wfxmlscanner.cpp

来自「IBM的解析xml的工具Xerces的源代码」· C++ 代码 · 共 1,883 行 · 第 1/5 页

/*
 * Copyright 2002,2003-2004 The Apache Software Foundation.
 * 
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 * 
 *      http://www.apache.org/licenses/LICENSE-2.0
 * 
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

/*
  * $Id: WFXMLScanner.cpp,v 1.25 2004/09/28 21:27:38 peiyongz Exp $
 */


// ---------------------------------------------------------------------------
//  Includes
// ---------------------------------------------------------------------------
#include <xercesc/internal/WFXMLScanner.hpp>
#include <xercesc/util/Janitor.hpp>
#include <xercesc/util/RuntimeException.hpp>
#include <xercesc/util/UnexpectedEOFException.hpp>
#include <xercesc/sax/InputSource.hpp>
#include <xercesc/framework/XMLDocumentHandler.hpp>
#include <xercesc/framework/XMLEntityHandler.hpp>
#include <xercesc/framework/XMLPScanToken.hpp>
#include <xercesc/framework/XMLValidityCodes.hpp>
#include <xercesc/internal/EndOfEntityException.hpp>
#include <xercesc/util/OutOfMemoryException.hpp>

XERCES_CPP_NAMESPACE_BEGIN

// ---------------------------------------------------------------------------
//  WFXMLScanner: Constructors and Destructor
// ---------------------------------------------------------------------------
WFXMLScanner::WFXMLScanner( XMLValidator* const  valToAdopt
                          , GrammarResolver* const grammarResolver
                          , MemoryManager* const manager) :

    XMLScanner(valToAdopt, grammarResolver, manager)
    , fElementIndex(0)
    , fElements(0)
    , fEntityTable(0)
    , fAttrNameHashList(0)
    , fAttrNSList(0)
    , fElementLookup(0)
{
    try
    {
        commonInit();
    }
    catch(const OutOfMemoryException&)
    {
        throw;
    }
    catch(...)
    {
        cleanUp();
        throw;
    }
}

WFXMLScanner::WFXMLScanner( XMLDocumentHandler* const docHandler
                          , DocTypeHandler* const     docTypeHandler
                          , XMLEntityHandler* const   entityHandler
                          , XMLErrorReporter* const   errHandler
                          , XMLValidator* const       valToAdopt
                          , GrammarResolver* const    grammarResolver
                          , MemoryManager* const      manager) :

    XMLScanner(docHandler, docTypeHandler, entityHandler, errHandler, valToAdopt, grammarResolver, manager)
    , fElementIndex(0)
    , fElements(0)
    , fEntityTable(0)
    , fAttrNameHashList(0)
    , fAttrNSList(0)
    , fElementLookup(0)
{
    try
    {	
        commonInit();
    }
    catch(const OutOfMemoryException&)
    {
        throw;
    }
    catch(...)
    {
        cleanUp();
        throw;
    }
}

WFXMLScanner::~WFXMLScanner()
{
    cleanUp();
}

// ---------------------------------------------------------------------------
//  XMLScanner: Getter methods
// ---------------------------------------------------------------------------
NameIdPool<DTDEntityDecl>* WFXMLScanner::getEntityDeclPool()
{
    return 0;
}

const NameIdPool<DTDEntityDecl>* WFXMLScanner::getEntityDeclPool() const
{
    return 0;
}

// ---------------------------------------------------------------------------
//  WFXMLScanner: Main entry point to scan a document
// ---------------------------------------------------------------------------
void WFXMLScanner::scanDocument(const InputSource& src)
{
    //  Bump up the sequence id for this parser instance. This will invalidate
    //  any previous progressive scan tokens.
    fSequenceId++;

    try
    {
        //  Reset the scanner and its plugged in stuff for a new run. This
        //  resets all the data structures, creates the initial reader and
        //  pushes it on the stack, and sets up the base document path.
        scanReset(src);

        // If we have a document handler, then call the start document
        if (fDocHandler)
            fDocHandler->startDocument();

        //  Scan the prolog part, which is everything before the root element
        //  including the DTD subsets.
        scanProlog();

        //  If we got to the end of input, then its not a valid XML file.
        //  Else, go on to scan the content.
        if (fReaderMgr.atEOF())
        {
            emitError(XMLErrs::EmptyMainEntity);
        }
        else
        {
            // Scan content, and tell it its not an external entity
            if (scanContent())
            {
                // That went ok, so scan for any miscellaneous stuff
                if (!fReaderMgr.atEOF())
                    scanMiscellaneous();
            }
        }

        // If we have a document handler, then call the end document
        if (fDocHandler)
            fDocHandler->endDocument();

        // Reset the reader manager to close all files, sockets, etc...
        fReaderMgr.reset();
    }
    //  NOTE:
    //
    //  In all of the error processing below, the emitError() call MUST come
    //  before the flush of the reader mgr, or it will fail because it tries
    //  to find out the position in the XML source of the error.
    catch(const XMLErrs::Codes)
    {
        // This is a 'first fatal error' type exit, so reset and fall through
        fReaderMgr.reset();
    }
    catch(const XMLValid::Codes)
    {
        // This is a 'first fatal error' type exit, so reset and fall through
        fReaderMgr.reset();
    }
    catch(const XMLException& excToCatch)
    {
        //  Emit the error and catch any user exception thrown from here. Make
        //  sure in all cases we flush the reader manager.
        fInException = true;
        try
        {
            if (excToCatch.getErrorType() == XMLErrorReporter::ErrType_Warning)
                emitError
                (
                    XMLErrs::XMLException_Warning
                    , excToCatch.getType()
                    , excToCatch.getMessage()
                );
            else if (excToCatch.getErrorType() >= XMLErrorReporter::ErrType_Fatal)
                emitError
                (
                    XMLErrs::XMLException_Fatal
                    , excToCatch.getType()
                    , excToCatch.getMessage()
                );
            else
                emitError
                (
                    XMLErrs::XMLException_Error
                    , excToCatch.getType()
                    , excToCatch.getMessage()
                );
        }
        catch(const OutOfMemoryException&)
        {
            throw;
        }
        catch(...)
        {
            // Flush the reader manager and rethrow user's error
            fReaderMgr.reset();
            throw;
        }

        // If it returned, then reset the reader manager and fall through
        fReaderMgr.reset();
    }
    catch(const OutOfMemoryException&)
    {
        throw;
    }
    catch(...)
    {
        // Reset and rethrow
        fReaderMgr.reset();
        throw;
    }
}


bool WFXMLScanner::scanNext(XMLPScanToken& token)
{
    // Make sure this token is still legal
    if (!isLegalToken(token))
        ThrowXMLwithMemMgr(RuntimeException, XMLExcepts::Scan_BadPScanToken, fMemoryManager);

    // Find the next token and remember the reader id
    unsigned int orgReader;
    XMLTokens curToken;
    bool retVal = true;

    try
    {
        while (true)
        {
            //  We have to handle any end of entity exceptions that happen here.
            //  We could be at the end of X nested entities, each of which will
            //  generate an end of entity exception as we try to move forward.
            try
            {
                curToken = senseNextToken(orgReader);
                break;
            }
            catch(const EndOfEntityException& toCatch)
            {
                // Send an end of entity reference event
                if (fDocHandler)
                    fDocHandler->endEntityReference(toCatch.getEntity());
            }
        }

        if (curToken == Token_CharData)
        {
            scanCharData(fCDataBuf);
        }
        else if (curToken == Token_EOF)
        {
            if (!fElemStack.isEmpty())
            {
                const ElemStack::StackElem* topElem = fElemStack.popTop();
                emitError
                (
                    XMLErrs::EndedWithTagsOnStack
                    , topElem->fThisElement->getFullName()
                );
            }

            retVal = false;
        }
        else
        {
            // Its some sort of markup
            bool gotData = true;
            switch(curToken)
            {
                case Token_CData :
                    // Make sure we are within content
                    if (fElemStack.isEmpty())
                        emitError(XMLErrs::CDATAOutsideOfContent);
                    scanCDSection();
                    break;

                case Token_Comment :
                    scanComment();
                    break;

                case Token_EndTag :
                    scanEndTag(gotData);
                    break;

                case Token_PI :
                    scanPI();
                    break;

                case Token_StartTag :
                    if (fDoNamespaces)
                        scanStartTagNS(gotData);
                    else
                        scanStartTag(gotData);
                    break;

                default :
                    fReaderMgr.skipToChar(chOpenAngle);
                    break;
            }

            if (orgReader != fReaderMgr.getCurrentReaderNum())
                emitError(XMLErrs::PartialMarkupInEntity);

            // If we hit the end, then do the miscellaneous part
            if (!gotData)
            {
                // That went ok, so scan for any miscellaneous stuff
                scanMiscellaneous();

                if (fDocHandler)
                    fDocHandler->endDocument();
            }
        }
    }
    //  NOTE:
    //
    //  In all of the error processing below, the emitError() call MUST come
    //  before the flush of the reader mgr, or it will fail because it tries
    //  to find out the position in the XML source of the error.
    catch(const XMLErrs::Codes)
    {
        // This is a 'first failure' exception, so reset and return failure
        fReaderMgr.reset();
        return false;
    }
    catch(const XMLValid::Codes)
    {
        // This is a 'first fatal error' type exit, so reset and reuturn failure
        fReaderMgr.reset();
        return false;
    }
    catch(const XMLException& excToCatch)
    {
        //  Emit the error and catch any user exception thrown from here. Make
        //  sure in all cases we flush the reader manager.
        fInException = true;
        try
        {
            if (excToCatch.getErrorType() == XMLErrorReporter::ErrType_Warning)
                emitError
                (
                    XMLErrs::XMLException_Warning
                    , excToCatch.getType()
                    , excToCatch.getMessage()
                );
            else if (excToCatch.getErrorType() >= XMLErrorReporter::ErrType_Fatal)
                emitError
                (
                    XMLErrs::XMLException_Fatal
                    , excToCatch.getType()
                    , excToCatch.getMessage()
                );
            else
                emitError
                (