xmlreader.hpp

经典开源游戏glest的源代码

    //      ask for buffers of text from it and will handle making some
    //      sense of it.
    //
    //  fSwapped
    //      If the encoding is one of the ones we do intrinsically, and its
    //      in a different byte order from our native order, then this is
    //      set to remind us to byte swap it during transcoding.
    //
    //  fThrowAtEnd
    //      Indicates whether the reader manager should throw an end of entity
    //      exception at the end of this reader instance. This is usually
    //      set for top level external entity references. It overrides the
    //      reader manager's global flag that controls throwing at the end
    //      of entities. Defaults to false.
    //
    //  fTranscoder
    //      If the encoding is not one that we handle intrinsically, then
    //      we use an an external transcoder to do it. This class is an
    //      abstraction that allows us to use pluggable external transcoding
    //      services (via XMLTransService in util.)
    //
    //  fType
    //      Indicates whether this reader represents a PE or not. If this
    //      flag is true and the fInLiteral flag is false, then we will put
    //      out an extra space at the end.
    //
    //  fgCharCharsTable;
    //      Pointer to XMLChar table, depends on XML version
    //
    //  fNEL
    //      Boolean indicates if NEL and LSEP should be recognized as NEL
    //
    //  fXMLVersion
    //      Enum to indicate if this Reader is conforming to XML 1.0 or XML 1.1
    // -----------------------------------------------------------------------
    unsigned int                fCharIndex;
    XMLCh                       fCharBuf[kCharBufSize];
    unsigned int                fCharsAvail;
    unsigned char               fCharSizeBuf[kCharBufSize];
    unsigned int                fCharOfsBuf[kCharBufSize];
    XMLSSize_t                  fCurCol;
    XMLSSize_t                  fCurLine;
    XMLRecognizer::Encodings    fEncoding;
    XMLCh*                      fEncodingStr;
    bool                        fForcedEncoding;
    bool                        fNoMore;
    XMLCh*                      fPublicId;
    unsigned int                fRawBufIndex;
    XMLByte                     fRawByteBuf[kRawBufSize];
    unsigned int                fRawBytesAvail;
    unsigned int                fReaderNum;
    RefFrom                     fRefFrom;
    bool                        fSentTrailingSpace;
    Sources                     fSource;
    unsigned int                fSrcOfsBase;
    bool                        fSrcOfsSupported;
    bool                        fCalculateSrcOfs;
    XMLCh*                      fSystemId;
    BinInputStream*             fStream;
    bool                        fSwapped;
    bool                        fThrowAtEnd;
    XMLTranscoder*              fTranscoder;
    Types                       fType;
    XMLByte*                    fgCharCharsTable;
    bool                        fNEL;
    XMLVersion                  fXMLVersion;
    MemoryManager*              fMemoryManager;
};


// ---------------------------------------------------------------------------
//  XMLReader: Public, query methods
// ---------------------------------------------------------------------------
inline bool XMLReader::isNameChar(const XMLCh toCheck) const
{
    return ((fgCharCharsTable[toCheck] & gNameCharMask) != 0);
}

inline bool XMLReader::isNCNameChar(const XMLCh toCheck) const
{
    return ((fgCharCharsTable[toCheck] & gNCNameCharMask) != 0);
}

inline bool XMLReader::isPlainContentChar(const XMLCh toCheck) const
{
    return ((fgCharCharsTable[toCheck] & gPlainContentCharMask) != 0);
}


inline bool XMLReader::isFirstNameChar(const XMLCh toCheck) const
{
    return ((fgCharCharsTable[toCheck] & gFirstNameCharMask) != 0);
}

inline bool XMLReader::isFirstNCNameChar(const XMLCh toCheck) const
{
    return (((fgCharCharsTable[toCheck] & gFirstNameCharMask) != 0) 
            && (toCheck != chColon));
}

inline bool XMLReader::isSpecialStartTagChar(const XMLCh toCheck) const
{
    return ((fgCharCharsTable[toCheck] & gSpecialStartTagCharMask) != 0);
}

inline bool XMLReader::isXMLChar(const XMLCh toCheck) const
{
    return ((fgCharCharsTable[toCheck] & gXMLCharMask) != 0);
}

inline bool XMLReader::isXMLLetter(const XMLCh toCheck) const
{
    return (((fgCharCharsTable[toCheck] & gFirstNameCharMask) != 0)
            && (toCheck != chColon) && (toCheck != chUnderscore));
}

inline bool XMLReader::isWhitespace(const XMLCh toCheck) const
{
    return ((fgCharCharsTable[toCheck] & gWhitespaceCharMask) != 0);
}

inline bool XMLReader::isControlChar(const XMLCh toCheck) const
{
    return ((fgCharCharsTable[toCheck] & gControlCharMask) != 0);
}

// ---------------------------------------------------------------------------
//  XMLReader: Buffer management methods
// ---------------------------------------------------------------------------
inline unsigned long XMLReader::charsLeftInBuffer() const
{
    return fCharsAvail - fCharIndex;
}


// ---------------------------------------------------------------------------
//  XMLReader: Getter methods
// ---------------------------------------------------------------------------
inline XMLSSize_t XMLReader::getColumnNumber() const
{
    return fCurCol;
}

inline const XMLCh* XMLReader::getEncodingStr() const
{
    return fEncodingStr;
}

inline XMLSSize_t XMLReader::getLineNumber() const
{
    return fCurLine;
}

inline bool XMLReader::getNoMoreFlag() const
{
    return fNoMore;
}

inline const XMLCh* XMLReader::getPublicId() const
{
    return fPublicId;
}

inline unsigned int XMLReader::getReaderNum() const
{
    return fReaderNum;
}

inline XMLReader::RefFrom XMLReader::getRefFrom() const
{
    return fRefFrom;
}

inline XMLReader::Sources XMLReader::getSource() const
{
    return fSource;
}

inline const XMLCh* XMLReader::getSystemId() const
{
    return fSystemId;
}

inline bool XMLReader::getThrowAtEnd() const
{
    return fThrowAtEnd;
}

inline XMLReader::Types XMLReader::getType() const
{
    return fType;
}

// ---------------------------------------------------------------------------
//  XMLReader: Setter methods
// ---------------------------------------------------------------------------
inline void XMLReader::setReaderNum(const unsigned int newNum)
{
    fReaderNum = newNum;
}

inline void XMLReader::setThrowAtEnd(const bool newValue)
{
    fThrowAtEnd = newValue;
}

inline void XMLReader::setXMLVersion(const XMLVersion version)
{
    fXMLVersion = version;
    if (version == XMLV1_1) {
        fNEL = true;
        fgCharCharsTable = XMLChar1_1::fgCharCharsTable1_1;
    }
    else {
        fNEL = XMLChar1_0::enableNEL;
        fgCharCharsTable = XMLChar1_0::fgCharCharsTable1_0;
    }

}



// ---------------------------------------------------------------------------
//
//  XMLReader: movePlainContentChars()
//
//       Move as many plain (no special handling of any sort required) content
//       characters as possible from this reader to the supplied destination buffer.
//
//       This is THE hottest performance spot in the parser.
//
// ---------------------------------------------------------------------------
inline void XMLReader::movePlainContentChars(XMLBuffer &dest)
{
    unsigned int count = fCharIndex;

    while (fCharIndex < fCharsAvail)
    {
        if (!isPlainContentChar(fCharBuf[fCharIndex]))
            break;
        fCharIndex++;
    }

    if (count != fCharIndex)
    {
        fCurCol    += (fCharIndex - count);
        dest.append(&fCharBuf[count], fCharIndex - count);
    }
}


// ---------------------------------------------------------------------------
//  XMLReader: getNextCharIfNot() method inlined for speed
// ---------------------------------------------------------------------------
inline bool XMLReader::getNextCharIfNot(const XMLCh chNotToGet, XMLCh& chGotten)
{
    //
    //  See if there is at least a char in the buffer. Else, do the buffer
    //  reload logic.
    //
    if (fCharIndex >= fCharsAvail)
    {
        // If fNoMore is set, then we have nothing else to give
        if (fNoMore)
            return false;

        // Try to refresh
        if (!refreshCharBuffer())
            return false;
    }

    // Check the next char
    if (fCharBuf[fCharIndex] == chNotToGet)
        return false;

    // Its not the one we want to skip so bump the index
    chGotten = fCharBuf[fCharIndex++];

    // Handle end of line normalization and line/col member maintenance.
    //
    // we can have end-of-line combinations with a leading
    // chCR(xD), chLF(xA), chNEL(x85), or chLineSeparator(x2028)
    //
    // 0000000000001101 chCR
    // 0000000000001010 chLF
    // 0000000010000101 chNEL
    // 0010000000101000 chLineSeparator
    // -----------------------
    // 1101111101010000 == ~(chCR|chLF|chNEL|chLineSeparator)
    //
    // if the result of the logical-& operation is
    // true  : 'curCh' can not be chCR, chLF, chNEL or chLineSeparator
    // false : 'curCh' can be chCR, chLF, chNEL or chLineSeparator
    //
    if ( chGotten & (XMLCh) ~(chCR|chLF|chNEL|chLineSeparator) )
    {
        fCurCol++;
    } else
    {
        handleEOL(chGotten, false);
    }

    return true;
}

// ---------------------------------------------------------------------------
//  XMLReader: getNextChar() method inlined for speed
// ---------------------------------------------------------------------------
inline bool XMLReader::getNextChar(XMLCh& chGotten)
{
    //
    //  See if there is at least a char in the buffer. Else, do the buffer
    //  reload logic.
    //
    if (fCharIndex >= fCharsAvail)
    {
        // If fNoMore is set, then we have nothing else to give
        if (fNoMore)
            return false;

        // Try to refresh
        if (!refreshCharBuffer())
            return false;
    }

    chGotten = fCharBuf[fCharIndex++];

    // Handle end of line normalization and line/col member maintenance.
    //
    // we can have end-of-line combinations with a leading
    // chCR(xD), chLF(xA), chNEL(x85), or chLineSeparator(x2028)
    //
    // 0000000000001101 chCR
    // 0000000000001010 chLF
    // 0000000010000101 chNEL
    // 0010000000101000 chLineSeparator
    // -----------------------
    // 1101111101010000 == ~(chCR|chLF|chNEL|chLineSeparator)
    //
    // if the result of the logical-& operation is
    // true  : 'curCh' can not be chCR, chLF, chNEL or chLineSeparator
    // false : 'curCh' can be chCR, chLF, chNEL or chLineSeparator
    //
    if ( chGotten & (XMLCh) ~(chCR|chLF|chNEL|chLineSeparator) )
    {
        fCurCol++;
    } else
    {
        handleEOL(chGotten, false);
    }

    return true;
}


// ---------------------------------------------------------------------------
//  XMLReader: peekNextChar() method inlined for speed
// ---------------------------------------------------------------------------
inline bool XMLReader::peekNextChar(XMLCh& chGotten)
{
    //
    //  If there is something still in the buffer, get it. Else do the reload
    //  scenario.
    //
    if (fCharIndex >= fCharsAvail)
    {
        // Try to refresh the buffer
        if (!refreshCharBuffer())
        {
            chGotten = chNull;
            return false;
        }
    }

    chGotten = fCharBuf[fCharIndex];

    //
    //  Even though we are only peeking, we have to act the same as the
    //  normal char get method in regards to newline normalization, though
    //  its not as complicated as the actual character getting method's.
    //
    if ((chGotten == chCR || (fNEL && (chGotten == chNEL || chGotten == chLineSeparator)))
        && (fSource == Source_External))
        chGotten = chLF;

    return true;
}

XERCES_CPP_NAMESPACE_END

#endif