Node Interface

Table 4-8 summarizes the methods of available through the Node interface.

XmlDomAppendChild

Appends the node to the end of the parent's list of children and returns the new node. If newChild is a DocumentFragment, all of its children are appended in original order; the DocumentFragment node itself is not.

xmlnode* XmlDomAppendChild(
   xmlctx *xctx, 
   xmlnode *parent, 
   xmlnode *newChild)

xctx

IN

parent

IN

newChild

IN

(xmlnode *) node added

XmlDomCleanNode

Frees parts of the node which were allocated by DOM itself, but does not recurse to children or touch the node's attributes. After freeing part of the node (such as name), a DOM call to get that part (such as XmlDomGetNodeName) should return a NULL pointer. Used to manage the allocations of a node parts of which are controlled by DOM, and part by the user. Calling clean frees all allocations may by DOM and leaves the user's allocations alone. The user is responsible for freeing their own allocations.

void XmlDomCleanNode(
   xmlctx *xctx, 
   xmlnode *node)

xctx

IN

node

IN

XmlDomCloneNode

Creates and returns a duplicate of a node. The duplicate node has no parent. Cloning an element copies all attributes and their values, including those generated by the XML processor to represent defaulted attributes, but it does not copy any text it contains unless it is a deep clone, since the text is contained in a child text node. Cloning any other type of node simply returns a copy of the node. Note that a clone of an unspecified attribute node is specified. If deep is TRUE, all children of the node are recursively cloned, and the cloned node will have cloned children; a non-deep clone will have no children.

xmlnode* XmlDomCloneNode(
   xmlctx *xctx, 
   xmlnode *node, 
   boolean deep)

xctx

IN

node

IN

deep

IN

(xmlnode *) duplicate (cloned) node

XmlDomFreeNode

Free a node allocated with XmlDomCreateXXX. Frees all resources associated with a node, then frees the node itself. Certain parts of the node are under DOM control, and some parts may be under user control. DOM keeps flags tracking who owns what, and only frees its own allocations. The user is responsible for freeing their own parts of the node before calling XmlDomFreeNode.

void XmlDomFreeNode(
   xmlctx *xctx, 
   xmlnode *node)

xctx

IN

node

IN

XmlDomGetAttrs

Returns a NamedNodeMap of attributes of an element node, or NULL if it has no attributes. For other node types, NULL is returned. Note that if an element once had attributes, but they have all been removed, an empty list will be returned. So, presence of the list does not mean the element has attributes. You must check the size of the list with XmlDomNumAttrs or use XmlDomHasChildNodes first.

xmlnamedmap* XmlDomGetAttrs(
   xmlctx *xctx, 
   xmlelemnode *elem)

xctx

IN

elem

IN

(xmlnamedmap *) NamedNodeMap of node's attributes

XmlDomGetChildNodes

Returns a list of the node's children, or NULL if it has no children. Only Element, Document, DTD, and DocumentFragment nodes may have children; all other types will return NULL.

Note that an empty list may be returned if the node once had children, but all have been removed! That is, the list may exist but have no members. So, presence of the list alone does not mean the node has children. You must check the size of the list with XmlDomNumChildNodes or use XmlDomHasChildNodes first.

The xmlnodelist structure is opaque and can only be manipulated with functions in the NodeList interface.

The returned list is live; all changes in the original node are reflected immediately.

xmlnodelist* XmlDomGetChildNodes(
   xmlctx *xctx, 
   xmlnode *node)

xctx

IN

node

IN

(xmlnodelist *) live NodeList containing all children of node

XmlDomGetDefaultNS

Gets the default namespace for a node.

oratext* XmlDomGetDefaultNS(
   xmlctx *xctx, 
   xmlnode *node)

xctx

IN

node

IN

(oratext *) default namespace for node [data encoding; may be NULL]

XmlDomGetFirstChild

Returns the first child of a node, or NULL if the node has no children. Only Element, Document, DTD, and DocumentFragment nodes may have children; all other types will return NULL.

xmlnode* XmlDomGetFirstChild(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(xmlnode *) first child of node

XmlDomGetFirstPfnsPair

This function is to allow implementations an opportunity to speedup the iteration of all available prefix-URI bindings available on a given node. It returns a state structure and the prefix and URI of the first prefix-URI mapping. The state structure should be passed to XmlDomGetNextPfnsPair on the remaining pairs.

xmlpfnspair* XmlDomGetFirstPfnsPair(
   xmlctx *xctx, 
   xmlnode *node, 
   oratext **prefix,
   oratext **uri)

xctx

IN

node

IN

prefix

OUT

uri

OUT

(xmlpfnspair *) iterating object or NULL of no prefixes

XmlDomGetLastChild

Returns the last child of a node, or NULL if the node has no children. Only Element, Document, DTD, and DocumentFragment nodes may have children; all other types will return NULL.

xmlnode* XmlDomGetLastChild(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(xmlnode *) last child of node

XmlDomGetNextPfnsPair

This function is to allow implementations an opportunity to speedup the iteration of all available prefix-URI bindings available on a given node. Given an iterator structure from XmlDomGetFirstPfnsPair, returns the next prefix-URI mapping; repeat calls to XmlDomGetNextPfnsPair until NULL is returned.

xmlpfnspair* XmlDomGetNextPfnsPair(
   xmlctx *xctx
   xmlpfnspair *pfns,
   oratext **prefix,
   oratext **uri)

xctx

IN

node

IN

prefix

OUT

uri

OUT

(xmlpfnspair *) iterating object, NULL when no more pairs

XmlDomGetNextSibling

Returns the node following a node at the same level in the DOM tree. That is, for each child of a parent node, the next sibling of that child is the child which comes after it. If a node is the last child of its parent, or has no parent, NULL is returned.

xmlnode* XmlDomGetNextSibling(
   xmlctx *xctx, 
   xmlnode *node)

xctx

IN

node

IN

(xmlnode *) node immediately following node at same level

XmlDomGetNodeLocal

Returns the namespace local name for a node as a NULL-terminated string. If the node's name is not fully qualified (has no prefix), then the local name is the same as the name.

A length-encoded version is available as XmlDomGetNodeLocalLen which returns the local name as a pointer and length, for use if the data is known to use XMLType backing store.

oratext* XmlDomGetNodeLocal(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(oratext *) local name of node [data encoding]

XmlDomGetNodeLocalLen

Returns the namespace local name for a node as a length-encoded string. If the node's name is not fully qualified (has no prefix), then the local name is the same as the name.

A NULL-terminated version is available as XmlDomGetNodeLocal which returns the local name as NULL-terminated string. If the backing store is known to be XMLType, then the node's data will be stored internally as length-encoded. Using the length-based Get functions will avoid having to copy and NULL-terminate the data.

If both the input buffer is non-NULL and the input buffer length is nonzero, then the value will be stored in the input buffer. Else, the implementation will return its own buffer.

If the actual length is greater than buflen, then a truncated value will be copied into the buffer and len will return the actual length.

oratext* XmlDomGetNodeLocalLen(
   xmlctx *xctx,
   xmlnode *node,
   oratext *buf,
   ub4 buflen, 
   ub4 *len)

xctx

IN

node

IN

buf

IN

buflen

IN

len

OUT

(oratext *) local name of node [data encoding]

XmlDomGetNodeName

Returns the (fully-qualified) name of a node (in the data encoding) as a NULL-terminated string, for example bar\0 or foo:bar\0.

Note that some node types have fixed names: "#text", "#cdata-section", "#comment", "#document", "#document-fragment".

A node's name cannot be changed once it is created, so there is no matching SetNodeName function.

A length-based version is available as XmlDomGetNodeNameLen which returns the node name as a pointer and length, for use if the data is known to use XMLType backing store.

oratext* XmlDomGetNodeName(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(oratext *) name of node [data encoding]

XmlDomGetNodeNameLen

Returns the (fully-qualified) name of a node (in the data encoding) as a length-encoded string, for example "bar", 3 or "foo:bar", 7.

Note that some node types have fixed names: "#text", "#cdata-section", "#comment", "#document", "#document-fragment".

A node's name cannot be changed once it is created, so there is no matching SetNodeName function.

A NULL-terminated version is available as XmlDomGetNodeName which returns the node name as NULL-terminated string. If the backing store is known to be XMLType, then the node's name will be stored internally as length-encoded. Using the length-encoded GetXXX functions will avoid having to copy and NULL-terminate the name.

If both the input buffer is non-NULL and the input buffer length is nonzero, then the value will be stored in the input buffer. Else, the implementation will return its own buffer.

If the actual length is greater than buflen, then a truncated value will be copied into the buffer and len will return the actual length.

oratext* XmlDomGetNodeNameLen(
   xmlctx *xctx,
   xmlnode *node,
   oratext *buf,
   ub4 buflen, 
   ub4 *len)

xctx

IN

node

IN

buf

IN

buflen

IN

len

OUT

(oratext *) name of node, with length of name set in 'len'

XmlDomGetNodePrefix

Returns the namespace prefix for a node (as a NULL-terminated string). If the node's name is not fully qualified (has no prefix), NULL is returned.

oratext* XmlDomGetNodePrefix(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(oratext *) namespace prefix of node [data encoding; may be NULL]

XmlDomGetNodeType

Returns the type code of a node. The type names and numeric values match the DOM specification:

• ELEMENT_NODE=1

• ATTRIBUTE_NODE=2

• TEXT_NODE=3

• CDATA_SECTION_NODE=4

• ENTITY_REFERENCE_NODE=5

• ENTITY_NODE=6

• PROCESSING_INSTRUCTION_NODE=7

• COMMENT_NODE=8

• DOCUMENT_NODE=9

• DOCUMENT_TYPE_NODE=10

• DOCUMENT_FRAGMENT_NODE=11

• NOTATION_NODE=12

Additional Oracle extension node types are as follows:

• ELEMENT_DECL_NODE

• ATTR_DECL_NODE

• CP_ELEMENT_NODE

• CP_CHOICE_NODE

• CP_PCDATA_NODE

• CP_STAR_NODE

• CP_PLUS_NODE

• CP_OPT_NODE

xmlnodetype XmlDomGetNodeType(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(xmlnodetype) numeric type-code of the node

XmlDomGetNodeURI

Returns the namespace URI for a node (in the data encoding) as a NULL-terminated string. If the node's name is not qualified (does not contain a namespace prefix), it will have the default namespace in effect when the node was created (which may be NULL).

A length-encoded version is available as XmlDomGetNodeURILen which returns the URI as a pointer and length, for use if the data is known to use XMLType backing store.

oratext* XmlDomGetNodeURI(
   xmlctx *xctx, 
   xmlnode *node)

xctx

IN

node

IN

(oratext *) namespace URI of node [data encoding; may be NULL]

XmlDomGetNodeURILen

Returns the namespace URI for a node (in the data encoding) as length-encoded string. If the node's name is not qualified (does not contain a namespace prefix), it will have the default namespace in effect when the node was created (which may be NULL).

A NULL-terminated version is available as XmlDomGetNodeURI which returns the URI value as NULL-terminated string. If the backing store is known to be XMLType, then the node's data will be stored internally as length-encoded. Using the length-based Get functions will avoid having to copy and NULL-terminate the data.

If both the input buffer is non-NULL and the input buffer length is nonzero, then the value will be stored in the input buffer. Else, the implementation will return its own buffer.

If the actual length is greater than buflen, then a truncated value will be copied into the buffer and len will return the actual length.

oratext* XmlDomGetNodeURILen(
   xmlctx *xctx,
   xmlnode *node,
   oratext *buf,
   ub4 buflen, 
   ub4 *len)

xctx

IN

node

IN

buf

IN

buflen

IN

len

OUT

(oratext *) namespace URI of node [data encoding; may be NULL]

XmlDomGetNodeValue

Returns the "value" (associated character data) for a node as a NULL-terminated string. Character and general entities will have been replaced. Only Attr, CDATA, Comment, ProcessingInstruction and Text nodes have values, all other node types have NULL value.

A length-encoded version is available as XmlDomGetNodeValueLen which returns the node value as a pointer and length, for use if the data is known to use XMLType backing store.

oratext* XmlDomGetNodeValue(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(oratext *) value of node

XmlDomGetNodeValueLen

Returns the "value" (associated character data) for a node as a length-encoded string. Character and general entities will have been replaced. Only Attr, CDATA, Comment, PI and Text nodes have values, all other node types have NULL value.

A NULL-terminated version is available as XmlDomGetNodeValue which returns the node value as NULL-terminated string. If the backing store is known to be XMLType, then the node's data will be stored internally as length-encoded. Using the length-based Get functions will avoid having to copy and NULL-terminate the data.

If both the input buffer is non-NULL and the input buffer length is nonzero, then the value will be stored in the input buffer. Else, the implementation will return its own buffer.

If the actual length is greater than buflen, then a truncated value will be copied into the buffer and len will return the actual length.

oratext* XmlDomGetNodeValueLen(
   xmlctx *xctx, 
   xmlnode *node, 
   oratext *buf, 
   ub4 buflen, 
   ub4 *len)

xctx

IN

node

IN

buf

IN

buflen

IN

len

OUT

(oratext *) value of node

XmlDomGetNodeValueStream

Returns the large data for a node and sends it in pieces to the user's output stream. For very large values, it is not always possible to store them [efficiently] as a single contiguous chunk. This function is used to access chunked data of that type. Only XMLType chunks its data (sometimes); XDK's data is always contiguous.

xmlerr XmlDomGetNodeValueStream(
   xmlctx *xctx, 
   xmlnode *node, 
   xmlostream *ostream)

xctx

IN

node

IN

ostream

IN

(xmlerr) numeric error code, 0 on success

XmlDomGetOwnerDocument

Returns the Document node associated with a node. Each node may belong to only one document, or may not be associated with any document at all (such as immediately after XmlDomCreateElem, and so on). The "owning" document [node] is returned, or NULL for an orphan node.

xmldocnode* XmlDomGetOwnerDocument(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(xmldocnode *) document node is in

XmlDomGetParentNode

Returns a node's parent node. All nodes types except Attr, Document, DocumentFragment, Entity, and Notation may have a parent (these five exceptions always have a NULL parent). If a node has just been created but not yet added to the DOM tree, or if it has been removed from the DOM tree, its parent is also NULL.

xmlnode* XmlDomGetParentNode(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(xmlnode *) parent of node

XmlDomGetPrevSibling

Returns the node preceding a node at the same level in the DOM tree. That is, for each child of a parent node, the previous sibling of that child is the child which came before it. If a node is the first child of its parent, or has no parent, NULL is returned.

xmlnode* XmlDomGetPrevSibling(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(xmlnode *) node immediately preceding node at same level

XmlDomGetSourceEntity

Returns the external entity node whose inclusion caused the creation of the given node.

xmlentnode* XmlDomGetSourceEntity(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(xmlentnode *) entity node if the input is from an external entity

XmlDomGetSourceLine

Returns the line# in the original source where the node started. The first line in every input is line #1.

ub4 XmlDomGetSourceLine(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(ub4) line number of node in original input source

XmlDomGetSourceLocation

Return source location (path, URI, and so on) of node. Note this will be in the compiler encoding, not the data encoding!

oratext* XmlDomGetSourceLocation(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(oratext *) full path of input source [in compiler encoding]

XmlDomHasAttrs

Test if an element has attributes. Returns TRUE if any attributes of any sort are defined (namespace or regular).

boolean XmlDomHasAttrs(
   xmlctx *xctx,
   xmlelemnode *elem)

xctx

IN

elem

IN

(boolean) TRUE if element has attributes

XmlDomHasChildNodes

Test if a node has children. Only Element, Document, DTD, and DocumentFragment nodes may have children. Note that just because XmlDomGetChildNodes returns a list does not mean the node actually has children, since the list may be empty, so a non-NULL return from XmlDomGetChildNodes should not be used as a test.

boolean XmlDomHasChildNodes(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(boolean) TRUE if the node has any children

XmlDomInsertBefore

Inserts the node newChild before the existing child node refChild in the parent node. If refChild is NULL, appends to parent's children as for each XmlDomAppendChild; otherwise it must be a child of the given parent. If newChild is a DocumentFragment, all of its children are inserted (in the same order) before refChild; the DocumentFragment node itself is not. If newChild is already in the DOM tree, it is first removed from its current position.

xmlnode* XmlDomInsertBefore(
   xmlctx *xctx,
   xmlnode *parent, 
   xmlnode *newChild,
   xmlnode *refChild)

xctx

IN

parent

IN

newChild

IN

refChild

IN

(xmlnode *) node being inserted

XmlDomNormalize

Normalizes the subtree rooted at an element, merges adjacent Text nodes children of elements. Note that adjacent Text nodes will never be created during a normal parse, only after manipulation of the document with DOM calls.

void XmlDomNormalize(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

XmlDomNumAttrs

Returns the number of attributes of an element. Note that just because a list is returned by XmlDomGetAttrs does not mean it contains any attributes; it may be an empty list with zero length.

ub4 XmlDomNumAttrs(
   xmlctx *xctx,
   xmlelemnode *elem)

xctx

IN

elem

IN

(ub4) number of attributes of node

XmlDomNumChildNodes

Returns the number of children of a node. Only Element, Document, DTD, and DocumentFragment nodes may have children, all other types return 0. Note that just because XmlDomGetChildNodes returns a list does not mean that it contains any children; it may be an empty list with zero length.

ub4 XmlDomNumChildNodes(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(ub4) number of children of node

XmlDomPrefixToURI

Given a namespace prefix and a node, returns the namespace URI mapped to that prefix. If the given node doesn't have a matching prefix, its parent is tried, then its parent, and so on, all the way to the root node. If the prefix is undefined, NULL is returned.

oratext* XmlDomPrefixToURI(
   xmlctx *xctx,
   xmlnode *node,
   oratext *prefix)

xctx

IN

node

IN

prefix

IN

(oratext *) URI for prefix [data encoding; NULL if no match]

XmlDomRemoveChild

Removes a node from its parent's list of children and returns it. The node is orphaned; its parent will be NULL after removal.

xmlnode* XmlDomRemoveChild(
   xmlctx *xctx, 
   xmlnode *oldChild)

xctx

IN

oldChild

IN

(xmlnode *) node removed

XmlDomReplaceChild

Replaces the child node oldChild with the new node newChild in oldChild's parent, and returns oldChild (which is now orphaned, with a NULL parent). If newChild is a DocumentFragment, all of its children are inserted in place of oldChild; the DocumentFragment node itself is not. If newChild is already in the DOM tree, it is first removed from its current position.

xmlnode* XmlDomReplaceChild(
   xmlctx *xctx, 
   xmlnode *newChild,
   xmlnode *oldChild)

xctx

IN

newChild

IN

oldChild

IN

(xmlnode *) node replaced

XmlDomSetDefaultNS

Set the default namespace for a node

void XmlDomSetDefaultNS(
   xmlctx *xctx,
   xmlnode *node,
   oratext *defns)

xctx

IN

node

IN

defns

IN

XmlDomSetNodePrefix

Sets the namespace prefix of node (as NULL-terminated string). Does not verify the prefix is defined. Just causes a new qualified name to be formed from the new prefix and the old local name; the new qualified name will be under DOM control and should not be managed by the user.

void XmlDomSetNodePrefix(
   xmlctx *xctx,
   xmlnode *node,
   oratext *prefix)

xctx

IN

node

IN

prefix

OUT

XmlDomSetNodeValue

Sets a node's value (character data) as a NULL-terminated string. Does not allow setting the value to NULL. Only Attr, CDATA, Comment, PI and Text nodes have values; trying to set the value of another type of node is a no-op. The new value must be in the data encoding. It is not verified, converted, or checked.

The value is not copied, its pointer is just stored. The user is responsible for persistence and freeing of that data.

xmlerr XmlDomSetNodeValue(
   xmlctx *xctx,
   xmlnode *node,
   oratext *value)

xctx

IN

node

IN

value

IN

(xmlerr) numeric error code, 0 on success

XmlDomSetNodeValueLen

Sets the value (associated character data) for a node as a length-encoded string.

A NULL-terminated version is available as XmlDomSetNodeValue which takes the node value as a NULL-terminated string. If the backing store is known to be XMLType, then the node's data will be stored internally as length-encoded. Using the length-based Set functions will avoid having to copy and NULL-terminate the data.

xmlerr XmlDomSetNodeValueLen(
   xmlctx *xctx, 
   xmlnode *node, 
   oratext *value, 
   ub4 len)

xctx

IN

node

IN

value

IN

len

IN

(xmlerr) numeric error code, 0 on success

XmlDomSetNodeValueStream

Sets the large "value" (character data) for a node piecemeal from an input stream. For very large values, it is not always possible to store them [efficiently] as a single contiguous chunk. This function is used to store chunked data of that type. Used only for XMLType data; XDK's data is always contiguous.

xmlerr XmlDomSetNodeValueStream(
   xmlctx *xctx, 
   xmlnode *node,
   xmlistream *istream)

xctx

IN

node

IN

istream

IN

(xmlerr) numeric error code, 0 on success

XmlDomValidate

Given a root node, validates it against the current DTD.

xmlerr XmlDomValidate(
   xmlctx *xctx,
   xmlnode *node)

xctx

IN

node

IN

(xmlerr) error code, XMLERR_OK [0] means node is valid