win32transservice.cpp

来自「IBM的解析xml的工具Xerces的源代码」· C++ 代码 · 共 1,004 行 · 第 1/2 页

XMLTranscoder*
Win32TransService::makeNewXMLTranscoder(const   XMLCh* const            encodingName
                                        ,       XMLTransService::Codes& resValue
                                        , const unsigned int            blockSize
                                        ,       MemoryManager* const    manager)
{
    const unsigned int upLen = 1024;
    XMLCh upEncoding[upLen + 1];

    //
    //  Get an upper cased copy of the encoding name, since we use a hash
    //  table and we store them all in upper case.
    //
    ::wcsncpy(upEncoding, encodingName, upLen);
    upEncoding[upLen] = 0;
    _wcsupr(upEncoding);

    // Now to try to find this guy in the CP map
    CPMapEntry* theEntry = fCPMap->get(upEncoding);

    // If not found, then return a null pointer
    if (!theEntry)
    {
        resValue = XMLTransService::UnsupportedEncoding;
        return 0;
    }

    // We found it, so return a Win32 transcoder for this encoding
    return new (manager) Win32Transcoder
    (
        encodingName
        , theEntry->getWinCP()
        , theEntry->getIEEncoding()
        , blockSize
        , manager
    );
}








//---------------------------------------------------------------------------
//
//  class Win32Transcoder Implementation ...
//
//---------------------------------------------------------------------------


// ---------------------------------------------------------------------------
//  Win32Transcoder: Constructors and Destructor
// ---------------------------------------------------------------------------
Win32Transcoder::Win32Transcoder(const  XMLCh* const    encodingName
                                , const unsigned int    winCP
                                , const unsigned int    ieCP
                                , const unsigned int    blockSize
                                , MemoryManager* const manager) :

    XMLTranscoder(encodingName, blockSize, manager)
    , fIECP(ieCP)
    , fWinCP(winCP)
{
}

Win32Transcoder::~Win32Transcoder()
{
}


// ---------------------------------------------------------------------------
//  Win32Transcoder: The virtual transcoder API
// ---------------------------------------------------------------------------
unsigned int
Win32Transcoder::transcodeFrom( const   XMLByte* const      srcData
                                , const unsigned int        srcCount
                                ,       XMLCh* const        toFill
                                , const unsigned int        maxChars
                                ,       unsigned int&       bytesEaten
                                ,       unsigned char* const charSizes)
{
    // Get temp pointers to the in and out buffers, and the chars sizes one
    XMLCh*          outPtr = toFill;
    const XMLByte*  inPtr  = srcData;
    unsigned char*  sizesPtr = charSizes;

    // Calc end pointers for each of them
    XMLCh*          outEnd = toFill + maxChars;
    const XMLByte*  inEnd  = srcData + srcCount;

    //
    //  Now loop until we either get our max chars, or cannot get a whole
    //  character from the input buffer.
    //
    bytesEaten = 0;
    while ((outPtr < outEnd) && (inPtr < inEnd))
    {
        //
        //  If we are looking at a leading byte of a multibyte sequence,
        //  then we are going to eat 2 bytes, else 1.
        //
        const unsigned int toEat = ::IsDBCSLeadByteEx(fWinCP, *inPtr) ?
                                    2 : 1;

        // Make sure a whol char is in the source
        if (inPtr + toEat > inEnd)
            break;

        // Try to translate this next char and check for an error
        const unsigned int converted = ::MultiByteToWideChar
        (
            fWinCP
            , MB_PRECOMPOSED | MB_ERR_INVALID_CHARS
            , (const char*)inPtr
            , toEat
            , outPtr
            , 1
        );

        if (converted != 1)
        {
            if (toEat == 1)
            {
                XMLCh tmpBuf[17];
                XMLString::binToText((unsigned int)(*inPtr), tmpBuf, 16, 16, getMemoryManager());
                ThrowXMLwithMemMgr2
                (
                    TranscodingException
                    , XMLExcepts::Trans_BadSrcCP
                    , tmpBuf
                    , getEncodingName()
                    , getMemoryManager()
                );
            }
             else
            {
                ThrowXMLwithMemMgr(TranscodingException, XMLExcepts::Trans_BadSrcSeq, getMemoryManager());
            }
        }

        // Update the char sizes array for this round
        *sizesPtr++ = toEat;

        // And update the bytes eaten count
        bytesEaten += toEat;

        // And update our in/out ptrs
        inPtr += toEat;
        outPtr++;
    }

    // Return the chars we output
    return (outPtr - toFill);
}


unsigned int
Win32Transcoder::transcodeTo(const  XMLCh* const    srcData
                            , const unsigned int    srcCount
                            ,       XMLByte* const  toFill
                            , const unsigned int    maxBytes
                            ,       unsigned int&   charsEaten
                            , const UnRepOpts       options)
{
    // Get pointers to the start and end of each buffer
    const XMLCh*    srcPtr = srcData;
    const XMLCh*    srcEnd = srcData + srcCount;
    XMLByte*        outPtr = toFill;
    XMLByte*        outEnd = toFill + maxBytes;

    //
    //  Now loop until we either get our max chars, or cannot get a whole
    //  character from the input buffer.
    //
    //  NOTE: We have to use a loop for this unfortunately because the
    //  conversion API is too dumb to tell us how many chars it converted if
    //  it couldn't do the whole source.
    //
    BOOL usedDef;
    while ((outPtr < outEnd) && (srcPtr < srcEnd))
    {
        //
        //  Do one char and see if it made it.
        const unsigned int bytesStored = ::WideCharToMultiByte
        (
            fWinCP
            , WC_COMPOSITECHECK | WC_SEPCHARS
            , srcPtr
            , 1
            , (char*)outPtr
            , outEnd - outPtr
            , 0
            , &usedDef
        );

        // If we didn't transcode anything, then we are done
        if (!bytesStored)
            break;

        //
        //  If the defaault char was used and the options indicate that
        //  this isn't allowed, then throw.
        //
        if (usedDef && (options == UnRep_Throw))
        {
            XMLCh tmpBuf[17];
            XMLString::binToText((unsigned int)*srcPtr, tmpBuf, 16, 16, getMemoryManager());
            ThrowXMLwithMemMgr2
            (
                TranscodingException
                , XMLExcepts::Trans_Unrepresentable
                , tmpBuf
                , getEncodingName()
                , getMemoryManager()
            );
        }

        // Update our pointers
        outPtr += bytesStored;
        srcPtr++;
    }

    // Update the chars eaten
    charsEaten = srcPtr - srcData;

    // And return the bytes we stored
    return outPtr - toFill;
}


bool Win32Transcoder::canTranscodeTo(const unsigned int toCheck) const
{
    //
    //  If the passed value is really a surrogate embedded together, then
    //  we need to break it out into its two chars. Else just one.
    //
    XMLCh           srcBuf[2];
    unsigned int    srcCount = 1;
    if (toCheck & 0xFFFF0000)
    {
        srcBuf[0] = XMLCh((toCheck >> 10) + 0xD800);
        srcBuf[1] = XMLCh(toCheck & 0x3FF) + 0xDC00;
        srcCount++;
    }
     else
    {
        srcBuf[0] = XMLCh(toCheck);
    }

    //
    //  Use a local temp buffer that would hold any sane multi-byte char
    //  sequence and try to transcode this guy into it.
    //
    char tmpBuf[64];

    BOOL usedDef;
    const unsigned int bytesStored = ::WideCharToMultiByte
    (
        fWinCP
        , WC_COMPOSITECHECK | WC_SEPCHARS
        , srcBuf
        , srcCount
        , tmpBuf
        , 64
        , 0
        , &usedDef
    );

    if (!bytesStored || usedDef)
        return false;

    return true;
}




//---------------------------------------------------------------------------
//
//  class Win32Transcoder Implementation ...
//
//---------------------------------------------------------------------------

// ---------------------------------------------------------------------------
//  Win32LCPTranscoder: Constructors and Destructor
// ---------------------------------------------------------------------------
Win32LCPTranscoder::Win32LCPTranscoder()
{
}

Win32LCPTranscoder::~Win32LCPTranscoder()
{
}


// ---------------------------------------------------------------------------
//  Win32LCPTranscoder: Implementation of the virtual transcoder interface
// ---------------------------------------------------------------------------
unsigned int Win32LCPTranscoder::calcRequiredSize(const char* const srcText
                                                  , MemoryManager* const manager)
{
    if (!srcText)
        return 0;

    return ::MultiByteToWideChar(CP_ACP, MB_PRECOMPOSED, srcText, -1, NULL, 0);
}


unsigned int Win32LCPTranscoder::calcRequiredSize(const XMLCh* const srcText
                                                  , MemoryManager* const manager)
{
    if (!srcText)
        return 0;

    return ::WideCharToMultiByte(CP_ACP, 0, srcText, -1, NULL, 0, NULL, NULL);
}

// Return value using global operator new
// Revisit: deprecate ?
char* Win32LCPTranscoder::transcode(const XMLCh* const toTranscode)
{
    if (!toTranscode)
        return 0;

    char* retVal = 0;
    if (*toTranscode)
    {
        // Calc the needed size
        const unsigned int neededLen = calcRequiredSize(toTranscode);

        // Allocate a buffer of that size plus one for the null and transcode
        retVal = new char[neededLen + 1];
        ::WideCharToMultiByte(CP_ACP, 0, (LPCWSTR)toTranscode, -1, retVal, neededLen+1, NULL, NULL);

        // And cap it off anyway just to make sure
        retVal[neededLen] = 0;
    }
     else
    {
        retVal = new char[1];
        retVal[0] = 0;
    }
    return retVal;
}

char* Win32LCPTranscoder::transcode(const XMLCh* const toTranscode,
                                    MemoryManager* const manager)
{
    if (!toTranscode)
        return 0;

    char* retVal = 0;
    if (*toTranscode)
    {
        // Calc the needed size
        const unsigned int neededLen = calcRequiredSize(toTranscode, manager);

        // Allocate a buffer of that size plus one for the null and transcode
        retVal = (char*) manager->allocate((neededLen + 1) * sizeof(char)); //new char[neededLen + 1];
        ::WideCharToMultiByte(CP_ACP, 0, (LPCWSTR)toTranscode, -1, retVal, neededLen+1, NULL, NULL);

        // And cap it off anyway just to make sure
        retVal[neededLen] = 0;
    }
     else
    {
        retVal = (char*) manager->allocate(sizeof(char)); //new char[1];
        retVal[0] = 0;
    }
    return retVal;
}

// Return value using global operator new
// Revisit: deprecate ?
XMLCh* Win32LCPTranscoder::transcode(const char* const toTranscode)
{
    if (!toTranscode)
        return 0;

    XMLCh* retVal = 0;
    if (*toTranscode)
    {
        // Calculate the buffer size required
        const unsigned int neededLen = calcRequiredSize(toTranscode);
        if (neededLen == 0)
        {
            retVal = new XMLCh[1];
            retVal[0] = 0;
            return retVal;
        }

        // Allocate a buffer of that size plus one for the null and transcode
        retVal = new XMLCh[neededLen + 1];
        ::MultiByteToWideChar(CP_ACP, MB_PRECOMPOSED, toTranscode, -1, (LPWSTR)retVal, neededLen + 1);

        // Cap it off just to make sure. We are so paranoid!
        retVal[neededLen] = 0;
    }
     else
    {
        retVal = new XMLCh[1];
        retVal[0] = 0;
    }
    return retVal;
}

XMLCh* Win32LCPTranscoder::transcode(const char* const toTranscode,
                                     MemoryManager* const manager)
{
    if (!toTranscode)
        return 0;

    XMLCh* retVal = 0;
    if (*toTranscode)
    {
        // Calculate the buffer size required
        const unsigned int neededLen = calcRequiredSize(toTranscode, manager);
        if (neededLen == 0)
        {
            retVal = (XMLCh*) manager->allocate(sizeof(XMLCh)); //new XMLCh[1];
            retVal[0] = 0;
            return retVal;
        }

        // Allocate a buffer of that size plus one for the null and transcode
        retVal = (XMLCh*) manager->allocate((neededLen + 1) * sizeof(XMLCh)); //new XMLCh[neededLen + 1];
        ::MultiByteToWideChar(CP_ACP, MB_PRECOMPOSED, toTranscode, -1, (LPWSTR)retVal, neededLen + 1);

        // Cap it off just to make sure. We are so paranoid!
        retVal[neededLen] = 0;
    }
     else
    {
        retVal = (XMLCh*) manager->allocate(sizeof(XMLCh)); //new XMLCh[1];
        retVal[0] = 0;
    }
    return retVal;
}


bool Win32LCPTranscoder::transcode( const   char* const     toTranscode
                                    ,       XMLCh* const    toFill
                                    , const unsigned int    maxChars
                                    , MemoryManager* const  manager)
{
    // Check for a couple of psycho corner cases
    if (!toTranscode || !maxChars)
    {
        toFill[0] = 0;
        return true;
    }

    if (!*toTranscode)
    {
        toFill[0] = 0;
        return true;
    }

    // This one has a fixed size output, so try it and if it fails it fails
    if ( 0 == ::MultiByteToWideChar(CP_ACP, MB_PRECOMPOSED, toTranscode, -1, (LPWSTR)toFill, maxChars + 1) )
        return false;
    return true;
}


bool Win32LCPTranscoder::transcode( const   XMLCh* const    toTranscode
                                    ,       char* const     toFill
                                    , const unsigned int    maxBytes
                                    , MemoryManager* const  manager)
{
    // Watch for a couple of pyscho corner cases
    if (!toTranscode || !maxBytes)
    {
        toFill[0] = 0;
        return true;
    }

    if (!*toTranscode)
    {
        toFill[0] = 0;
        return true;
    }

    // This one has a fixed size output, so try it and if it fails it fails
    if ( 0 == ::WideCharToMultiByte(CP_ACP, 0, (LPCWSTR)toTranscode, -1, toFill, maxBytes + 1, NULL, NULL) )
        return false;

    // Cap it off just in case
    toFill[maxBytes] = 0;
    return true;
}


XERCES_CPP_NAMESPACE_END