It can be used to instantiate a validating or non-validating * parser, by setting a member flag.
* * we basically re-use the existing SAX1 parser code, but provide a * new implementation of XMLContentHandler that raises the new * SAX2 style events * */ class PARSERS_EXPORT SAX2XMLReaderImpl : public SAX2XMLReader // , public Parser , public XMLDocumentHandler , public XMLErrorReporter , public XMLEntityHandler , public DocTypeHandler { public : SAX2XMLReaderImpl() ; ~SAX2XMLReaderImpl() ; // ----------------------------------------------------------------------- // Implementation of the XMLDocumentHandler interface // ----------------------------------------------------------------------- /** @name Implementation of the XMLDocumentHandler Interface. */ //@{ /** * This method is used to report all the characters scanned * by the parser. The driver will invoke the 'characters' * method of the user installed SAX Document Handler. * *If any advanced callback handlers are installed, the * corresponding 'docCharacters' method will also be invoked.
* * @param chars A const pointer to a Unicode string representing the * character data. * @param length The length of the Unicode string returned in 'chars'. * @param cdataSection A flag indicating if the characters represent * content from the CDATA section. * @see DocumentHandler#characters */ virtual void docCharacters ( const XMLCh* const chars , const unsigned int length , const bool cdataSection ); /** * This method is used to report any comments scanned by the parser. * This method is a no-op unless, unless an advanced callback handler * is installed, in which case the corresponding 'docComment' method * is invoked. * * @param comment A const pointer to a null terminated Unicode * string representing the comment text. */ virtual void docComment ( const XMLCh* const comment ); /** * This method is used to report any PI scanned by the parser. * *Any PI's occurring before any 'content' are not reported * to any SAX handler as per the specification. However, all * PI's within content are reported via the SAX Document Handler's * 'processingInstruction' method. * *
If any advanced callback handlers are installed, the * corresponding 'docPI' method will be invoked.
* * @param target A const pointer to a Unicode string representing the * target of the PI declaration. * @param data A const pointer to a Unicode string representing the * data of the PI declaration. See the PI production rule * in the XML specification for details. * * @see DocumentHandler#processingInstruction */ virtual void docPI ( const XMLCh* const target , const XMLCh* const data ); /** * This method is used to indicate the end of root element * was just scanned by the parser. Corresponding 'endDocument' * method of the user installed SAX Document Handler will also * be invoked. * *In addition, if any advanced callback handlers are installed, * the corresponding 'endDocument' method is invoked.
* * @see DocumentHandler#endDocument */ virtual void endDocument(); /** * This method is used to indicate the end tag of an element. * The driver will invoke the corresponding 'endElement' method of * the SAX Document Handler interface. * *If any advanced callback handlers are installed, the * corresponding 'endElement' method is also invoked.
* * @param elemDecl A const reference to the object containing element * declaration information. * @param urlId An id referring to the namespace prefix, if * namespaces setting is switched on. * @param isRoot A flag indicating whether this element was the * root element. * @see DocumentHandler#endElement */ virtual void endElement ( const XMLElementDecl& elemDecl , const unsigned int urlId , const bool isRoot ); /** * This method is used to indicate that an end of an entity reference * was just scanned. * *If any advanced callback handlers are installed, the * corresponding 'endEnityReference' method is invoked.
* * @param entDecl A const reference to the object containing the * entity declaration information. */ virtual void endEntityReference ( const XMLEntityDecl& entDecl ); /** * This method is used to report all the whitespace characters, * which are determined to be 'ignorable'. This distinction * between characters is only made, if validation is enabled. * Corresponding 'ignorableWhitespace' method of the user installed * SAX Document Handler interface is called. * *Any whitespace before content is not reported to the SAX * Document Handler method, as per the SAX specification. * However, if any advanced callback handlers are installed, the * corresponding 'ignorableWhitespace' method is invoked.
* * @param chars A const pointer to a Unicode string representing the * ignorable whitespace character data. * @param length The length of the Unicode string 'chars'. * @param cdataSection A flag indicating if the characters represent * content from the CDATA section. * @see DocumentHandler#ignorableWhitespace */ virtual void ignorableWhitespace ( const XMLCh* const chars , const unsigned int length , const bool cdataSection ); /** * This method allows the user installed Document Handler and * any advanced callback handlers to 'reset' themselves. */ virtual void resetDocument(); /** * This method is used to report the start of the parsing process. * The corresponding user installed SAX Document Handler's method * 'startDocument' is invoked. * *If any advanced callback handlers are installed, then the * corresponding 'startDocument' method is also called.
* * @see DocumentHandler#startDocument */ virtual void startDocument(); /** * This method is used to report the start of an element. It is * called at the end of the element, by which time all attributes * specified are also parsed. The corresponding user installed * SAX Document Handler's method 'startElement' is invoked. * *If any advanced callback handlers are installed, then the * corresponding 'startElement' method is also called.
* * @param elemDecl A const reference to the object containing element * declaration information. * @param urlId An id referring to the namespace prefix, if * namespaces setting is switched on. * @param elemPrefix A const pointer to a Unicode string containing * the namespace prefix for this element. Applicable * only when namespace processing is enabled. * @param attrList A const reference to the object containing the * list of attributes just scanned for this element. * @param attrCount A count of number of attributes in the list * specified by the parameter 'attrList'. * @param isEmpty A flag indicating whether this is an empty element * or not. * @param isRoot A flag indicating whether this element was the * root element. * @see DocumentHandler#startElement */ virtual void startElement ( const XMLElementDecl& elemDecl , const unsigned int urlId , const XMLCh* const elemPrefix , const RefVectorOfIf any advanced callback handlers are installed, the * corresponding 'endEnityReference' method is invoked.
* * @param entDecl A const reference to the object containing the * entity declaration information. */ virtual void startEntityReference ( const XMLEntityDecl& entDecl ); /** * This method is used to report the XML decl scanned by the parser. * Refer to the XML specification to see the meaning of parameters. * * This method is a no-op for this SAX driver * implementation. * * @param versionStr A const pointer to a Unicode string representing * version string value. * @param encodingStr A const pointer to a Unicode string representing * the encoding string value. * @param standaloneStr A const pointer to a Unicode string * representing the standalone string value. * @param actualEncodingStr A const pointer to a Unicode string * representing the actual encoding string * value. */ virtual void XMLDecl ( const XMLCh* const versionStr , const XMLCh* const encodingStr , const XMLCh* const standaloneStr , const XMLCh* const actualEncodingStr ); //@} // ----------------------------------------------------------------------- // Implementation of the XMLReader interface // ----------------------------------------------------------------------- /** @name Getter methods */ //@{ /** Get error count from the last parse operation. * * This method returns the error count from the last parse * operation. Note that this count is actually stored in the * scanner, so this method simply returns what the * scanner reports. * * @return number of errors encountered during the latest * parse operation. */ virtual int getErrorCount() const; /** * This method returns the installed content handler. * * @return A pointer to the installed content handler object. */ virtual ContentHandler* getContentHandler() const; /** * This method returns the installed DTD handler. * * @return A pointer to the installed DTD handler object. */ virtual DTDHandler* getDTDHandler() const ; /** * This method returns the installed entity resolver. * * @return A pointer to the installed entity resolver object. */ virtual EntityResolver* getEntityResolver() const ; /** * This method returns the installed error handler. * * @return A pointer to the installed error handler object. */ virtual ErrorHandler* getErrorHandler() const ; /** * This method returns the installed lexical handler. * * @return A pointer to the installed lexical handler object. */ virtual LexicalHandler* getLexicalHandler() const ; /** * This method returns the installed declaration handler. * * @return A pointer to the installed declaration handler object. */ virtual DeclHandler* getDeclarationHandler() const ; /** * This method returns the state of the parser's * exit-on-First-Fatal-Error flag. * * @return true, if the parser is currently configured to * exit on the first fatal error, false otherwise. * * @see #setExitOnFirstFatalError */ virtual bool getExitOnFirstFatalError() const; /** * This method returns the state of the parser's * validation-constraint-fatal flag. * * @return true, if the parser is currently configured to * set validation constraint errors as fatal, false * otherwise. * * @see #setValidationContraintFatal */ virtual bool getValidationConstraintFatal() const; /** @name Implementation of SAX 2.0 interface's. */ //@{ /** * This method invokes the parsing process on the XML file specified * by the InputSource parameter. * * @param source A const reference to the InputSource object which * points to the XML file to be parsed. */ virtual void parse(const InputSource& source); /** * This method invokes the parsing process on the XML file specified by * the Unicode string parameter 'systemId'. * * @param systemId A const XMLCh pointer to the Unicode string which * contains the path to the XML file to be parsed. */ virtual void parse(const XMLCh* const systemId); /** * This method invokes the parsing process on the XML file specified by * the native char* string parameter 'systemId'. * * @param systemId A const char pointer to a native string which * contains the path to the XML file to be parsed. */ virtual void parse(const char* const systemId); /** * This method installs the user specified SAX Document Handler * callback function on parser. * * @param handler A pointer to the document handler to be called * when the parser comes across 'document' events * as per the SAX specification. */ virtual void setContentHandler(ContentHandler* const handler); /** * Allow an application to register a DTD event handler. * * If the application does not register a DTD handler, all DTD * events reported by the SAX parser will be silently ignored (this * is the default behaviour implemented by HandlerBase). * * Applications may register a new or different handler in the middle * of a parse, and the SAX parser must begin using the new handler * immediately. * * @param handler The DTD handler. * @see DTDHandler#DTDHandler * @see HandlerBase#HandlerBase */ virtual void setDTDHandler(DTDHandler* const handler) ; /** * Allow an application to register a custom entity resolver. * * If the application does not register an entity resolver, the * SAX parser will resolve system identifiers and open connections * to entities itself (this is the default behaviour implemented in * DefaultHandler). * * Applications may register a new or different entity resolver * in the middle of a parse, and the SAX parser must begin using * the new resolver immediately. * * @param resolver The object for resolving entities. * @see EntityResolver#EntityResolver * @see DefaultHandler#DefaultHandler */ virtual void setEntityResolver(EntityResolver* const resolver) ; /** * Allow an application to register an error event handler. * * If the application does not register an error event handler, * all error events reported by the SAX parser will be silently * ignored, except for fatalError, which will throw a SAXException * (this is the default behaviour implemented by HandlerBase). * * Applications may register a new or different handler in the * middle of a parse, and the SAX parser must begin using the new * handler immediately. * * @param handler The error handler. * @see ErrorHandler#ErrorHandler * @see SAXException#SAXException * @see HandlerBase#HandlerBase */ virtual void setErrorHandler(ErrorHandler* const handler) ; /** * Allow an application to register a lexical event handler. * * If the application does not register a lexical handler, * all events reported by the SAX parser will be silently * ignored. (this is the default behaviour implemented by HandlerBase). * * Applications may register a new or different handler in the * middle of a parse, and the SAX parser must begin using the new * handler immediately. * * @param handler The error handler. * @see LexicalHandler#LexicalHandler * @see SAXException#SAXException * @see HandlerBase#HandlerBase */ virtual void setLexicalHandler(LexicalHandler* const handler) ; /** * Allow an application to register a declaration event handler. * * If the application does not register a declaration handler, * all events reported by the SAX parser will be silently * ignored. (this is the default behaviour implemented by DefaultHandler). * * Applications may register a new or different handler in the * middle of a parse, and the SAX parser must begin using the new * handler immediately. * * @param handler The DTD declaration handler. * @see DeclHandler#DeclHandler * @see SAXException#SAXException * @see DefaultHandler#DefaultHandler */ virtual void setDeclarationHandler(DeclHandler* const handler); /** * This method allows users to set the parser's behaviour when it * encounters the first fatal error. If set to true, the parser * will exit at the first fatal error. If false, then it will * report the error and continue processing. * *The default value is 'true' and the parser exits on the * first fatal error.
* * @param newState The value specifying whether the parser should * continue or exit when it encounters the first * fatal error. * * @see #getExitOnFirstFatalError */ virtual void setExitOnFirstFatalError(const bool newState); /** * This method allows users to set the parser's behaviour when it * encounters a validtion constraint error. If set to true, and the * the parser is set to exit when it encounter the first fatal error, * the parser will exit at the first encounter. If false, then it will * report the error and continue processing. * *The default value is 'false'.
* * @param newState The value specifying whether the parser should * continue or exit when it encounters a validation * constraint error. * * @see #getValidationConstraintFatal */ virtual void setValidationConstraintFatal(const bool newState); /** * Set the state of any feature in a SAX2 XMLReader. * Supported features in SAX2 for xerces-c are: *The methods in the advanced callback interface represent * Xerces-C extensions. There is no specification for this interface.
* * @param toInstall A pointer to the users advanced callback handler. * * @see #removeAdvDocHandler */ virtual void installAdvDocHandler(XMLDocumentHandler* const toInstall); /** * This method removes the 'advanced' document handler callback from * the underlying parser scanner. If no handler is installed, advanced * callbacks are not invoked by the scanner. * @param toRemove A pointer to the advanced callback handler which * should be removed. * * @see #installAdvDocHandler */ virtual bool removeAdvDocHandler(XMLDocumentHandler* const toRemove); //@} // ----------------------------------------------------------------------- // Progressive scan methods // ----------------------------------------------------------------------- /** @name Progressive scan methods */ //@{ /** Begin a progressive parse operation * * This method is used to start a progressive parse on a XML file. * To continue parsing, subsequent calls must be to the parseNext * method. * * It scans through the prolog and returns a token to be used on * subsequent scanNext() calls. If the return value is true, then the * token is legal and ready for further use. If it returns false, then * the scan of the prolog failed and the token is not going to work on * subsequent scanNext() calls. * * @param systemId A pointer to a Unicode string represting the path * to the XML file to be parsed. * @param toFill A token maintaing state information to maintain * internal consistency between invocation of 'parseNext' * calls. * @param reuseGrammar The flag indicating whether the existing Grammar * should be reused or not for this parsing run. * If true, there cannot be any internal subset. * * @return 'true', if successful in parsing the prolog. It indicates the * user can go ahead with parsing the rest of the file. It * returns 'false' to indicate that the parser could parse the * prolog (which means the token will not be valid.) * * @see #parseNext * @see #parseFirst(char*,...) * @see #parseFirst(InputSource&,...) */ virtual bool parseFirst ( const XMLCh* const systemId , XMLPScanToken& toFill , const bool reuseGrammar = false ); /** Begin a progressive parse operation * * This method is used to start a progressive parse on a XML file. * To continue parsing, subsequent calls must be to the parseNext * method. * * It scans through the prolog and returns a token to be used on * subsequent scanNext() calls. If the return value is true, then the * token is legal and ready for further use. If it returns false, then * the scan of the prolog failed and the token is not going to work on * subsequent scanNext() calls. * * @param systemId A pointer to a regular native string represting * the path to the XML file to be parsed. * @param toFill A token maintaing state information to maintain * internal consIstency between invocation of 'parseNext' * calls. * @param reuseGrammar The flag indicating whether the existing Grammar * should be reused or not for this parsing run. * If true, there cannot be any internal subset. * * @return 'true', if successful in parsing the prolog. It indicates the * user can go ahead with parsing the rest of the file. It * returns 'false' to indicate that the parser could not parse * the prolog. * * @see #parseNext * @see #parseFirst(XMLCh*,...) * @see #parseFirst(InputSource&,...) */ virtual bool parseFirst ( const char* const systemId , XMLPScanToken& toFill , const bool reuseGrammar = false ); /** Begin a progressive parse operation * * This method is used to start a progressive parse on a XML file. * To continue parsing, subsequent calls must be to the parseNext * method. * * It scans through the prolog and returns a token to be used on * subsequent scanNext() calls. If the return value is true, then the * token is legal and ready for further use. If it returns false, then * the scan of the prolog failed and the token is not going to work on * subsequent scanNext() calls. * * @param source A const reference to the InputSource object which * points to the XML file to be parsed. * @param toFill A token maintaing state information to maintain * internal consistency between invocation of 'parseNext' * calls. * @param reuseGrammar The flag indicating whether the existing Grammar * should be reused or not for this parsing run. * If true, there cannot be any internal subset. * * @return 'true', if successful in parsing the prolog. It indicates the * user can go ahead with parsing the rest of the file. It * returns 'false' to indicate that the parser could not parse * the prolog. * * @see #parseNext * @see #parseFirst(XMLCh*,...) * @see #parseFirst(char*,...) */ virtual bool parseFirst ( const InputSource& source , XMLPScanToken& toFill , const bool reuseGrammar = false ); /** Continue a progressive parse operation * * This method is used to continue with progressive parsing of * XML files started by a call to 'parseFirst' method. * * It parses the XML file and stops as soon as it comes across * a XML token (as defined in the XML specification). Relevant * callback handlers are invoked as required by the SAX * specification. * * @param token A token maintaing state information to maintain * internal consistency between invocation of 'parseNext' * calls. * * @return 'true', if successful in parsing the next XML token. * It indicates the user can go ahead with parsing the rest * of the file. It returns 'false' to indicate that the parser * could not find next token as per the XML specification * production rule. * * @see #parseFirst(XMLCh*,...) * @see #parseFirst(char*,...) * @see #parseFirst(InputSource&,...) */ virtual bool parseNext(XMLPScanToken& token); /** Reset the parser after a progressive parse * * If a progressive parse loop exits before the end of the document * is reached, the parser has no way of knowing this. So it will leave * open any files or sockets or memory buffers that were in use at * the time that the parse loop exited. * * The next parse operation will cause these open files and such to * be closed, but the next parse operation might occur at some unknown * future point. To avoid this problem, you should reset the parser if * you exit the loop early. * * If you exited because of an error, then this cleanup will be done * for you. Its only when you exit the file prematurely of your own * accord, because you've found what you wanted in the file most * likely. * * @param token A token maintaing state information to maintain * internal consistency between invocation of 'parseNext' * calls. */ virtual void parseReset(XMLPScanToken& token); //@} private : // ----------------------------------------------------------------------- // Unimplemented constructors and operators // ----------------------------------------------------------------------- SAX2XMLReaderImpl(const SAX2XMLReaderImpl&); void operator=(const SAX2XMLReaderImpl&); // ----------------------------------------------------------------------- // Private data members // // fAttrList // A temporary implementation of the basic SAX2 Attributes // interface. We use this one over and over on each startElement // event to allow SAX-like access to the element attributes. // // fDocHandler // The installed SAX content handler, if any. Null if none. // // fnamespacePrefix // Indicates whether the namespace-prefix feature is on or off. // // fautoValidation // Indicates whether automatic validation is on or off // // fValidation // Indicates whether the 'validation' core features is on or off // // fReuseGrammar // Tells the parser whether it should reuse the grammar or not. // If true, there cannot be any internal subset. // // fStringBuffers // Any temporary strings we need are pulled out of this pool // // fPrefixes // A Stack of the current namespace prefixes that need calls to // endPrefixMapping // // fPrefixCounts // A Stack of the number of prefixes that need endPrefixMapping // calls for that element // // fDTDHandler // The installed SAX DTD handler, if any. Null if none. // // fElemDepth // This is used to track the element nesting depth, so that we can // know when we are inside content. This is so we can ignore char // data outside of content. // // fEntityResolver // The installed SAX entity handler, if any. Null if none. // // fErrorHandler // The installed SAX error handler, if any. Null if none. // // fLexicalHandler // The installed SAX lexical handler, if any. Null if none. // // fDecllHandler // The installed SAX declaration handler, if any. Null if none. // // fAdvDHCount // fAdvDHList // fAdvDHListSize // This is an array of pointers to XMLDocumentHandlers, which is // how we see installed advanced document handlers. There will // usually not be very many at all, so a simple array is used // instead of a collection, for performance. It will grow if needed, // but that is unlikely. // // The count is how many handlers are currently installed. The size // is how big the array itself is (for expansion purposes.) When // count == size, is time to expand. // // fParseInProgress // This flag is set once a parse starts. It is used to prevent // multiple entrance or reentrance of the parser. // // fScanner // The scanner being used by this parser. It is created internally // during construction. // // fHasExternalSubset // Indicate if the document has external DTD subset. // // ----------------------------------------------------------------------- VecAttributesImpl fAttrList ; ContentHandler* fDocHandler ; RefVectorOf