www.alphaworks.ibm.comwww.ibm.com/developerwww.ibm.com

Home

Readme
Download

Build







Migration

Releases

Feedback

Y2K Compliance


CVS Repository
Mail Archive

API Docs for SAX and DOM
 

Main Page   Class Hierarchy   Alphabetical List   Compound List   File List   Compound Members   File Members  

XMLValidator Class Reference

This abstract class provides the interface for all validators. More...

#include <XMLValidator.hpp>

Collaboration diagram for XMLValidator:

Collaboration graph
class_XMLErrorReporter
[legend]
List of all members.

Public Methods

Destructor
virtual ~XMLValidator ()
 The derived class should clean up its allocated data, then this class will do the same for data allocated at this level.

Virtual validator interface
virtual int checkContent (XMLElementDecl *const elemDecl,QName **const children,const unsigned int childCount)=0
 The derived class should look up its declaration of the passed element from its element pool. More...

virtual bool checkRootElement (const unsigned int elemId)=0
 For those validators that contrain the possible root elements of a document to only particular elements, they should use this call to validate that the passed root element id is a legal root element.

virtual void faultInAttr (XMLAttr &toFill,const XMLAttDef &attDef)const=0
 The derived class should fault in the passed XMLAttr value. More...

virtual void preContentValidation (bool reuseGrammar)=0
 This method is called by the scanner after a Grammar is scanned.

virtual void postParseValidation ()=0
 This method is called by the scanner after the parse has completed. More...

virtual void reset ()=0
 This method is called by the scanner before a new document is about to start. More...

virtual bool requiresNamespaces ()const=0
 The derived class should return a boolean that indicates whether it requires namespace processing or not. More...

virtual void validateAttrValue (const XMLAttDef *attDef,const XMLCh *const attrValue)=0
 The derived class should apply any rules to the passed attribute value that are above and beyond those defined by XML 1.0. More...

virtual void validateElement (const XMLElementDecl *elemDef)=0
 The derived class should apply any rules to the passed element decl that are above and beyond those defined by XML 1.0. More...

virtual Grammar* getGrammar ()const=0
 Retrieve the Grammar used.

virtual void setGrammar (Grammar *aGrammar)=0
 Set the Grammar.

Virtual Schema handler interface
virtual bool handlesDTD ()const=0
 This method allows the scanner to ask the validator if it handles DTDs or not.

virtual bool handlesSchema ()const=0
 This method allows the scanner to ask the validator if it handles Schema or not.

Setter methods
void setScannerInfo (XMLScanner *const owningScanner,ReaderMgr *const readerMgr,XMLBufferMgr *const bufMgr)
void setErrorReporter (XMLErrorReporter *const errorReporter)
 This method is called to set an error reporter on the validator via which it will report any errors it sees during parsing or validation. More...

Error emittor methods
void emitError (const XMLValid::Codes toEmit)
 This call is a convenience by which validators can emit errors. More...

void emitError (const XMLValid::Codes toEmit,const XMLCh *const text1,const XMLCh *const text2=0,const XMLCh *const text3=0,const XMLCh *const text4=0)
void emitError (const XMLValid::Codes toEmit,const char *const text1,const char *const text2=0,const char *const text3=0,const char *const text4=0)

Protected Methods

 XMLValidator (XMLErrorReporter *const errReporter=0)
const XMLBufferMgr* getBufMgr ()const
XMLBufferMgr* getBufMgr ()
const ReaderMgr* getReaderMgr ()const
ReaderMgr* getReaderMgr ()
const XMLScanner* getScanner ()const
XMLScanner* getScanner ()

Detailed Description

This abstract class provides the interface for all validators.

This is the simple amount of API that all validators must honor, in order for the scanner to use them to do validation. All validators will actually contain much more functionality than is accessible via this common API, but that functionality requires that you know what type of validator you are dealing with.

Basically, at this level, the primary concern is to be able to query core information about elements and attributes. Adding decls to the validator requires that you go through the derived interface because they all have their own decl types. At this level, we can return information via the base decl classes, from which each validator derives its own decl classes.


Constructor & Destructor Documentation

XMLValidator::~XMLValidator ( ) [virtual]
 

The derived class should clean up its allocated data, then this class will do the same for data allocated at this level.

XMLValidator::XMLValidator ( XMLErrorReporter *const errReporter = 0 ) [protected]
 


Member Function Documentation

int XMLValidator::checkContent ( XMLElementDecl *const elemDecl,
QName **const children,
const unsigned int childCount ) [pure virtual]
 

The derived class should look up its declaration of the passed element from its element pool.

It should then use the content model description contained in that element declaration to validate that the passed list of child elements are valid for that content model. The count can be zero, indicating no child elements.

Note that whitespace and text content are not validated here. Those are handled by the scanner. So only element ids are provided here.

Parameters:
elemId   The pool id of the element whose content is to be checked.
children   An array of element QName which represent the elements found within the parent element, i.e. the content to be validated.
childCount   The number of elements in the childIds array. It can be zero if the element had none.

bool XMLValidator::checkRootElement ( const unsigned int elemId ) [pure virtual]
 

For those validators that contrain the possible root elements of a document to only particular elements, they should use this call to validate that the passed root element id is a legal root element.

void XMLValidator::emitError ( const XMLValid::Codes toEmit,
const char *const text1,
const char *const text2 = 0,
const char *const text3 = 0,
const char *const text4 = 0 )
 

void XMLValidator::emitError ( const XMLValid::Codes toEmit,
const XMLCh *const text1,
const XMLCh *const text2 = 0,
const XMLCh *const text3 = 0,
const XMLCh *const text4 = 0 )
 

void XMLValidator::emitError ( const XMLValid::Codes toEmit )
 

This call is a convenience by which validators can emit errors.

Most of the grunt work of loading the text, getting the current source location, ect... is handled here.

If the loaded text has replacement parameters, then text strings can be passed. These will be used to replace the tokens {0}, {1}, {2}, and {3} in the order passed. So text1 will replace {0}, text2 will replace {1}, and so forth.

Parameters:
toEmit   The error code to emit. it must be one of the defined validator error codes.
textX   Up to four replacement parameters. They can be provided as either XMLCh strings, or local code page strings which will be transcoded internally.

void XMLValidator::faultInAttr ( XMLAttr & toFill,
const XMLAttDef & attDef ) const [pure virtual]
 

The derived class should fault in the passed XMLAttr value.

It should use the passeed attribute definition (which is passed via the base type so it must often be downcast to the appropriate type for the derived validator class), to fill in the passed attribute. This is done as a performance enhancement since the derived class has more direct access to the information.

XMLBufferMgr * XMLValidator::getBufMgr ( ) [protected]
 

const XMLBufferMgr * XMLValidator::getBufMgr ( ) const [protected]
 

Grammar * XMLValidator::getGrammar ( ) const [pure virtual]
 

Retrieve the Grammar used.

ReaderMgr * XMLValidator::getReaderMgr ( ) [protected]
 

const ReaderMgr * XMLValidator::getReaderMgr ( ) const [protected]
 

XMLScanner * XMLValidator::getScanner ( ) [protected]
 

const XMLScanner * XMLValidator::getScanner ( ) const [protected]
 

bool XMLValidator::handlesDTD ( ) const [pure virtual]
 

This method allows the scanner to ask the validator if it handles DTDs or not.

bool XMLValidator::handlesSchema ( ) const [pure virtual]
 

This method allows the scanner to ask the validator if it handles Schema or not.

void XMLValidator::postParseValidation ( ) [pure virtual]
 

This method is called by the scanner after the parse has completed.

It gives the validator a chance to check certain things that can only be checked after the whole document has been parsed, such as referential integrity of ID/IDREF pairs and so forth. The validator should just issue errors for any problems it finds.

void XMLValidator::preContentValidation ( bool reuseGrammar ) [pure virtual]
 

This method is called by the scanner after a Grammar is scanned.

bool XMLValidator::requiresNamespaces ( ) const [pure virtual]
 

The derived class should return a boolean that indicates whether it requires namespace processing or not.

Some do and some allow it to be optional. This flag is used to control whether the client code's requests to disable namespace processing can be honored or not.

void XMLValidator::reset ( ) [pure virtual]
 

This method is called by the scanner before a new document is about to start.

It gives the validator a change to reset itself in preperation for another validation pass.

void XMLValidator::setErrorReporter ( XMLErrorReporter *const errorReporter )
 

This method is called to set an error reporter on the validator via which it will report any errors it sees during parsing or validation.

This is generally called by the owning scanner.

Parameters:
errorReporter   A pointer to the error reporter to use. This is not adopted, just referenced so the caller remains responsible for its cleanup, if any.

void XMLValidator::setGrammar ( Grammar * aGrammar ) [pure virtual]
 

Set the Grammar.

void XMLValidator::setScannerInfo ( XMLScanner *const owningScanner,
ReaderMgr *const readerMgr,
XMLBufferMgr *const bufMgr )
 

Parameters:
owningScanner   This is a pointer to the scanner to which the validator belongs. The validator will often need to query state data from the scanner.
readerMgr   This is a pointer to the reader manager that is being used by the scanner.
bufMgr   This is the buffer manager of the scanner. This is provided as a convenience so that the validator doesn't have to create its own buffer manager during the parse process.

void XMLValidator::validateAttrValue ( const XMLAttDef * attDef,
const XMLCh *const attrValue ) [pure virtual]
 

The derived class should apply any rules to the passed attribute value that are above and beyond those defined by XML 1.0.

The scanner itself will impose XML 1.0 rules, based on the type of the attribute. This will generally be used to check things such as range checks and other datatype related validation.

If the value breaks any rules as defined by the derived class, it should just issue errors as usual.

void XMLValidator::validateElement ( const XMLElementDecl * elemDef ) [pure virtual]
 

The derived class should apply any rules to the passed element decl that are above and beyond those defined by XML 1.0.

If the value breaks any rules as defined by the derived class, it should just issue errors as usual.


The documentation for this class was generated from the following file:


Copyright © 2000 The Apache Software Foundation. All Rights Reserved.