XMLReader.cpp

/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You 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.
 */

/*
 * $Id$
 */

// ---------------------------------------------------------------------------
//  Includes
// ---------------------------------------------------------------------------
#include <xercesc/internal/XMLReader.hpp>
#include <xercesc/util/BitOps.hpp>
#include <xercesc/util/BinInputStream.hpp>
#include <xercesc/util/PlatformUtils.hpp>
#include <xercesc/util/RuntimeException.hpp>
#include <xercesc/util/TransService.hpp>
#include <xercesc/util/XMLEBCDICTranscoder.hpp>
#include <xercesc/util/XMLString.hpp>
#include <xercesc/util/Janitor.hpp>

XERCES_CPP_NAMESPACE_BEGIN

// ---------------------------------------------------------------------------
//  XMLReader: Query Methods
// ---------------------------------------------------------------------------
//  Checks whether all of the chars in the passed buffer are whitespace or
//  not. Breaks out on the first non-whitespace.
//
bool XMLReader::isAllSpaces(const   XMLCh* const    toCheck
                            , const XMLSize_t       count) const
{
    const XMLCh* curCh = toCheck;
    const XMLCh* endPtr = toCheck + count;
    while (curCh < endPtr)
    {
        if (!(fgCharCharsTable[*curCh++] & gWhitespaceCharMask))
            return false;
    }
    return true;
}


//
//  Checks whether at least one of the chars in the passed buffer are whitespace or
//  not.
//
bool XMLReader::containsWhiteSpace(const   XMLCh* const    toCheck
                            , const XMLSize_t     count) const
{
    const XMLCh* curCh = toCheck;
    const XMLCh* endPtr = toCheck + count;
    while (curCh < endPtr)
    {
        if (fgCharCharsTable[*curCh++] & gWhitespaceCharMask)
            return true;
    }
    return false;
}

//
//  This one is not called terribly often, so call the XMLChar utility
//
bool XMLReader::isPublicIdChar(const XMLCh toCheck) const
{
    if (fXMLVersion == XMLV1_1)
        return XMLChar1_1::isPublicIdChar(toCheck);
    else
        return XMLChar1_0::isPublicIdChar(toCheck);
}

// ---------------------------------------------------------------------------
//  XMLReader: Constructors and Destructor
// ---------------------------------------------------------------------------
XMLReader::XMLReader(const  XMLCh* const          pubId
                    , const XMLCh* const          sysId
                    ,       BinInputStream* const streamToAdopt
                    , const RefFrom               from
                    , const Types                 type
                    , const Sources               source
                    , const bool                  throwAtEnd
                    , const bool                  calculateSrcOfs
                    , const XMLVersion            version
                    ,       MemoryManager* const  manager) :
    fCharIndex(0)
    , fCharsAvail(0)
    , fCurCol(1)
    , fCurLine(1)
    , fEncodingStr(0)
    , fForcedEncoding(false)
    , fNoMore(false)
    , fPublicId(XMLString::replicate(pubId, manager))
    , fRawBufIndex(0)
    , fRawBytesAvail(0)
    , fReaderNum(0xFFFFFFFF)
    , fRefFrom(from)
    , fSentTrailingSpace(false)
    , fSource(source)
    , fSrcOfsBase(0)
    , fSrcOfsSupported(false)
    , fCalculateSrcOfs(calculateSrcOfs)
    , fSystemId(XMLString::replicate(sysId, manager))
    , fStream(streamToAdopt)
    , fSwapped(false)
    , fThrowAtEnd(throwAtEnd)
    , fTranscoder(0)
    , fType(type)
    , fMemoryManager(manager)
{
    setXMLVersion(version);

    // Do an initial load of raw bytes
    refreshRawBuffer();

    // Ask the transcoding service if it supports src offset info
    fSrcOfsSupported = XMLPlatformUtils::fgTransService->supportsSrcOfs();

    //
    //  Use the recognizer class to get a basic sense of what family of
    //  encodings this file is in. We'll start off with a reader of that
    //  type, and update it later if needed when we read the XMLDecl line.
    //
    fEncoding = XMLRecognizer::basicEncodingProbe(fRawByteBuf, fRawBytesAvail);

    #if defined(XERCES_DEBUG)
    if ((fEncoding < XMLRecognizer::Encodings_Min)
    ||  (fEncoding > XMLRecognizer::Encodings_Max))
    {
        ThrowXMLwithMemMgr(RuntimeException, XMLExcepts::Reader_BadAutoEncoding, fMemoryManager);
    }
    #endif

    fEncodingStr = XMLString::replicate(XMLRecognizer::nameForEncoding(fEncoding, fMemoryManager), fMemoryManager);

    // Check whether the fSwapped flag should be set or not
    checkForSwapped();

    //
    //  This will check to see if the first line is an XMLDecl and, if
    //  so, decode that first line manually one character at a time. This
    //  leaves enough characters in the buffer that the high level code
    //  can get through the Decl and call us back with the real encoding.
    //
    doInitDecode();

    //
    //  NOTE: We won't create a transcoder until we either get a call to
    //  setEncoding() or we get a call to refreshCharBuffer() and no
    //  transcoder has been set yet.
    //
}


XMLReader::XMLReader(const  XMLCh* const          pubId
                    , const XMLCh* const          sysId
                    ,       BinInputStream* const streamToAdopt
                    , const XMLCh* const          encodingStr
                    , const RefFrom               from
                    , const Types                 type
                    , const Sources               source
                    , const bool                  throwAtEnd
                    , const bool                  calculateSrcOfs
                    , const XMLVersion            version
                    ,       MemoryManager* const  manager) :
    fCharIndex(0)
    , fCharsAvail(0)
    , fCurCol(1)
    , fCurLine(1)
    , fEncoding(XMLRecognizer::UTF_8)
    , fEncodingStr(0)
    , fForcedEncoding(true)
    , fNoMore(false)
    , fPublicId(XMLString::replicate(pubId, manager))
    , fRawBufIndex(0)
    , fRawBytesAvail(0)
    , fReaderNum(0xFFFFFFFF)
    , fRefFrom(from)
    , fSentTrailingSpace(false)
    , fSource(source)
    , fSrcOfsBase(0)
    , fSrcOfsSupported(false)
    , fCalculateSrcOfs(calculateSrcOfs)
    , fSystemId(XMLString::replicate(sysId, manager))
    , fStream(streamToAdopt)
    , fSwapped(false)
    , fThrowAtEnd(throwAtEnd)
    , fTranscoder(0)
    , fType(type)
    , fMemoryManager(manager)
{
    setXMLVersion(version);

    // Do an initial load of raw bytes
    refreshRawBuffer();

    // Copy the encoding string to our member
    fEncodingStr = XMLString::replicate(encodingStr, fMemoryManager);
    XMLString::upperCaseASCII(fEncodingStr);

    // Ask the transcoding service if it supports src offset info
    fSrcOfsSupported = XMLPlatformUtils::fgTransService->supportsSrcOfs();

    //
    //  Map the passed encoding name to one of our enums. If it does not
    //  match one of the intrinsic encodings, it will come back 'other',
    //  which tells us to create a transcoder based reader.
    //
    fEncoding = XMLRecognizer::encodingForName(fEncodingStr);

    //  test the presence of the BOM and remove it from the source
    switch(fEncoding)
    {
        case XMLRecognizer::UCS_4B :
        case XMLRecognizer::UCS_4L :
        {
            if (fRawBytesAvail > 4 &&
                ((fRawByteBuf[0] == 0x00) && (fRawByteBuf[1] == 0x00) && (fRawByteBuf[2] == 0xFE) && (fRawByteBuf[3] == 0xFF)) ||
                ((fRawByteBuf[0] == 0xFF) && (fRawByteBuf[1] == 0xFE) && (fRawByteBuf[2] == 0x00) && (fRawByteBuf[3] == 0x00))  )
            {
                fRawBufIndex += 4;
            }
            break;
        }
        case XMLRecognizer::UTF_8 :
        {
            // Look at the raw buffer as short chars
            const char* asChars = (const char*)fRawByteBuf;

            if (fRawBytesAvail > XMLRecognizer::fgUTF8BOMLen &&
                XMLString::compareNString(  asChars
                                            , XMLRecognizer::fgUTF8BOM
                                            , XMLRecognizer::fgUTF8BOMLen) == 0)
            {
                fRawBufIndex += XMLRecognizer::fgUTF8BOMLen;
            }
            break;
        }
        case XMLRecognizer::UTF_16B :
        case XMLRecognizer::UTF_16L :
        {
            if (fRawBytesAvail < 2)
                break;

            const UTF16Ch* asUTF16 = (const UTF16Ch*)&fRawByteBuf[fRawBufIndex];
            if ((*asUTF16 == chUnicodeMarker) || (*asUTF16 == chSwappedUnicodeMarker))
            {
                fRawBufIndex += sizeof(UTF16Ch);
            }
            break;
        }
        case XMLRecognizer::EBCDIC:
        case XMLRecognizer::US_ASCII:
        case XMLRecognizer::XERCES_XMLCH:
        case XMLRecognizer::OtherEncoding:
        case XMLRecognizer::Encodings_Count:
        {
            // silence warning about enumeration not being used
            break;
        }
    }

    // Check whether the fSwapped flag should be set or not
    checkForSwapped();

    //
    //  Create a transcoder for the encoding. Since the encoding has been
    //  forced, this will be the one we will use, period.
    //
    XMLTransService::Codes failReason;
    if (fEncoding == XMLRecognizer::OtherEncoding)
    {
        //
        //  fEncodingStr not  pre-recognized, use it
        //  directly for transcoder
        //
        fTranscoder = XMLPlatformUtils::fgTransService->makeNewTranscoderFor
        (
            fEncodingStr
            , failReason
            , kCharBufSize
            , fMemoryManager
        );
    }
     else
    {
        //
        //  Use the recognized fEncoding to create the transcoder
        //
        fTranscoder = XMLPlatformUtils::fgTransService->makeNewTranscoderFor
        (
            fEncoding
            , failReason
            , kCharBufSize
            , fMemoryManager
        );

    }

    if (!fTranscoder)
    {
        ThrowXMLwithMemMgr1
        (
            TranscodingException
            , XMLExcepts::Trans_CantCreateCvtrFor
            , fEncodingStr
            , fMemoryManager
        );
    }

    //
    //  Note that, unlike above, we do not do an initial decode of the
    //  first line. We take the caller's word that the encoding is correct
    //  and just assume that the first bulk decode (kicked off by the first
    //  get of a character) will work.
    //
    //  So we do here the slipping in of the leading space if required.
    //
    if ((fType == Type_PE) && (fRefFrom == RefFrom_NonLiteral))
    {
        // This represents no data from the source
        fCharSizeBuf[fCharsAvail] = 0;
        fCharOfsBuf[fCharsAvail] = 0;
        fCharBuf[fCharsAvail++] = chSpace;
    }
}


XMLReader::XMLReader(const  XMLCh* const          pubId
                    , const XMLCh* const          sysId
                    ,       BinInputStream* const streamToAdopt
                    , XMLRecognizer::Encodings    encodingEnum
                    , const RefFrom               from
                    , const Types                 type
                    , const Sources               source
                    , const bool                  throwAtEnd
                    , const bool                  calculateSrcOfs
                    , const XMLVersion            version
                    ,       MemoryManager* const  manager) :
    fCharIndex(0)
    , fCharsAvail(0)
    , fCurCol(1)
    , fCurLine(1)
    , fEncoding(XMLRecognizer::UTF_8)
    , fEncodingStr(0)
    , fForcedEncoding(true)
    , fNoMore(false)
    , fPublicId(XMLString::replicate(pubId, manager))
    , fRawBufIndex(0)
    , fRawBytesAvail(0)
    , fReaderNum(0xFFFFFFFF)
    , fRefFrom(from)
    , fSentTrailingSpace(false)
    , fSource(source)
    , fSrcOfsBase(0)
    , fSrcOfsSupported(false)
    , fCalculateSrcOfs(calculateSrcOfs)
    , fSystemId(XMLString::replicate(sysId, manager))
    , fStream(streamToAdopt)
    , fSwapped(false)
    , fThrowAtEnd(throwAtEnd)
    , fTranscoder(0)
    , fType(type)
    , fMemoryManager(manager)
{
    setXMLVersion(version);

    // Do an initial load of raw bytes
    refreshRawBuffer();

    // Ask the transcoding service if it supports src offset info
    fSrcOfsSupported = XMLPlatformUtils::fgTransService->supportsSrcOfs();

    //
    //  Use the passed encoding code
    //
    fEncoding = encodingEnum;
    fEncodingStr = XMLString::replicate(XMLRecognizer::nameForEncoding(fEncoding, fMemoryManager), fMemoryManager);

    // Check whether the fSwapped flag should be set or not
    checkForSwapped();

    //
    //  Create a transcoder for the encoding. Since the encoding has been
    //  forced, this will be the one we will use, period.
    //
    XMLTransService::Codes failReason;
    fTranscoder = XMLPlatformUtils::fgTransService->makeNewTranscoderFor
    (
        fEncoding
        , failReason
        , kCharBufSize
        , fMemoryManager
    );

    if (!fTranscoder)
    {
        ThrowXMLwithMemMgr1
        (
            TranscodingException
            , XMLExcepts::Trans_CantCreateCvtrFor
            , fEncodingStr
            , fMemoryManager
        );
    }

    //
    //  Note that, unlike above, we do not do an initial decode of the
    //  first line. We take the caller's word that the encoding is correct
    //  and just assume that the first bulk decode (kicked off by the first
    //  get of a character) will work.
    //
    //  So we do here the slipping in of the leading space if required.
    //
    if ((fType == Type_PE) && (fRefFrom == RefFrom_NonLiteral))
    {
        // This represents no data from the source
        fCharSizeBuf[fCharsAvail] = 0;
        fCharOfsBuf[fCharsAvail] = 0;
        fCharBuf[fCharsAvail++] = chSpace;
    }
}


XMLReader::~XMLReader()
{
    fMemoryManager->deallocate(fEncodingStr);
    fMemoryManager->deallocate(fPublicId);
    fMemoryManager->deallocate(fSystemId);
    delete fStream;
    delete fTranscoder;
}


// ---------------------------------------------------------------------------
//  XMLReader: Character buffer management methods
// ---------------------------------------------------------------------------
XMLFilePos XMLReader::getSrcOffset() const
{
    if (!fSrcOfsSupported || !fCalculateSrcOfs)
        ThrowXMLwithMemMgr(RuntimeException, XMLExcepts::Reader_SrcOfsNotSupported, fMemoryManager);

    //
    //  Take the current source offset and add in the sizes that we've
    //  eaten from the source so far.
    //
    if( fCharIndex == 0 ) {
        return fSrcOfsBase;
    }

    if( fCharIndex < fCharsAvail ) {

        return (fSrcOfsBase + fCharOfsBuf[fCharIndex]);
    }

    return (fSrcOfsBase + fCharOfsBuf[fCharIndex-1] + fCharSizeBuf[fCharIndex-1]);
}


bool XMLReader::refreshCharBuffer()
{
    // If the no more flag is set, then don't both doing anything
    if (fNoMore)
        return false;

    XMLSize_t startInd;

    // See if we have any existing chars.
    const XMLSize_t spareChars = fCharsAvail - fCharIndex;

    // If we are full, then don't do anything.
    if (spareChars == kCharBufSize)
        return true;

    //
    //  If no transcoder has been created yet, then we never saw the
    //  any encoding="" string and the encoding was not forced, so lets
    //  create one now. We know that it won't change now.
    //
    //  However, note that if we autosensed EBCDIC, then we have to
    //  consider it an error if we never got an encoding since we don't
    //  know what variant of EBCDIC it is.
    //
    if (!fTranscoder)
    {
        if (fEncoding == XMLRecognizer::EBCDIC)
            ThrowXMLwithMemMgr(RuntimeException, XMLExcepts::Reader_EncodingStrRequired, fMemoryManager);

        // Ask the transcoding service to make use a transcoder
        XMLTransService::Codes failReason;
        fTranscoder = XMLPlatformUtils::fgTransService->makeNewTranscoderFor
        (
            fEncodingStr
            , failReason
            , kCharBufSize
            , fMemoryManager
        );

        if (!fTranscoder)
        {
            ThrowXMLwithMemMgr1
            (
                TranscodingException
                , XMLExcepts::Trans_CantCreateCvtrFor
                , fEncodingStr
                , fMemoryManager
            );
        }
    }

    //
    //  Add the number of source bytes eaten so far to the base src
    //  offset member.
    //
    if (fCalculateSrcOfs) {
        for (startInd = 0; startInd < fCharIndex; startInd++)
            fSrcOfsBase += fCharSizeBuf[startInd];
    }

    //
    //  If there are spare chars, then move then down to the bottom. We
    //  have to move the char sizes down also.
    //
    startInd = 0;
    if (spareChars)
    {
        for (XMLSize_t index = fCharIndex; index < fCharsAvail; index++)
        {
            fCharBuf[startInd] = fCharBuf[index];
            fCharSizeBuf[startInd] = fCharSizeBuf[index];
            startInd++;
        }
    }

    //
    //  And then get more chars, starting after any spare chars that were
    //  left over from the last time.
    //
    fCharsAvail = xcodeMoreChars
    (
        &fCharBuf[startInd]
        , &fCharSizeBuf[startInd]
        , kCharBufSize - spareChars
    );

    // Add back in the spare chars
    fCharsAvail += spareChars;

    // Reset the buffer index to zero, so we start from the 0th char again
    fCharIndex = 0;

    //
    //  If no chars available, then we have to check for one last thing. If
    //  this is reader for a PE and its not being expanded inside a literal,
    //  then unget a trailing space. We use a boolean to avoid triggering
    //  this more than once.
    //
    if (!fCharsAvail
    &&  (fType == Type_PE)
    &&  (fRefFrom == RefFrom_NonLiteral)
    &&  !fSentTrailingSpace)
    {
        fCharBuf[0] = chSpace;
        fCharsAvail = 1;
        fSentTrailingSpace = true;
    }

    //
    //  If we get here with no more chars, then set the fNoMore flag which
    //  lets us optimize and know without checking that no more chars are
    //  available.
    //
    if (!fCharsAvail)
        fNoMore = true;

    //  Calculate fCharOfsBuf using the elements from fCharBufSize
    if (fCalculateSrcOfs)
    {
        unsigned int last = 0;
        fCharOfsBuf[0] = 0;
        for (XMLSize_t index = 1; index < fCharsAvail; ++index) {
            fCharOfsBuf[index] = last+fCharSizeBuf[index-1];
            last = fCharOfsBuf[index];
            // code was:
            // fCharOfsBuf[index] = fCharOfsBuf[index-1]+fCharSizeBuf[index-1];
            // but on Solaris 64 bit with sun studio 11 this didn't work as
            // every value of fCharOfsBuf[] was 1.
        }
    }

    return (fCharsAvail != 0);
}



// ---------------------------------------------------------------------------
//  XMLReader: Scanning methods
// ---------------------------------------------------------------------------
bool XMLReader::getName(XMLBuffer& toFill, const bool token)
{
    //  Ok, first lets see if we have chars in the buffer. If not, then lets
    //  reload.
    if (fCharIndex == fCharsAvail)
    {
        if (!refreshCharBuffer())
            return false;
    }

    XMLSize_t charIndex_start = fCharIndex;

    //  Lets check the first char for being a first name char. If not, then
    //  what's the point in living mannnn? Just give up now. We only do this
    //  if its a name and not a name token that they want.
    if (!token)
    {
        if (fXMLVersion == XMLV1_1 && ((fCharBuf[fCharIndex] >= 0xD800) && (fCharBuf[fCharIndex] <= 0xDB7F))) {
           // make sure one more char is in the buffer, the transcoder
           // should put only a complete surrogate pair into the buffer
           assert(fCharIndex+1 < fCharsAvail);
           if ((fCharBuf[fCharIndex+1] < 0xDC00) || (fCharBuf[fCharIndex+1] > 0xDFFF))
               return false;

            // Looks ok, so lets eat it
            fCharIndex += 2;
        }
        else {
            if (!isFirstNameChar(fCharBuf[fCharIndex]))
                return false;

            // Looks ok, so lets eat it
            fCharIndex ++;
        }

    }

    //  And now we loop until we run out of data in this reader or we hit
    //  a non-name char.
    while (true)
    {
        if (fXMLVersion == XMLV1_1)
        {
            while (fCharIndex < fCharsAvail)
            {
                //  Check the current char and take it if its a name char. Else
                //  break out.
                if ( (fCharBuf[fCharIndex] >= 0xD800) && (fCharBuf[fCharIndex] <= 0xDB7F) )
                {
                    // make sure one more char is in the buffer, the transcoder
                    // should put only a complete surrogate pair into the buffer
                    assert(fCharIndex+1 < fCharsAvail);
                    if ( (fCharBuf[fCharIndex+1] < 0xDC00) ||
                         (fCharBuf[fCharIndex+1] > 0xDFFF)  )
                        break;
                    fCharIndex += 2;

                }
                else
                {
                    if (!isNameChar(fCharBuf[fCharIndex]))
                        break;
                    fCharIndex++;
                }
            }
        }
        else // XMLV1_0
        {
            while (fCharIndex < fCharsAvail)
            {
                if (!isNameChar(fCharBuf[fCharIndex]))
                    break;
                fCharIndex++;
            }
        }

        // we have to copy the accepted character(s), and update column
        if (fCharIndex != charIndex_start)
        {
            fCurCol += (XMLFileLoc)(fCharIndex - charIndex_start);
            toFill.append(&fCharBuf[charIndex_start], fCharIndex - charIndex_start);
        }

        // something is wrong if there is still something in the buffer
        // or if we don't get no more, then break out.
        if ((fCharIndex < fCharsAvail) ||
             !refreshCharBuffer())
            break;

        charIndex_start = fCharIndex;
    }

    return !toFill.isEmpty();
}

bool XMLReader::getQName(XMLBuffer& toFill, int* colonPosition)
{
    XMLSize_t charIndex_start;
    bool checkNextCharacterForFirstNCName = true;

    // We are only looking for two iterations (i.e. 'NCANAME':'NCNAME').
    // We will stop when we finished scanning for a QName (i.e. either a second
    // colon or an invalid char).
    *colonPosition = -1;
    for (;;) {

        //  Ok, first lets see if we have chars in the buffer. If not, then lets
        //  reload.
        if (fCharIndex == fCharsAvail) {
            if (!refreshCharBuffer()) {
                break;
            }
        }

        charIndex_start = fCharIndex;
        if (checkNextCharacterForFirstNCName) {

            checkNextCharacterForFirstNCName = false;
            //  Lets check the first char for being a first name char. If not, then
            //  what's the point in living mannnn? Just give up now. We only do this
            //  if its a name and not a name token that they want.
            if (fXMLVersion == XMLV1_1
                && ((fCharBuf[fCharIndex] >= 0xD800) && (fCharBuf[fCharIndex] <= 0xDB7F))) {
                // make sure one more char is in the buffer, the transcoder
                // should put only a complete surrogate pair into the buffer
                assert(fCharIndex+1 < fCharsAvail);
                if ((fCharBuf[fCharIndex+1] < 0xDC00) || (fCharBuf[fCharIndex+1] > 0xDFFF))
                    return false;

                // Looks ok, so lets eat it
                fCharIndex += 2;
            }
            else {
                if (!isFirstNCNameChar(fCharBuf[fCharIndex])) {
                    return false;
                }

                // Looks ok, so lets eat it
                fCharIndex++;
            }
        }

        while (fCharIndex < fCharsAvail) {
            //  Check the current char and take it if its a name char. Else
            //  break out.
            if ( (fCharBuf[fCharIndex] >= 0xD800) && (fCharBuf[fCharIndex] <= 0xDB7F) )
            {
                // make sure one more char is in the buffer, the transcoder
                // should put only a complete surrogate pair into the buffer
                assert(fCharIndex+1 < fCharsAvail);
                if ( (fXMLVersion == XMLV1_0) ||
                     (fCharBuf[fCharIndex+1] < 0xDC00) ||
                     (fCharBuf[fCharIndex+1] > 0xDFFF)  ) {
                    break;
                }

                fCharIndex += 2;
                continue;
            }

            if ((fgCharCharsTable[fCharBuf[fCharIndex]] & gNCNameCharMask) == 0)
                break;

            fCharIndex++;
        }

        // we have to copy the accepted character(s), and update column
        if (fCharIndex != charIndex_start)
        {
            fCurCol += (XMLFileLoc)(fCharIndex - charIndex_start);
            toFill.append(&fCharBuf[charIndex_start], fCharIndex - charIndex_start);
        }

        // something is wrong if there is still something in the buffer
        // or if we don't get no more, then break out.
        if (fCharIndex < fCharsAvail) {
            if (fCharBuf[fCharIndex] != chColon) {
                break;
            }

            if (*colonPosition != -1) {
                return false;
            }

            *colonPosition = (int)toFill.getLen();
            toFill.append(chColon);
            fCharIndex++;
            fCurCol++;
            checkNextCharacterForFirstNCName = true;
        }
    }

    if (checkNextCharacterForFirstNCName) {
        return false;
    }

    return !toFill.isEmpty();
}

bool XMLReader::getSpaces(XMLBuffer& toFill)
{
    //
    //  We just loop until we either hit a non-space or the end of this
    //  entity. We return true if we returned because of a non-space and
    //  false if because of end of entity.
    //
    //  NOTE:   We have to maintain line/col info here and we have to do
    //          whitespace normalization if we are not already internalized.
    //
    while (true)
    {
        // Loop through the current chars in the buffer
        while (fCharIndex < fCharsAvail)
        {
            // Get the current char out of the buffer
            XMLCh curCh = fCharBuf[fCharIndex];

            //
            //  See if its a white space char. If so, then process it. Else
            //  we've hit a non-space and need to return.
            //
            if (isWhitespace(curCh))
            {
                // Eat this char
                fCharIndex++;

                //
                //  'curCh' is a whitespace(x20|x9|xD|xA), so we only can have
                //  end-of-line combinations with a leading chCR(xD) or chLF(xA)
                //
                //  100000 x20
                //  001001 x9
                //  001010 chLF
                //  001101 chCR
                //  -----------
                //  000110 == (chCR|chLF) & ~(0x9|0x20)
                //
                //  if the result of thelogical-& operation is
                //  true  : 'curCh' must be xA  or xD
                //  false : 'curCh' must be x20 or x9
                //
                if ( ( curCh & (chCR|chLF) & ~(0x9|0x20) ) == 0 )
                {
                    fCurCol++;
                } else
                {
                    handleEOL(curCh, false);
                }

                // Ok we can add this guy to our buffer
                toFill.append(curCh);
            }
             else
            {
                // Return true to indicate we broke out due to a whitespace
                return true;
            }
        }

        //
        //  We've eaten up the current buffer, so lets try to reload it. If
        //  we don't get anything new, then break out. If we do, then we go
        //  back to the top to keep getting spaces.
        //
        if (!refreshCharBuffer())
            break;
    }
    return false;
}


bool XMLReader::getUpToCharOrWS(XMLBuffer& toFill, const XMLCh toCheck)
{
    while (true)
    {
        // Loop through the current chars in the buffer
        while (fCharIndex < fCharsAvail)
        {
            // Get the current char out of the buffer
            XMLCh curCh = fCharBuf[fCharIndex];

            //
            //  See if its not a white space or our target char, then process
            //  it. Else, we need to return.
            //
            if (!isWhitespace(curCh) && (curCh != toCheck))
            {
                // Eat this char
                fCharIndex++;

                //
                //  'curCh' is not a whitespace(x20|x9|xD|xA), so we only can
                //  have end-of-line combinations with a leading chNEL(x85) or
                //  chLineSeparator(x2028)
                //
                //  0010000000101000 chLineSeparator
                //  0000000010000101 chNEL
                //  ---------------------
                //  1101111101010010 == ~(chNEL|chLineSeparator)
                //
                //  if the result of the logical-& operation is
                //  true  : 'curCh' can not be chNEL or chLineSeparator
                //  false : 'curCh' can be chNEL or chLineSeparator
                //
                if ( curCh & (XMLCh) ~(chNEL|chLineSeparator) )
                {
                    fCurCol++;
                } else
                {
                    handleEOL(curCh, false);
                }

                // Add it to our buffer
                toFill.append(curCh);
            }
             else
            {
                return true;
            }
        }

        //
        //  We've eaten up the current buffer, so lets try to reload it. If
        //  we don't get anything new, then break out. If we do, then we go
        //  back to the top to keep getting spaces.
        //
        if (!refreshCharBuffer())
            break;
    }

    // We never hit any non-space and ate up the whole reader
    return false;

}

bool XMLReader::skipIfQuote(XMLCh& chGotten)
{
    if (fCharIndex == fCharsAvail)
    {
        if (!refreshCharBuffer())
            return false;
    }

    const XMLCh curCh = fCharBuf[fCharIndex];
    if ((curCh == chDoubleQuote) || (curCh == chSingleQuote))
    {
        chGotten = curCh;
        fCharIndex++;
        fCurCol++;
        return true;
    }
    return false;
}


bool XMLReader::skipSpaces(bool& skippedSomething, bool inDecl)
{
    // Remember the current line and column
    XMLFileLoc orgLine = fCurLine;
    XMLFileLoc orgCol  = fCurCol;

    //  We enter a loop where we skip over spaces until we hit the end of
    //  this reader or a non-space value. The return indicates whether we
    //  hit the non-space (true) or the end (false).
    while (true)
    {
        // Loop through the current chars in the buffer
        while (fCharIndex < fCharsAvail)
        {
            //  See if its a white space char. If so, then process it. Else
            //  we've hit a non-space and need to return.
            if (isWhitespace(fCharBuf[fCharIndex]))
            {
                // Get the current char out of the buffer and eat it
                XMLCh curCh = fCharBuf[fCharIndex++];

                //
                //  'curCh' is a whitespace(x20|x9|xD|xA), so we only can have
                //  end-of-line combinations with a leading chCR(xD) or chLF(xA)
                //
                //  100000 x20
                //  001001 x9
                //  001010 chLF
                //  001101 chCR
                //  -----------
                //  000110 == (chCR|chLF) & ~(0x9|0x20)
                //
                //  if the result of the logical-& operation is
                //  true  : 'curCh' must be xA  or xD
                //  false : 'curCh' must be x20 or x9