net.sf.saxon.pull
Class PullFromIterator
public
class
PullFromIterator
extends Object
implements PullProvider
This class delivers any XPath sequence through the pull interface. Atomic values
in the sequence are supplied unchanged, as are top-level text, comment, attribute,
namespace, and processing-instruction nodes. Elements and documents appearing in
the input sequence are supplied as a sequence of events that walks recursively
down the subtree rooted at that node. The input is supplied in the form of a
SequenceIterator.
public void close()
Close the event reader. This indicates that no further events are required.
It is not necessary to close an event reader after
PullFromIterator has
been reported, but it is recommended to close it if reading terminates
prematurely. Once an event reader has been closed, the effect of further
calls on next() is undefined.
public int current()
Get the event most recently returned by next(), or by other calls that change
the position, for example getStringValue() and skipToMatchingEnd(). This
method does not change the position of the PullProvider.
Returns: the current event
Get an atomic value. This call may be used only when the last event reported was
ATOMIC_VALUE. This indicates that the PullProvider is reading a sequence that contains
a free-standing atomic value; it is never used when reading the content of a node.
Get the attributes associated with the current element. This method must
be called only after a START_ELEMENT event has been notified. The contents
of the returned AttributeCollection are guaranteed to remain unchanged
until the next START_ELEMENT event, but may be modified thereafter. The object
should not be modified by the client.
Attributes may be read before or after reading the namespaces of an element,
but must not be read after the first child node has been read, or after calling
one of the methods skipToEnd(), getStringValue(), or getTypedValue().
Returns: an AttributeCollection representing the attributes of the element
that has just been notified.
public int getFingerprint()
Get the fingerprint of the name of the element. This is similar to the nameCode, except that
it does not contain any information about the prefix: so two elements with the same fingerprint
have the same name, excluding prefix. This method
can be used after the
PullFromIterator,
PullFromIterator,
PullFromIterator,
PullFromIterator, or
PullFromIterator events.
If called at other times, the result is undefined and may result in an IllegalStateException.
If called when the current node is an unnamed namespace node (a node representing the default namespace)
the returned value is -1.
Returns: the fingerprint. The fingerprint can be used to obtain the local name
and namespace URI from the name pool.
public int getNameCode()
Get the nameCode identifying the name of the current node. This method
can be used after the
PullFromIterator,
PullFromIterator,
PullFromIterator, or
PullFromIterator events. With some PullProvider implementations,
including this one, it can also be used after
PullFromIterator.
If called at other times, the result is undefined and may result in an IllegalStateException.
If called when the current node is an unnamed namespace node (a node representing the default namespace)
the returned value is -1.
Returns: the nameCode. The nameCode can be used to obtain the prefix, local name,
and namespace URI from the name pool.
Get the namespace declarations associated with the current element. This method must
be called only after a START_ELEMENT event has been notified. In the case of a top-level
START_ELEMENT event (that is, an element that either has no parent node, or whose parent
is not included in the sequence being read), the NamespaceDeclarations object returned
will contain a namespace declaration for each namespace that is in-scope for this element
node. In the case of a non-top-level element, the NamespaceDeclarations will contain
a set of namespace declarations and undeclarations, representing the differences between
this element and its parent.
It is permissible for this method to return namespace declarations that are redundant.
The NamespaceDeclarations object is guaranteed to remain unchanged until the next START_ELEMENT
event, but may then be overwritten. The object should not be modified by the client.
Namespaces may be read before or after reading the attributes of an element,
but must not be read after the first child node has been read, or after calling
one of the methods skipToEnd(), getStringValue(), or getTypedValue().
*
Get configuration information.
public SourceLocator getSourceLocator()
Get the location of the current event.
For an event stream representing a real document, the location information
should identify the location in the lexical XML source. For a constructed document, it should
identify the location in the query or stylesheet that caused the node to be created.
A value of null can be returned if no location information is available.
public CharSequence getStringValue()
Get the string value of the current element, text node, processing-instruction,
or top-level attribute or namespace node, or atomic value.
In other situations the result is undefined and may result in an IllegalStateException.
If the most recent event was a PullFromIterator, this method causes the content
of the element to be read. The current event on completion of this method will be the
corresponding PullFromIterator. The next call of next() will return the event following
the END_ELEMENT event.
Returns: the String Value of the node in question, defined according to the rules in the
XPath data model.
public int getTypeAnnotation()
Get the type annotation of the current attribute or element node, or atomic value.
The result of this method is undefined unless the most recent event was START_ELEMENT,
ATTRIBUTE, or ATOMIC_VALUE.
Returns: the type annotation. This code is the fingerprint of a type name, which may be
resolved to a SchemaType by access to the Configuration.
public int next()
Get the next event
Returns: an integer code indicating the type of event. The code
PullFromIterator is returned at the end of the sequence.
Set configuration information. This must only be called before any events
have been read.
public int skipToMatchingEnd()
Skip the current subtree. This method may be called only immediately after
a START_DOCUMENT or START_ELEMENT event. This call returns the matching
END_DOCUMENT or END_ELEMENT event; the next call on next() will return
the event following the END_DOCUMENT or END_ELEMENT.