodeImpl n,
                           java.lang.Object data)

Store user data related to a given node This is a place where we could use weak references! Indeed, the node here won't be GC'ed as long as some user data is attached to it, since the userData table will have a reference to the node.

Overrides:

setUserData in class CoreDocumentImpl

getUserData

protected java.lang.Object getUserData(NodeImpl n)

Retreive user data related to a given node

Overrides:

getUserData in class CoreDocumentImpl

createNodeIterator

public NodeIterator createNodeIterator(Node root,
                                       short whatToShow,
                                       NodeFilter filter)

NON-DOM extension: Create and return a NodeIterator. The NodeIterator is added to a list of NodeIterators so that it can be removed to free up the DOM Nodes it references.

Parameters:

root - The root of the iterator.

whatToShow - The whatToShow mask.

filter - The NodeFilter installed. Null means no filter.

See Also:

#removeNodeIterator, #removeNodeIterators

createNodeIterator

public NodeIterator createNodeIterator(Node root,
                                       int whatToShow,
                                       NodeFilter filter,
                                       boolean entityReferenceExpansion)

Create and return a NodeIterator. The NodeIterator is added to a list of NodeIterators so that it can be removed to free up the DOM Nodes it references.

Specified by:

createNodeIterator in interface DocumentTraversal

Parameters:

root - The root of the iterator.

whatToShow - The whatToShow mask.

filter - The NodeFilter installed. Null means no filter.

entityReferenceExpansion - true to expand the contents of EntityReference nodes

Since:

WD-DOM-Level-2-19990923

See Also:

#removeNodeIterator, #removeNodeIterators

createTreeWalker

public TreeWalker createTreeWalker(Node root,
                                   short whatToShow,
                                   NodeFilter filter)

NON-DOM extension: Create and return a TreeWalker.

Parameters:

root - The root of the iterator.

whatToShow - The whatToShow mask.

filter - The NodeFilter installed. Null means no filter.

createTreeWalker

public TreeWalker createTreeWalker(Node root,
                                   int whatToShow,
                                   NodeFilter filter,
                                   boolean entityReferenceExpansion)

Create and return a TreeWalker.

Specified by:

createTreeWalker in interface DocumentTraversal

Parameters:

root - The root of the iterator.

whatToShow - The whatToShow mask.

filter - The NodeFilter installed. Null means no filter.

entityReferenceExpansion - true to expand the contents of EntityReference nodes

Since:

WD-DOM-Level-2-19990923

createRange

public Range createRange()

Description copied from interface: DocumentRange

This interface can be obtained from the object implementing the Document interface using binding-specific casting methods.

Specified by:

createRange in interface DocumentRange

Following copied from interface: org.w3c.dom.ranges.DocumentRange

Returns:

The initial state of the Range returned from this method is such that both of its boundary-points are positioned at the beginning of the corresponding Document, before any content. The Range returned can only be used to select content associated with this Document, or with DocumentFragments and Attrs for which this Document is the ownerDocument.

createEvent

public Event createEvent(java.lang.String type)
                  throws DOMException

Introduced in DOM Level 2. Optional.

Create and return Event objects.

Specified by:

createEvent in interface DocumentEvent

Parameters:

type - The eventType parameter specifies the type of Event interface to be created. If the Event interface specified is supported by the implementation this method will return a new Event of the interface type requested. If the Event is to be dispatched via the dispatchEvent method the appropriate event init method must be called after creation in order to initialize the Event's values. As an example, a user wishing to synthesize some kind of Event would call createEvent with the parameter "Events". The initEvent method could then be called on the newly created Event to set the specific type of Event to be dispatched and set its context information.

Returns:

Newly created Event

Throws:

DOMException - NOT_SUPPORTED_ERR: Raised if the implementation does not support the type of Event interface requested

Since:

WD-DOM-Level-2-19990923

setEventListeners

protected void setEventListeners(NodeImpl n,
                                 java.util.Vector listeners)

Store event listener registered on a given node This is another place where we could use weak references! Indeed, the node here won't be GC'ed as long as some listener is registered on it, since the eventsListeners table will have a reference to the node.

getEventListeners

protected java.util.Vector getEventListeners(NodeImpl n)

Retreive event listener registered on a given node

addEventListener

protected void addEventListener(NodeImpl node,
                                java.lang.String type,
                                EventListener listener,
                                boolean useCapture)

Introduced in DOM Level 2.

Register an event listener with this Node. A listener may be independently registered as both Capturing and Bubbling, but may only be registered once per role; redundant registrations are ignored.

Overrides:

addEventListener in class CoreDocumentImpl

Parameters:

node - node to add listener to

type - Event name (NOT event group!) to listen for.

listener - Who gets called when event is dispatched

useCapture - True iff listener is registered on capturing phase rather than at-target or bubbling

removeEventListener

protected void removeEventListener(NodeImpl node,
                                   java.lang.String type,
                                   EventListener listener,
                                   boolean useCapture)

Introduced in DOM Level 2.

Deregister an event listener previously registered with this Node. A listener must be independently removed from the Capturing and Bubbling roles. Redundant removals (of listeners not currently registered for this role) are ignored.

Overrides:

removeEventListener in class CoreDocumentImpl

Parameters:

node - node to remove listener from

type - Event name (NOT event group!) to listen for.

listener - Who gets called when event is dispatched

useCapture - True iff listener is registered on capturing phase rather than at-target or bubbling

dispatchEvent

protected boolean dispatchEvent(NodeImpl node,
                                Event event)

Introduced in DOM Level 2.

Distribution engine for DOM Level 2 Events.

Event propagation runs as follows:

1. Event is dispatched to a particular target node, which invokes this code. Note that the event's stopPropagation flag is cleared when dispatch begins; thereafter, if it has been set before processing of a node commences, we instead immediately advance to the DEFAULT phase.
2. The node's ancestors are established as destinations for events. For capture and bubble purposes, node ancestry is determined at the time dispatch starts. If an event handler alters the document tree, that does not change which nodes will be informed of the event.
3. CAPTURING_PHASE: Ancestors are scanned, root to target, for Capturing listeners. If found, they are invoked (see below).
4. AT_TARGET: Event is dispatched to NON-CAPTURING listeners on the target node. Note that capturing listeners on this node are _not_ invoked.
5. BUBBLING_PHASE: Ancestors are scanned, target to root, for non-capturing listeners.
6. Default processing: Some DOMs have default behaviors bound to specific nodes. If this DOM does, and if the event's preventDefault flag has not been set, we now return to the target node and process its default handler for this event, if any.

Note that registration of handlers during processing of an event does not take effect during this phase of this event; they will not be called until the next time this node is visited by dispatchEvent. On the other hand, removals take effect immediately.

If an event handler itself causes events to be dispatched, they are processed synchronously, before processing resumes on the event which triggered them. Please be aware that this may result in events arriving at listeners "out of order" relative to the actual sequence of requests.

Note that our implementation resets the event's stop/prevent flags when dispatch begins. I believe the DOM's intent is that event objects be redispatchable, though it isn't stated in those terms.

Overrides:

dispatchEvent in class CoreDocumentImpl

Parameters:

node - node to dispatch to

event - the event object to be dispatched to registered EventListeners

Returns:

true if the event's preventDefault() method was invoked by an EventListener; otherwise false.

dispatchEventToSubtree

protected void dispatchEventToSubtree(NodeImpl node,
                                      Node n,
                                      Event e)

NON-DOM INTERNAL: DOMNodeInsertedIntoDocument and ...RemovedFrom... are dispatched to an entire subtree. This is the distribution code therefor. They DO NOT bubble, thanks be, but may be captured.

***** At the moment I'm being sloppy and using the normal capture dispatcher on every node. This could be optimized hugely by writing a capture engine that tracks our position in the tree to update the capture chain without repeated chases up to root.

Parameters:

node - node to dispatch to

n - node which was directly inserted or removed

e - event to be sent to that node and its subtree

dispatchAggregateEvents

protected void dispatchAggregateEvents(NodeImpl node,
                                       org.apache.xerces.dom.DocumentImpl.EnclosingAttr ea)

NON-DOM INTERNAL: Convenience wrapper for calling dispatchAggregateEvents when the context was established by savedEnclosingAttr.

Parameters:

node - node to dispatch to

ea - description of Attr affected by current operation

dispatchAggregateEvents

protected void dispatchAggregateEvents(NodeImpl node,
                                       AttrImpl enclosingAttr,
                                       java.lang.String oldvalue,
                                       short change)

NON-DOM INTERNAL: Generate the "aggregated" post-mutation events DOMAttrModified and DOMSubtreeModified. Both of these should be issued only once for each user-requested mutation operation, even if that involves multiple changes to the DOM. For example, if a DOM operation makes multiple changes to a single Attr before returning, it would be nice to generate only one DOMAttrModified, and multiple changes over larger scope but within a recognizable single subtree might want to generate only one DOMSubtreeModified, sent to their lowest common ancestor.

To manage this, use the "internal" versions of insert and remove with MUTATION_LOCAL, then make an explicit call to this routine at the higher level. Some examples now exist in our code.

Parameters:

node - The node to dispatch to

enclosingAttr - The Attr node (if any) whose value has been changed as a result of the DOM operation. Null if none such.

oldValue - The String value previously held by the enclosingAttr. Ignored if none such.

change - Type of modification to the attr. See MutationEvent.attrChange

saveEnclosingAttr

protected void saveEnclosingAttr(NodeImpl node)

NON-DOM INTERNAL: Pre-mutation context check, in preparation for later generating DOMAttrModified events. Determines whether this node is within an Attr

Parameters:

node - node to get enclosing attribute for

Returns:

either a description of that Attr, or null if none such.