jdk7u.jaxp.org.apache.xerces.external.impl

Class XMLDocumentFragmentScannerImpl.FragmentContentDriver

java.lang.Object

jdk7u.jaxp.org.apache.xerces.external.impl.XMLDocumentFragmentScannerImpl.FragmentContentDriver

All Implemented Interfaces:

XMLDocumentFragmentScannerImpl.Driver

Direct Known Subclasses:

XMLDocumentScannerImpl.ContentDriver

Enclosing class:

XMLDocumentFragmentScannerImpl

protected class XMLDocumentFragmentScannerImpl.FragmentContentDriver
extends Object
implements XMLDocumentFragmentScannerImpl.Driver

Driver to handle content scanning. This driver is capable of reading the fragment of XML document. When it has finished reading fragment of XML documents, it can pass the job of reading to another driver. This class has been modified as per the new design which is more suited to efficiently build pull parser. Lot of performance improvements have been done and the code has been added to support stax functionality/features.

Author:

Neeraj Bajaj, Sun Microsystems, Andy Clark, IBM, Eric Ye, IBM

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	XMLDocumentFragmentScannerImpl.FragmentContentDriver()

Method Summary

Methods

Modifier and Type	Method and Description
protected void	checkLimit(XMLStringBuffer buffer) Add the count of the content buffer and check if the accumulated value exceeds the limit
void	decideSubState() SCANNER_STATE_CONTENT and SCANNER_STATE_START_OF_MARKUP are two super states of the parser.
protected boolean	elementDepthIsZeroHook() Element depth iz zero.
protected void	endOfFileHook(EOFException e) End of file hook.
int	next() Drives the parser to the next state/event on the input.
protected boolean	scanForDoctypeHook() Scan for DOCTYPE hook.
protected boolean	scanRootElementHook() Scan for root element hook.

Methods inherited from class java.lang.Object

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

Constructor Detail

XMLDocumentFragmentScannerImpl.FragmentContentDriver

protected XMLDocumentFragmentScannerImpl.FragmentContentDriver()

Method Detail

decideSubState

public void decideSubState()
                    throws IOException

SCANNER_STATE_CONTENT and SCANNER_STATE_START_OF_MARKUP are two super states of the parser. At any point of time when in doubt over the current state of the parser, the state should be set to SCANNER_STATE_CONTENT. Parser will automatically revive itself and will set state of the parser to one of its sub state. sub states are defined in the parser on the basis of different XML component like SCANNER_STATE_ENTITY_REFERENCE , SCANNER_STATE_START_ELEMENT, SCANNER_STATE_CDATA etc.. These sub states help the parser to have fine control over the parsing. These are the different milepost, parser stops at each sub state (milepost). Based on this state it is decided if paresr needs to stop at next milepost ??

Throws:

IOException

next

public int next()
         throws IOException,
                XNIException

Drives the parser to the next state/event on the input. Parser is guaranteed to stop at the next state/event. Internally XML document is divided into several states. Each state represents a sections of XML document. When this functions returns normally, it has read the section of XML document and returns the state corresponding to section of document which has been read. For optimizations, a particular driver can read ahead of the section of document (state returned) just read and can maintain a different internal state. State returned corresponds to Stax states.

Specified by:

next in interface XMLDocumentFragmentScannerImpl.Driver

Returns:

state representing the section of document just read.

Throws:

IOException - Thrown on i/o error.

XNIException - Thrown on parse error.

checkLimit

protected void checkLimit(XMLStringBuffer buffer)

Add the count of the content buffer and check if the accumulated value exceeds the limit

Parameters:

buffer - content buffer

scanForDoctypeHook

protected boolean scanForDoctypeHook()
                              throws IOException,
                                     XNIException

Scan for DOCTYPE hook. This method is a hook for subclasses to add code to handle scanning for a the "DOCTYPE" string after the string "

Returns:

True if the "DOCTYPE" was scanned; false if "DOCTYPE" was not scanned.

Throws:

IOException

XNIException

elementDepthIsZeroHook

protected boolean elementDepthIsZeroHook()
                                  throws IOException,
                                         XNIException

Element depth iz zero. This methos is a hook for subclasses to add code to handle when the element depth hits zero. When scanning a document fragment, an element depth of zero is normal. However, when scanning a full XML document, the scanner must handle the trailing miscellanous section of the document after the end of the document's root element.

Returns:

True if the caller should stop and return true which allows the scanner to switch to a new scanning driver. A return value of false indicates that the content driver should continue as normal.

Throws:

IOException

XNIException

scanRootElementHook

protected boolean scanRootElementHook()
                               throws IOException,
                                      XNIException

Scan for root element hook. This method is a hook for subclasses to add code that handles scanning for the root element. When scanning a document fragment, there is no "root" element. However, when scanning a full XML document, the scanner must handle the root element specially.

Returns:

True if the caller should stop and return true which allows the scanner to switch to a new scanning driver. A return value of false indicates that the content driver should continue as normal.

Throws:

IOException

XNIException

endOfFileHook

protected void endOfFileHook(EOFException e)
                      throws IOException,
                             XNIException

End of file hook. This method is a hook for subclasses to add code that handles the end of file. The end of file in a document fragment is OK if the markup depth is zero. However, when scanning a full XML document, an end of file is always premature.

Throws:

IOException

XNIException