xmldocumentscannerimpl.java

来自「JAVA 所有包」· Java 代码 · 共 1,499 行 · 第 1/4 页

    //
    
    /**
     * Driver to handle XMLDecl scanning.
     *
     * This class has been modified as per the new design which is more suited to
     * efficiently build pull parser. Lots of performance improvements have been done and
     * the code has been added to support stax functionality/features.
     *
     * @author Neeraj Bajaj, Sun Microsystems.
     *
     * @author Andy Clark, IBM
     */
    protected final class XMLDeclDriver
            implements Driver {
        
        //
        // Driver methods
        //
        
        
        public int next() throws IOException, XNIException {
            if(DEBUG_NEXT){
                System.out.println("NOW IN XMLDeclDriver");
            }
            
            // next driver is prolog regardless of whether there
            // is an XMLDecl in this document
            setScannerState(SCANNER_STATE_PROLOG);
            setDriver(fPrologDriver);
            
            //System.out.println("fEntityScanner = " + fEntityScanner);
            // scan XMLDecl
            try {
                if (fEntityScanner.skipString(xmlDecl)) {
                    fMarkupDepth++;
                    // NOTE: special case where document starts with a PI
                    //       whose name starts with "xml" (e.g. "xmlfoo")
                    if (XMLChar.isName(fEntityScanner.peekChar())) {
                        fStringBuffer.clear();
                        fStringBuffer.append("xml");
                        while (XMLChar.isName(fEntityScanner.peekChar())) {
                            fStringBuffer.append((char)fEntityScanner.scanChar());
                        }
                        String target = fSymbolTable.addSymbol(fStringBuffer.ch, fStringBuffer.offset, fStringBuffer.length);
                        //this function should fill the data.. and set the fEvent object to this event.
                        fStringBuffer.clear() ;
                        scanPIData(target, fStringBuffer);
                        //REVISIT:where else we can set this value to 'true'
                        fEntityManager.fCurrentEntity.mayReadChunks = true;
                        //return PI event since PI was encountered
                        return XMLEvent.PROCESSING_INSTRUCTION ;
                    }
                    // standard XML declaration
                    else {
                        scanXMLDeclOrTextDecl(false);
                        //REVISIT:where else we can set this value to 'true'
                        fEntityManager.fCurrentEntity.mayReadChunks = true;
                        return XMLEvent.START_DOCUMENT;
                    }
                } else{
                    //REVISIT:where else we can set this value to 'true'
                    fEntityManager.fCurrentEntity.mayReadChunks = true;
                    //In both case return the START_DOCUMENT. ony difference is that first block will
                    //cosume the XML declaration if any.
                    return XMLEvent.START_DOCUMENT;
                }
                
                
                //START_OF_THE_DOCUMENT
                
                
            }
            
            // premature end of file
            catch (EOFException e) {
                reportFatalError("PrematureEOF", null);
                return -1;
                //throw e;
            }
            
        }
    } // class XMLDeclDriver
    
    /**
     * Driver to handle prolog scanning.
     *
     * @author Andy Clark, IBM
     */
    protected final class PrologDriver
            implements Driver {
        
        /**
         * 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.
         *
         * @return state representing the section of document just read.
         *
         * @throws IOException  Thrown on i/o error.
         * @throws XNIException Thrown on parse error.
         */
        
        public int next() throws IOException, XNIException {
            //System.out.println("here in next");
            
            if(DEBUG_NEXT){
                System.out.println("NOW IN PrologDriver");
            }
            try {
                do {
                    switch (fScannerState) {
                        case SCANNER_STATE_PROLOG: {
                            fEntityScanner.skipSpaces();
                            if (fEntityScanner.skipChar('<')) {
                                setScannerState(SCANNER_STATE_START_OF_MARKUP);
                            } else if (fEntityScanner.skipChar('&')) {
                                setScannerState(SCANNER_STATE_REFERENCE);
                            } else {
                                setScannerState(SCANNER_STATE_CONTENT);
                            }
                            break;
                        }
                        
                        case SCANNER_STATE_START_OF_MARKUP: {
                            fMarkupDepth++;
                            
                            if (fEntityScanner.skipChar('?')) {
                                setScannerState(SCANNER_STATE_PI);
                            } else if (fEntityScanner.skipChar('!')) {
                                if (fEntityScanner.skipChar('-')) {
                                    if (!fEntityScanner.skipChar('-')) {
                                        reportFatalError("InvalidCommentStart",
                                                null);
                                    }
                                    setScannerState(SCANNER_STATE_COMMENT);
                                } else if (fEntityScanner.skipString(DOCTYPE)) {
                                    setScannerState(SCANNER_STATE_DOCTYPE);
                                    Entity entity = fEntityScanner.getCurrentEntity();
                                    if(entity instanceof Entity.ScannedEntity){
                                        fStartPos=((Entity.ScannedEntity)entity).position;
                                    }
                                    fReadingDTD=true;
                                    if(fDTDDecl == null)
                                        fDTDDecl = new XMLStringBuffer();
                                    fDTDDecl.append("<!DOCTYPE");
                                    
                                } else {
                                    reportFatalError("MarkupNotRecognizedInProlog",
                                            null);
                                }
                            } else if (XMLChar.isNameStart(fEntityScanner.peekChar())) {
                                setScannerState(SCANNER_STATE_ROOT_ELEMENT);
                                setDriver(fContentDriver);
                                //from now onwards this would be handled by fContentDriver,in the same next() call
                                return fContentDriver.next();
                                
                            } else {
                                reportFatalError("MarkupNotRecognizedInProlog",
                                        null);
                            }
                            break;
                        }
                    }
                } while (fScannerState == SCANNER_STATE_PROLOG || fScannerState == SCANNER_STATE_START_OF_MARKUP );
                
                switch(fScannerState){
                    /**
                    //this part is handled by FragmentContentHandler
                    case SCANNER_STATE_ROOT_ELEMENT: {
                        //we have read '<' and beginning of reading the start element tag
                        setScannerState(SCANNER_STATE_START_ELEMENT_TAG);
                        setDriver(fContentDriver);
                        //from now onwards this would be handled by fContentDriver,in the same next() call
                        return fContentDriver.next();
                    }
                     */
                    case SCANNER_STATE_COMMENT: {
                        //this function fills the data..
                        scanComment();
                        setScannerState(SCANNER_STATE_PROLOG);
                        return XMLEvent.COMMENT;
                        //setScannerState(SCANNER_STATE_PROLOG);
                        //break;
                    }
                    case SCANNER_STATE_PI: {
                        fContentBuffer.clear() ;
                        scanPI(fContentBuffer);
                        setScannerState(SCANNER_STATE_PROLOG);
                        return XMLEvent.PROCESSING_INSTRUCTION;
                    }

                    case SCANNER_STATE_DOCTYPE: {
                        
                        if (fDisallowDoctype) {
                            reportFatalError("DoctypeNotAllowed", null);
                        }
                        
                        if (fSeenDoctypeDecl) {
                            reportFatalError("AlreadySeenDoctype", null);
                        }
                        fSeenDoctypeDecl = true;
                        if(fDTDDriver == null){
                            fDTDDriver = new DTDDriver();
                        }
                        
                        // scanDoctypeDecl() sends XNI doctypeDecl event that
                        // in SAX is converted to startDTD() event.
                        if (scanDoctypeDecl()) {
                            setScannerState(SCANNER_STATE_DTD_INTERNAL_DECLS);
                            fSeenInternalSubset = true;
                            setDriver(fContentDriver);
                            return fDTDDriver.next();
                        }
                        
                        /** xxx:check this part again
                        if(fSeenDoctypeDecl){
                            Entity entity = fEntityScanner.getCurrentEntity();
                            if(entity instanceof Entity.ScannedEntity){
                                fEndPos = ((Entity.ScannedEntity)entity).position;
                            }
                            fReadingDTD = false;
                        }
                         */
                        
                        // handle external subset
                        if (fDoctypeSystemId != null) {
                            if (((fValidation || fLoadExternalDTD) 
                                && (fValidationManager == null || !fValidationManager.isCachedDTD()))) {
                                setScannerState(SCANNER_STATE_DTD_EXTERNAL);
                                setDriver(fContentDriver);
                                return fDTDDriver.next();
                                
                            }
                        }
                        else if (fExternalSubsetSource != null) {
                            if (((fValidation || fLoadExternalDTD) 
                                && (fValidationManager == null || !fValidationManager.isCachedDTD()))) {
                                // This handles the case of a DOCTYPE that had neither an internal subset or an external subset.
                                fDTDScanner.setInputSource(fExternalSubsetSource);
                                fExternalSubsetSource = null;
                                setScannerState(SCANNER_STATE_DTD_EXTERNAL_DECLS);
                                setDriver(fContentDriver);
                                return fDTDDriver.next();
                            }                       	
                        }

                        // Send endDTD() call if: 
                        // a) systemId is null or if an external subset resolver could not locate an external subset.
                        // b) "load-external-dtd" and validation are false
                        // c) DTD grammar is cached

                        // in XNI this results in 3 events:  doctypeDecl, startDTD, endDTD
                        // in SAX this results in 2 events: startDTD, endDTD
                        fDTDScanner.setInputSource(null);
                        setScannerState(SCANNER_STATE_PROLOG);
                        return XMLEvent.DTD;
                    }
                    
                    case SCANNER_STATE_CONTENT: {
                        reportFatalError("ContentIllegalInProlog", null);
                        fEntityScanner.scanChar();
                    }
                    case SCANNER_STATE_REFERENCE: {
                        reportFatalError("ReferenceIllegalInProlog", null);
                    }
                    
                    /**
                     * if (complete) {
                     * if (fEntityScanner.scanChar() != '<') {
                     * reportFatalError("RootElementRequired", null);
                     * }
                     * setScannerState(SCANNER_STATE_ROOT_ELEMENT);
                     * setDriver(fContentDriver);
                     * }
                     */
                }
            }
            // premature end of file
            catch (EOFException e) {
                reportFatalError("PrematureEOF", null);
                //xxx  what should be returned here.... ???
                return -1 ;
                //throw e;
            }
            //xxx  what should be returned here.... ???
            return -1;
            
        }
        
        
    } // class PrologDriver
    
    /**
     * Driver to handle the internal and external DTD subsets.
     *
     * @author Andy Clark, IBM
     */
    protected final class DTDDriver
            implements Driver {
        
        //
        // Driver methods
        //
        
        public int next() throws IOException, XNIException{
            // throw new XNIException("DTD Parsing is currently not supported");
            if(DEBUG_NEXT){
                System.out.println("Now in DTD Driver");
            }
            
            dispatch(true);
            
            if(DEBUG_NEXT){
                System.out.println("After calling dispatch(true) -- At this point whole DTD is read.");
            }
            
            //xxx: remove this hack and align this with reusing DTD components
            //currently this routine will only be executed from Stax
            if(fPropertyManager != null){
                dtdGrammarUtil =  new DTDGrammarUtil(((XMLDTDScannerImpl)fDTDScanner).getGrammar(),fSymbolTable, fNamespaceContext);
            }
            
            return XMLEvent.DTD ;
        }
        
        /**
         * Dispatch an XML "event".
         *
         * @param complete True if this driver is intended to scan
         *                 and dispatch as much as possible.
         *
         * @return True if there is more to dispatch either from this
         *          or a another driver.
         *
         * @throws IOException  Thrown on i/o error.
         * @throws XNIException Thrown on parse error.
         */
        public boolean dispatch(boolean complete)
        throws IOException, XNIException {
            fEntityManager.setEntityHandler(null);
            try {
                boolean again;
                XMLResourceIdentifierImpl resourceIdentifier = new XMLResourceIdentifierImpl();
                if( fDTDScanner == null){
                    
                    if (fEntityManager.getEntityScanner() instanceof XML11EntityScanner){
                        fDTDScanner = new XML11DTDScannerImpl();
                    } else
                    
                    fDTDScanner = new XMLDTDScannerImpl();
                    
                    ((XMLDTDScannerImpl)fDTDScanner).reset(fPropertyManager);
                }
                do {
                    again = false;
                    switch (fScannerState) {
                        case SCANNER_STATE_DTD_INTERNAL_DECLS: {
                            // REVISIT: Should there be a feature for
                            //          the "complete" parameter?
                            boolean completeDTD = true;
                            
                            boolean moreToScan = fDTDScanner.scanDTDInternalSubset(completeDTD, fStandalone, fHasExternalDTD && fLoadExternalDTD);
                            Entity entity = fEntityScanner.getCurrentEntity();
                            if(entity instanceof Entity.ScannedEntity){
                                fEndPos=((Entity.ScannedEntity)entity).position;
                            }
                            fReadingDTD=false;
                            if (!moreToScan) {