void DecimalDatatypeValidator::checkAdditionalFacetConstraintsBase(MemoryManager* const manager) const
{
DecimalDatatypeValidator *numBase = (DecimalDatatypeValidator*) getBaseValidator();
if (!numBase)
return;
int thisFacetsDefined = getFacetsDefined();
int baseFacetsDefined = numBase->getFacetsDefined();
// check 4.3.11.c1 error: totalDigits > base.totalDigits
// totalDigits != base.totalDigits if (base.fixed)
if (( thisFacetsDefined & DatatypeValidator::FACET_TOTALDIGITS) != 0)
{
if ( (( baseFacetsDefined & DatatypeValidator::FACET_TOTALDIGITS) != 0) &&
( fTotalDigits > numBase->fTotalDigits ))
{
XMLCh value1[BUF_LEN+1];
XMLCh value2[BUF_LEN+1];
XMLString::binToText(fTotalDigits, value1, BUF_LEN, 10, manager);
XMLString::binToText(numBase->fTotalDigits, value2, BUF_LEN, 10, manager);
ThrowXMLwithMemMgr2(InvalidDatatypeFacetException
, XMLExcepts::FACET_totalDigit_base_totalDigit
, value1
, value2
, manager);
}
if ( (( baseFacetsDefined & DatatypeValidator::FACET_TOTALDIGITS) != 0) &&
(( numBase->getFixed() & DatatypeValidator::FACET_TOTALDIGITS) != 0) &&
( fTotalDigits != numBase->fTotalDigits ))
{
XMLCh value1[BUF_LEN+1];
XMLCh value2[BUF_LEN+1];
XMLString::binToText(fTotalDigits, value1, BUF_LEN, 10, manager);
XMLString::binToText(numBase->fTotalDigits, value2, BUF_LEN, 10, manager);
ThrowXMLwithMemMgr2(InvalidDatatypeFacetException
, XMLExcepts::FACET_totalDigit_base_fixed
, value1
, value2
, manager);
}
}
if (( thisFacetsDefined & DatatypeValidator::FACET_FRACTIONDIGITS) != 0)
{
// check question error: fractionDigits > base.fractionDigits ???
if ( (( baseFacetsDefined & DatatypeValidator::FACET_FRACTIONDIGITS) != 0) &&
( fFractionDigits > numBase->fFractionDigits ))
{
XMLCh value1[BUF_LEN+1];
XMLCh value2[BUF_LEN+1];
XMLString::binToText(fFractionDigits, value1, BUF_LEN, 10, manager);
XMLString::binToText(numBase->fFractionDigits, value2, BUF_LEN, 10, manager);
ThrowXMLwithMemMgr2(InvalidDatatypeFacetException
, XMLExcepts::FACET_fractDigit_base_fractDigit
, value1
, value2
, manager);
}
// check question error: fractionDigits > base.totalDigits ???
if ( (( baseFacetsDefined & DatatypeValidator::FACET_TOTALDIGITS) != 0) &&
( fFractionDigits > numBase->fTotalDigits ))
{
XMLCh value1[BUF_LEN+1];
XMLCh value2[BUF_LEN+1];
XMLString::binToText(fFractionDigits, value1, BUF_LEN, 10, manager);
XMLString::binToText(numBase->fTotalDigits, value2, BUF_LEN, 10, manager);
ThrowXMLwithMemMgr2(InvalidDatatypeFacetException
, XMLExcepts::FACET_fractDigit_base_totalDigit
, value1
, value2
, manager);
}
// fractionDigits != base.fractionDigits if (base.fixed)
if ( (( baseFacetsDefined & DatatypeValidator::FACET_FRACTIONDIGITS) != 0) &&
(( numBase->getFixed() & DatatypeValidator::FACET_FRACTIONDIGITS) != 0) &&
( fFractionDigits != numBase->fFractionDigits ))
{
XMLCh value1[BUF_LEN+1];
XMLCh value2[BUF_LEN+1];
XMLString::binToText(fFractionDigits, value1, BUF_LEN, 10, manager);
XMLString::binToText(numBase->fFractionDigits, value2, BUF_LEN, 10, manager);
ThrowXMLwithMemMgr2(InvalidDatatypeFacetException
, XMLExcepts::FACET_fractDigit_base_fixed
, value1
, value2
, manager);
}
}
}
// -----------------------------------------------------------------------
// Abstract interface from AbstractNumericValidator
// -----------------------------------------------------------------------
void DecimalDatatypeValidator::checkContent(const XMLCh* const content
, ValidationContext* const context
, bool asBase
, MemoryManager* const manager)
{
//validate against base validator if any
DecimalDatatypeValidator *pBase = (DecimalDatatypeValidator*) this->getBaseValidator();
if (pBase)
pBase->checkContent(content, context, true, manager);
int thisFacetsDefined = getFacetsDefined();
// we check pattern first
if ( (thisFacetsDefined & DatatypeValidator::FACET_PATTERN ) != 0 )
{
if (getRegex()->matches(content, manager) ==false)
{
ThrowXMLwithMemMgr2(InvalidDatatypeValueException
, XMLExcepts::VALUE_NotMatch_Pattern
, content
, getPattern()
, manager);
}
}
// if this is a base validator, we only need to check pattern facet
// all other facet were inherited by the derived type
if (asBase)
return;
XMLBigDecimal compareDataValue(content, manager);
XMLBigDecimal* compareData = &compareDataValue;
if (getEnumeration())
{
XMLSize_t i=0;
XMLSize_t enumLength = getEnumeration()->size();
for ( ; i < enumLength; i++)
{
if (compareValues(compareData, (XMLBigDecimal*) getEnumeration()->elementAt(i)) ==0 )
break;
}
if (i == enumLength)
ThrowXMLwithMemMgr1(InvalidDatatypeValueException, XMLExcepts::VALUE_NotIn_Enumeration, content, manager);
}
boundsCheck(compareData, manager);
if ( (thisFacetsDefined & DatatypeValidator::FACET_FRACTIONDIGITS) != 0 )
{
if ( compareData->getScale() > fFractionDigits )
{
XMLCh value1[BUF_LEN+1];
XMLCh value2[BUF_LEN+1];
XMLString::binToText(compareData->getScale(), value1, BUF_LEN, 10, manager);
XMLString::binToText(fFractionDigits, value2, BUF_LEN, 10, manager);
ThrowXMLwithMemMgr3(InvalidDatatypeFacetException
, XMLExcepts::VALUE_exceed_fractDigit
, compareData->getRawData()
, value1
, value2
, manager);
}
}
if ( (thisFacetsDefined & DatatypeValidator::FACET_TOTALDIGITS) != 0 )
{
if ( compareData->getTotalDigit() > fTotalDigits )
{
XMLCh value1[BUF_LEN+1];
XMLCh value2[BUF_LEN+1];
XMLString::binToText(compareData->getTotalDigit(), value1, BUF_LEN, 10, manager);
XMLString::binToText(fTotalDigits, value2, BUF_LEN, 10, manager);
ThrowXMLwithMemMgr3(InvalidDatatypeFacetException
, XMLExcepts::VALUE_exceed_totalDigit
, compareData->getRawData()
, value1
, value2
, manager);
}
/***
E2-44 totalDigits
... by restricting it to numbers that are expressible as i x 10^-n
where i and n are integers such that |i| < 10^totalDigits and 0 <= n <= totalDigits.
***/
if ( compareData->getScale() > fTotalDigits )
{
XMLCh value1[BUF_LEN+1];
XMLCh value2[BUF_LEN+1];
XMLString::binToText(compareData->getScale(), value1, BUF_LEN, 10, manager);
XMLString::binToText(fTotalDigits, value2, BUF_LEN, 10, manager);
ThrowXMLwithMemMgr3(InvalidDatatypeFacetException
, XMLExcepts::VALUE_exceed_totalDigit
, compareData->getRawData()
, value1
, value2
, manager);
}
}
}
// ---------------------------------------------------------------------------
// XMLUTF8Transcoder: Implementation of the transcoder API
// ---------------------------------------------------------------------------
XMLSize_t
XMLUTF8Transcoder::transcodeFrom(const XMLByte* const srcData
, const XMLSize_t srcCount
, XMLCh* const toFill
, const XMLSize_t maxChars
, XMLSize_t& bytesEaten
, unsigned char* const charSizes)
{
// Watch for pathological scenario. Shouldn't happen, but...
if (!srcCount || !maxChars)
return 0;
//
// Get pointers to our start and end points of the input and output
// buffers.
//
const XMLByte* srcPtr = srcData;
const XMLByte* srcEnd = srcPtr + srcCount;
XMLCh* outPtr = toFill;
XMLCh* outEnd = outPtr + maxChars;
unsigned char* sizePtr = charSizes;
//
// We now loop until we either run out of input data, or room to store
// output chars.
//
while ((srcPtr < srcEnd) && (outPtr < outEnd))
{
// Special-case ASCII, which is a leading byte value of <= 127
if (*srcPtr <= 127)
{
// Handle ASCII in groups instead of single character at a time.
const XMLByte* srcPtr_save = srcPtr;
const XMLSize_t chunkSize = (srcEnd-srcPtr)<(outEnd-outPtr)?(srcEnd-srcPtr):(outEnd-outPtr);
for(XMLSize_t i=0;i<chunkSize && *srcPtr <= 127;++i)
*outPtr++ = XMLCh(*srcPtr++);
memset(sizePtr,1,srcPtr - srcPtr_save);
sizePtr += srcPtr - srcPtr_save;
if (srcPtr == srcEnd || outPtr == outEnd)
break;
}
// See how many trailing src bytes this sequence is going to require
const unsigned int trailingBytes = gUTFBytes[*srcPtr];
//
// If there are not enough source bytes to do this one, then we
// are done. Note that we done >= here because we are implicitly
// counting the 1 byte we get no matter what.
//
// If we break out here, then there is nothing to undo since we
// haven't updated any pointers yet.
//
if (srcPtr + trailingBytes >= srcEnd)
break;
// Looks ok, so lets build up the value
// or at least let's try to do so--remembering that
// we cannot assume the encoding to be valid:
// first, test first byte
if((gUTFByteIndicatorTest[trailingBytes] & *srcPtr) != gUTFByteIndicator[trailingBytes]) {
char pos[2] = {(char)0x31, 0};
char len[2] = {(char)(trailingBytes+0x31), 0};
char byte[2] = {(char)*srcPtr,0};
ThrowXMLwithMemMgr3(UTFDataFormatException, XMLExcepts::UTF8_FormatError, pos, byte, len, getMemoryManager());
}
/***
* http://www.unicode.org/reports/tr27/
*
* Table 3.1B. lists all of the byte sequences that are legal in UTF-8.
* A range of byte values such as A0..BF indicates that any byte from A0 to BF (inclusive)
* is legal in that position.
* Any byte value outside of the ranges listed is illegal.
* For example,
* the byte sequence <C0 AF> is illegal since C0 is not legal in the 1st Byte column.
* The byte sequence <E0 9F 80> is illegal since in the row
* where E0 is legal as a first byte,
* 9F is not legal as a second byte.
* The byte sequence <F4 80 83 92> is legal, since every byte in that sequence matches
* a byte range in a row of the table (the last row).
*
*
* Table 3.1B. Legal UTF-8 Byte Sequences
* Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
* =========================================================================
* U+0000..U+007F 00..7F
* -------------------------------------------------------------------------
* U+0080..U+07FF C2..DF 80..BF
*
* -------------------------------------------------------------------------
* U+0800..U+0FFF E0 A0..BF 80..BF
* --
*
* U+1000..U+FFFF E1..EF 80..BF 80..BF
*
* --------------------------------------------------------------------------
* U+10000..U+3FFFF F0 90..BF 80..BF 80..BF
* --
* U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF
* U+100000..U+10FFFF F4 80..8F 80..BF 80..BF
* --
* ==========================================================================
*
* Cases where a trailing byte range is not 80..BF are underlined in the table to
* draw attention to them. These occur only in the second byte of a sequence.
*
***/
XMLUInt32 tmpVal = 0;
switch(trailingBytes)
{
case 1 :
// UTF-8: [110y yyyy] [10xx xxxx]
// Unicode: [0000 0yyy] [yyxx xxxx]
//
// 0xC0, 0xC1 has been filtered out
checkTrailingBytes(*(srcPtr+1), 1, 1);
tmpVal = *srcPtr++;
tmpVal <<= 6;
tmpVal += *srcPtr++;
break;
case 2 :
// UTF-8: [1110 zzzz] [10yy yyyy] [10xx xxxx]
// Unicode: [zzzz yyyy] [yyxx xxxx]
//
if (( *srcPtr == 0xE0) && ( *(srcPtr+1) < 0xA0))
{
char byte0[2] = {(char)*srcPtr ,0};
char byte1[2] = {(char)*(srcPtr+1),0};
ThrowXMLwithMemMgr2(UTFDataFormatException
, XMLExcepts::UTF8_Invalid_3BytesSeq
, byte0
, byte1
, getMemoryManager());
}
checkTrailingBytes(*(srcPtr+1), 2, 1);
checkTrailingBytes(*(srcPtr+2), 2, 2);
//
// D36 (a) UTF-8 is the Unicode Transformation Format that serializes
// a Unicode code point as a sequence of one to four bytes,
// as specified in Table 3.1, UTF-8 Bit Distribution.
// (b) An illegal UTF-8 code unit sequence is any byte sequence that
// does not match the patterns listed in Table 3.1B, Legal UTF-8
// Byte Sequences.
// (c) An irregular UTF-8 code unit sequence is a six-byte sequence
// where the first three bytes correspond to a high surrogate,
// and the next three bytes correspond to a low surrogate.
// As a consequence of C12, these irregular UTF-8 sequences shall
// not be generated by a conformant process.
//
//irregular three bytes sequence
// that is zzzzyy matches leading surrogate tag 110110 or
// trailing surrogate tag 110111
// *srcPtr=1110 1101
// *(srcPtr+1)=1010 yyyy or
// *(srcPtr+1)=1011 yyyy
//
// 0xED 1110 1101
// 0xA0 1010 0000
if ((*srcPtr == 0xED) && (*(srcPtr+1) >= 0xA0))
{
char byte0[2] = {(char)*srcPtr, 0};
char byte1[2] = {(char)*(srcPtr+1),0};
ThrowXMLwithMemMgr2(UTFDataFormatException
, XMLExcepts::UTF8_Irregular_3BytesSeq
, byte0
, byte1
, getMemoryManager());
}
tmpVal = *srcPtr++;
tmpVal <<= 6;
tmpVal += *srcPtr++;
tmpVal <<= 6;
tmpVal += *srcPtr++;
break;
case 3 :
// UTF-8: [1111 0uuu] [10uu zzzz] [10yy yyyy] [10xx xxxx]*
// Unicode: [1101 10ww] [wwzz zzyy] (high surrogate)
// [1101 11yy] [yyxx xxxx] (low surrogate)
// * uuuuu = wwww + 1
//
if (((*srcPtr == 0xF0) && (*(srcPtr+1) < 0x90)) ||
((*srcPtr == 0xF4) && (*(srcPtr+1) > 0x8F)) )
{
char byte0[2] = {(char)*srcPtr ,0};
char byte1[2] = {(char)*(srcPtr+1),0};
ThrowXMLwithMemMgr2(UTFDataFormatException
, XMLExcepts::UTF8_Invalid_4BytesSeq
, byte0
, byte1
, getMemoryManager());
}
checkTrailingBytes(*(srcPtr+1), 3, 1);
checkTrailingBytes(*(srcPtr+2), 3, 2);
checkTrailingBytes(*(srcPtr+3), 3, 3);
tmpVal = *srcPtr++;
tmpVal <<= 6;
tmpVal += *srcPtr++;
tmpVal <<= 6;
tmpVal += *srcPtr++;
tmpVal <<= 6;
tmpVal += *srcPtr++;
break;
default: // trailingBytes > 3
/***
* The definition of UTF-8 in Annex D of ISO/IEC 10646-1:2000 also allows
* for the use of five- and six-byte sequences to encode characters that
* are outside the range of the Unicode character set; those five- and
* six-byte sequences are illegal for the use of UTF-8 as a transformation
* of Unicode characters. ISO/IEC 10646 does not allow mapping of unpaired
* surrogates, nor U+FFFE and U+FFFF (but it does allow other noncharacters).
***/
char len[2] = {(char)(trailingBytes+0x31), 0};
char byte[2] = {(char)*srcPtr,0};
ThrowXMLwithMemMgr2(UTFDataFormatException
, XMLExcepts::UTF8_Exceeds_BytesLimit
, byte
, len
, getMemoryManager());
break;
}
// since trailingBytes comes from an array, this logic is redundant
// default :
// ThrowXMLwithMemMgr(TranscodingException, XMLExcepts::Trans_BadSrcSeq);
//}
tmpVal -= gUTFOffsets[trailingBytes];
//
// If it will fit into a single char, then put it in. Otherwise
// encode it as a surrogate pair. If its not valid, use the
// replacement char.
//
if (!(tmpVal & 0xFFFF0000))
{
*sizePtr++ = trailingBytes + 1;
*outPtr++ = XMLCh(tmpVal);
}
else if (tmpVal > 0x10FFFF)
{
//
// If we've gotten more than 32 chars so far, then just break
// out for now and lets process those. When we come back in
// here again, we'll get no chars and throw an exception. This
// way, the error will have a line and col number closer to
// the real problem area.
//
if ((outPtr - toFill) > 32)
break;
ThrowXMLwithMemMgr(TranscodingException, XMLExcepts::Trans_BadSrcSeq, getMemoryManager());
}
else
{
//
// If we have enough room to store the leading and trailing
// chars, then lets do it. Else, pretend this one never
// happened, and leave it for the next time. Since we don't
// update the bytes read until the bottom of the loop, by
// breaking out here its like it never happened.
//
if (outPtr + 1 >= outEnd)
break;
// Store the leading surrogate char
tmpVal -= 0x10000;
*sizePtr++ = trailingBytes + 1;
*outPtr++ = XMLCh((tmpVal >> 10) + 0xD800);
//
// And then the trailing char. This one accounts for no
// bytes eaten from the source, so set the char size for this
// one to be zero.
//
*sizePtr++ = 0;
*outPtr++ = XMLCh((tmpVal & 0x3FF) + 0xDC00);
}
}
// Update the bytes eaten
bytesEaten = srcPtr - srcData;
// Return the characters read
return outPtr - toFill;
}
XMLSize_t
XMLUTF8Transcoder::transcodeTo( const XMLCh* const srcData
, const XMLSize_t srcCount
, XMLByte* const toFill
, const XMLSize_t maxBytes
, XMLSize_t& charsEaten
, const UnRepOpts options)
{
// Watch for pathological scenario. Shouldn't happen, but...
if (!srcCount || !maxBytes)
return 0;
//
// Get pointers to our start and end points of the input and output
// buffers.
//
const XMLCh* srcPtr = srcData;
const XMLCh* srcEnd = srcPtr + srcCount;
XMLByte* outPtr = toFill;
XMLByte* outEnd = toFill + maxBytes;
while (srcPtr < srcEnd)
{
//
// Tentatively get the next char out. We have to get it into a
// 32 bit value, because it could be a surrogate pair.
//
XMLUInt32 curVal = *srcPtr;
//
// If its a leading surrogate, then lets see if we have the trailing
// available. If not, then give up now and leave it for next time.
//
unsigned int srcUsed = 1;
if ((curVal >= 0xD800) && (curVal <= 0xDBFF))
{
if (srcPtr + 1 >= srcEnd)
break;
// Create the composite surrogate pair
curVal = ((curVal - 0xD800) << 10)
+ ((*(srcPtr + 1) - 0xDC00) + 0x10000);
// And indicate that we ate another one
srcUsed++;
}
// Figure out how many bytes we need
unsigned int encodedBytes;
if (curVal < 0x80)
encodedBytes = 1;
else if (curVal < 0x800)
encodedBytes = 2;
else if (curVal < 0x10000)
encodedBytes = 3;
else if (curVal < 0x110000)
encodedBytes = 4;
else
{
// If the options say to throw, then throw
if (options == UnRep_Throw)
{
XMLCh tmpBuf[17];
XMLString::binToText(curVal, tmpBuf, 16, 16, getMemoryManager());
ThrowXMLwithMemMgr2
(
TranscodingException
, XMLExcepts::Trans_Unrepresentable
, tmpBuf
, getEncodingName()
, getMemoryManager()
);
}
// Else, use the replacement character
*outPtr++ = chSpace;
srcPtr += srcUsed;
continue;
}
//
// If we cannot fully get this char into the output buffer,
// then leave it for the next time.
//
if (outPtr + encodedBytes > outEnd)
break;
// We can do it, so update the source index
srcPtr += srcUsed;
//
// And spit out the bytes. We spit them out in reverse order
// here, so bump up the output pointer and work down as we go.
//
outPtr += encodedBytes;
switch(encodedBytes)
{
case 6 : *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL);
curVal >>= 6;
case 5 : *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL);
curVal >>= 6;
case 4 : *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL);
curVal >>= 6;
case 3 : *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL);
curVal >>= 6;
case 2 : *--outPtr = XMLByte((curVal | 0x80UL) & 0xBFUL);
curVal >>= 6;
case 1 : *--outPtr = XMLByte
(
curVal | gFirstByteMark[encodedBytes]
);
}
// Add the encoded bytes back in again to indicate we've eaten them
outPtr += encodedBytes;
}
// Fill in the chars we ate
charsEaten = (srcPtr - srcData);
// And return the bytes we filled in
return (outPtr - toFill);
}
// ---------------------------------------------------------------------------
// XMLASCIITranscoder: Implementation of the transcoder API
// ---------------------------------------------------------------------------
unsigned int
XMLASCIITranscoder::transcodeFrom( const XMLByte* const srcData
, const unsigned int srcCount
, XMLCh* const toFill
, const unsigned int maxChars
, unsigned int& bytesEaten
, unsigned char* const charSizes)
{
// If debugging, make sure that the block size is legal
#if defined(XERCES_DEBUG)
checkBlockSize(maxChars);
#endif
//
// Calculate the max chars we can do here. Its the lesser of the
// max output chars and the source byte count.
//
const unsigned int countToDo = srcCount < maxChars ? srcCount : maxChars;
//
// Now loop through that many source chars and just cast each one
// over to the XMLCh format. Check each source that its really a
// valid ASCI char.
//
const XMLByte* srcPtr = srcData;
XMLCh* outPtr = toFill;
unsigned int countDone = 0;
for (; countDone < countToDo; countDone++)
{
// Do the optimistic work up front
if (*srcPtr < 0x80)
{
*outPtr++ = XMLCh(*srcPtr++);
continue;
}
//
// We got non source encoding char. If we got more than 32 chars,
// the just break out. We'll come back here later to hit this again
// and give an error much closer to the real source position.
//
if (countDone > 32)
break;
XMLCh tmpBuf[17];
XMLString::binToText((unsigned int)*srcPtr, tmpBuf, 16, 16, getMemoryManager());
ThrowXMLwithMemMgr2
(
TranscodingException
, XMLExcepts::Trans_Unrepresentable
, tmpBuf
, getEncodingName()
, getMemoryManager()
);
}
// Set the bytes we ate
bytesEaten = countDone;
// Set the char sizes to the fixed size
memset(charSizes, 1, countDone);
// Return the chars we transcoded
return countDone;
}
/***
* .empty stringPool
* .empty gramamrRegistry
***/
void XMLGrammarPoolImpl::deserializeGrammars(BinInputStream* const binIn)
{
MemoryManager *memMgr = getMemoryManager();
unsigned int stringCount = fStringPool->getStringCount();
if (stringCount)
{
/***
* it contains only the four predefined one, that is ok
* but we need to reset the string before deserialize it
*
***/
if ( stringCount <= 4 )
{
fStringPool->flushAll();
}
else
{
ThrowXMLwithMemMgr(XSerializationException, XMLExcepts::XSer_StringPool_NotEmpty, memMgr);
}
}
RefHashTableOfEnumerator<Grammar> grammarEnum(fGrammarRegistry, false, memMgr);
if (grammarEnum.hasMoreElements())
{
ThrowXMLwithMemMgr(XSerializationException, XMLExcepts::XSer_GrammarPool_NotEmpty, memMgr);
}
// This object will take care of cleaning up if an exception is
// thrown during deserialization.
JanitorMemFunCall<XMLGrammarPoolImpl> cleanup(this, &XMLGrammarPoolImpl::cleanUp);
try
{
XSerializeEngine serEng(binIn, this);
//version information
unsigned int StorerLevel;
serEng>>StorerLevel;
serEng.fStorerLevel = StorerLevel;
// The storer level must match the loader level.
//
if (StorerLevel != (unsigned int)MATCH_GRAMMAR_SERIALIZATION_LEVEL)
{
XMLCh StorerLevelChar[5];
XMLCh LoaderLevelChar[5];
XMLString::binToText(StorerLevel, StorerLevelChar, 4, 10, memMgr);
XMLString::binToText(MATCH_GRAMMAR_SERIALIZATION_LEVEL, LoaderLevelChar, 4, 10, memMgr);
ThrowXMLwithMemMgr2(XSerializationException
, XMLExcepts::XSer_Storer_Loader_Mismatch
, StorerLevelChar
, LoaderLevelChar
, memMgr);
}
//lock status
serEng>>fLocked;
//StringPool, don't use >>
fStringPool->serialize(serEng);
/***
* Deserialize RefHashTableOf<Grammar>* fGrammarRegistry;
***/
XTemplateSerializer::loadObject(&fGrammarRegistry, 29, true, serEng);
}
catch(const OutOfMemoryException&)
{
// This is a special case, because we don't want
// to execute cleanup code on out-of-memory
// conditions.
cleanup.release();
throw;
}
// Everything is OK, so we can release the cleanup object.
cleanup.release();
if (fLocked)
{
createXSModel();
}
}
// ---------------------------------------------------------------------------
// 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;
}
unsigned int
XML256TableTranscoder390::transcodeTo( const XMLCh* const srcData
, const unsigned int srcCount
, XMLByte* const toFill
, const unsigned int maxBytes
, unsigned int& charsEaten
, const UnRepOpts options)
{
// If debugging, make sure that the block size is legal
#if defined(XERCES_DEBUG)
checkBlockSize(maxBytes);
#endif
//
// Calculate the max chars we can do here. Its the lesser of the
// max output chars and the number of chars in the source.
//
const unsigned int countToDo = srcCount < maxBytes ? srcCount : maxBytes;
//
// Loop through the count we have to do and map each char via the
// lookup table.
//
const XMLCh* srcPtr = srcData;
const XMLCh* endPtr = (srcPtr + countToDo);
XMLByte* outPtr = toFill;
XMLByte nextOut;
while (srcPtr < endPtr)
{
//
// Get the next src char out to a temp, then do a binary search
// of the 'to' table for this entry.
//
if ((nextOut = xlatOneTo(*srcPtr)))
{
*outPtr++ = nextOut;
srcPtr++;
continue;
}
//
// Its not representable so, according to the options, either
// throw or use the replacement.
//
if (options == UnRep_Throw)
{
XMLCh tmpBuf[17];
XMLString::binToText((unsigned int)*srcPtr, tmpBuf, 16, 16, getMemoryManager());
ThrowXMLwithMemMgr2
(
TranscodingException
, XMLExcepts::Trans_Unrepresentable
, tmpBuf
, getEncodingName()
, getMemoryManager()
);
}
// Eat the source char and use the replacement char
srcPtr++;
*outPtr++ = 0x3F;
}
// Set the chars eaten
charsEaten = countToDo;
// Return the bytes we transcoded
return countToDo;
}
BinInputStream* XMLURL::makeNewStream() const
{
//
// If its a local host, then we short circuit it and use our own file
// stream support. Otherwise, we just let it fall through and let the
// installed network access object provide a stream.
//
if (fProtocol == XMLURL::File)
{
if (!fHost || !XMLString::compareIStringASCII(fHost, XMLUni::fgLocalHostString))
{
XMLCh* realPath = XMLString::replicate(fPath, fMemoryManager);
ArrayJanitor<XMLCh> basePathName(realPath, fMemoryManager);
//
// Need to manually replace any character reference %xx first
// HTTP protocol will be done automatically by the netaccessor
//
int end = XMLString::stringLen(realPath);
int percentIndex = XMLString::indexOf(realPath, chPercent, 0, fMemoryManager);
while (percentIndex != -1) {
if (percentIndex+2 >= end ||
!isHexDigit(realPath[percentIndex+1]) ||
!isHexDigit(realPath[percentIndex+2]))
{
XMLCh value1[4];
XMLString::moveChars(value1, &(realPath[percentIndex]), 3);
value1[3] = chNull;
ThrowXMLwithMemMgr2(MalformedURLException
, XMLExcepts::XMLNUM_URI_Component_Invalid_EscapeSequence
, realPath
, value1
, fMemoryManager);
}
unsigned int value = (xlatHexDigit(realPath[percentIndex+1]) * 16) + xlatHexDigit(realPath[percentIndex+2]);
realPath[percentIndex] = XMLCh(value);
int i =0;
for (i = percentIndex + 1; i < end - 2 ; i++)
realPath[i] = realPath[i+2];
realPath[i] = chNull;
end = i;
percentIndex = XMLString::indexOf(realPath, chPercent, percentIndex, fMemoryManager);
}
BinFileInputStream* retStrm = new (fMemoryManager) BinFileInputStream(realPath, fMemoryManager);
if (!retStrm->getIsOpen())
{
delete retStrm;
return 0;
}
return retStrm;
}
}
//
// If we don't have have an installed net accessor object, then we
// have to just throw here.
//
if (!XMLPlatformUtils::fgNetAccessor)
ThrowXMLwithMemMgr(MalformedURLException, XMLExcepts::URL_UnsupportedProto, fMemoryManager);
// Else ask the net accessor to create the stream
return XMLPlatformUtils::fgNetAccessor->makeNew(*this);
}
unsigned int
XML88591Transcoder390::transcodeTo(const XMLCh* const srcData
, const unsigned int srcCount
, XMLByte* const toFill
, const unsigned int maxBytes
, unsigned int& charsEaten
, const UnRepOpts options)
{
// If debugging, make sure that the block size is legal
#if defined(XERCES_DEBUG)
checkBlockSize(maxBytes);
#endif
//
// Calculate the max chars we can do here. Its the lesser of the
// max output bytes and the number of chars in the source.
//
const unsigned int countToDo = srcCount < maxBytes ? srcCount : maxBytes;
//
// Loop through the bytes to do and convert over each byte. Its just
// a downcast of the wide char, checking for unrepresentable chars.
//
const XMLCh* srcPtr = srcData;
const XMLCh* srcEnd = srcPtr + countToDo;
XMLByte* destPtr = toFill;
while (srcPtr < srcEnd)
{
// If its legal, take it and jump back to top
if (*srcPtr < 256)
{
*destPtr++ = XMLByte(*srcPtr++);
continue;
}
//
// Its not representable so use a replacement char. According to
// the options, either throw or use the replacement.
//
if (options == UnRep_Throw)
{
XMLCh tmpBuf[17];
XMLString::binToText((unsigned int)*srcPtr, tmpBuf, 16, 16, getMemoryManager());
ThrowXMLwithMemMgr2
(
TranscodingException
, XMLExcepts::Trans_Unrepresentable
, tmpBuf
, getEncodingName()
, getMemoryManager()
);
}
*destPtr++ = 0x1A;
srcPtr++;
}
// Set the chars eaten
charsEaten = countToDo;
// Return the bytes we transcoded
return countToDo;
}
RangeToken* RegxParser::parseCharacterClass(const bool useNRange) {
setParseContext(regexParserStateInBrackets);
processNext();
RangeToken* tok = 0;
bool isNRange = false;
if (getState() == REGX_T_CHAR && getCharData() == chCaret) {
isNRange = true;
processNext();
}
tok = fTokenFactory->createRange();
parserState type;
bool firstLoop = true;
bool wasDecoded;
while ( (type = getState()) != REGX_T_EOF) {
wasDecoded = false;
// single range | from-to-range | subtraction
if (type == REGX_T_CHAR && getCharData() == chCloseSquare && !firstLoop)
break;
XMLInt32 ch = getCharData();
bool end = false;
if (type == REGX_T_BACKSOLIDUS) {
switch(ch) {
case chLatin_d:
case chLatin_D:
case chLatin_w:
case chLatin_W:
case chLatin_s:
case chLatin_S:
case chLatin_i:
case chLatin_I:
case chLatin_c:
case chLatin_C:
{
tok->mergeRanges(getTokenForShorthand(ch));
end = true;
}
break;
case chLatin_p:
case chLatin_P:
{
RangeToken* tok2 = processBacksolidus_pP(ch);
if (tok2 == 0) {
ThrowXMLwithMemMgr(ParseException,XMLExcepts::Parser_Atom5, getMemoryManager());
}
tok->mergeRanges(tok2);
end = true;
}
break;
case chDash:
wasDecoded = true;
// fall thru to default.
default:
ch = decodeEscaped();
}
} // end if REGX_T_BACKSOLIDUS
else if (type == REGX_T_XMLSCHEMA_CC_SUBTRACTION && !firstLoop) {
if (isNRange)
{
tok = RangeToken::complementRanges(tok, fTokenFactory, fMemoryManager);
isNRange=false;
}
RangeToken* rangeTok = parseCharacterClass(false);
tok->subtractRanges(rangeTok);
if (getState() != REGX_T_CHAR || getCharData() != chCloseSquare) {
ThrowXMLwithMemMgr(ParseException,XMLExcepts::Parser_CC5, getMemoryManager());
}
break;
} // end if REGX_T_XMLSCHEMA...
processNext();
if (!end) {
if (type == REGX_T_CHAR
&& (ch == chOpenSquare
|| ch == chCloseSquare
|| (ch == chDash && getCharData() == chCloseSquare && firstLoop))) {
// if regex = [-] then invalid...
// '[', ']', '-' not allowed and should be escaped
XMLCh chStr[] = { ch, chNull };
ThrowXMLwithMemMgr2(ParseException,XMLExcepts::Parser_CC6, chStr, chStr, getMemoryManager());
}
if (ch == chDash && getCharData() == chDash && getState() != REGX_T_BACKSOLIDUS && !wasDecoded) {
XMLCh chStr[] = { ch, chNull };
ThrowXMLwithMemMgr2(ParseException,XMLExcepts::Parser_CC6, chStr, chStr, getMemoryManager());
}
if (getState() != REGX_T_CHAR || getCharData() != chDash) {
tok->addRange(ch, ch);
}
else {
processNext();
if ((type = getState()) == REGX_T_EOF)
ThrowXMLwithMemMgr(ParseException,XMLExcepts::Parser_CC2, getMemoryManager());
if (type == REGX_T_CHAR && getCharData() == chCloseSquare) {
tok->addRange(ch, ch);
tok->addRange(chDash, chDash);
}
else if (type == REGX_T_XMLSCHEMA_CC_SUBTRACTION) {
static const XMLCh dashStr[] = { chDash, chNull};
ThrowXMLwithMemMgr2(ParseException, XMLExcepts::Parser_CC6, dashStr, dashStr, getMemoryManager());
}
else {
XMLInt32 rangeEnd = getCharData();
XMLCh rangeEndStr[] = { rangeEnd, chNull };
if (type == REGX_T_CHAR) {
if (rangeEnd == chOpenSquare
|| rangeEnd == chCloseSquare
|| rangeEnd == chDash)
// '[', ']', '-' not allowed and should be escaped
ThrowXMLwithMemMgr2(ParseException, XMLExcepts::Parser_CC6, rangeEndStr, rangeEndStr, getMemoryManager());
}
else if (type == REGX_T_BACKSOLIDUS) {
rangeEnd = decodeEscaped();
}
processNext();
if (ch > rangeEnd) {
XMLCh chStr[] = { ch, chNull };
ThrowXMLwithMemMgr2(ParseException,XMLExcepts::Parser_Ope3, rangeEndStr, chStr, getMemoryManager());
}
tok->addRange(ch, rangeEnd);
}
}
}
firstLoop = false;
}
if (getState() == REGX_T_EOF)
ThrowXMLwithMemMgr(ParseException,XMLExcepts::Parser_CC2, getMemoryManager());
if (isNRange)
{
if(useNRange)
tok->setTokenType(Token::T_NRANGE);
else
tok = RangeToken::complementRanges(tok, fTokenFactory, fMemoryManager);
}
tok->sortRanges();
tok->compactRanges();
// If the case-insensitive option is enabled, we need to
// have the new RangeToken instance build its internal
// case-insensitive RangeToken.
if (RegularExpression::isSet(fOptions, RegularExpression::IGNORE_CASE))
{
tok->getCaseInsensitiveToken(fTokenFactory);
}
setParseContext(regexParserStateNormal);
processNext();
return tok;
}
void QNameDatatypeValidator::checkContent( const XMLCh* const content
, ValidationContext* const context
, bool asBase
, MemoryManager* const manager
)
{
//validate against base validator if any
QNameDatatypeValidator *pBaseValidator = (QNameDatatypeValidator*) this->getBaseValidator();
if (pBaseValidator)
pBaseValidator->checkContent(content, context, true, manager);
int thisFacetsDefined = getFacetsDefined();
// we check pattern first
if ( (thisFacetsDefined & DatatypeValidator::FACET_PATTERN ) != 0 )
{
if (getRegex()->matches(content, manager) ==false)
{
ThrowXMLwithMemMgr2(InvalidDatatypeValueException
, XMLExcepts::VALUE_NotMatch_Pattern
, content
, getPattern()
, manager);
}
}
// if this is a base validator, we only need to check pattern facet
// all other facet were inherited by the derived type
if (asBase)
return;
checkValueSpace(content, manager);
if (context) {
int colonPos = XMLString::indexOf(content, chColon);
if (colonPos > 0) {
XMLCh* prefix = XMLString::replicate(content, manager);
ArrayJanitor<XMLCh> jan(prefix, manager);
normalizeContent(prefix, manager);
prefix[colonPos] = chNull;
if (context->isPrefixUnknown(prefix)) {
ThrowXMLwithMemMgr1(InvalidDatatypeValueException
, XMLExcepts::VALUE_QName_Invalid2
, content
, manager);
}
}
}
if ((thisFacetsDefined & DatatypeValidator::FACET_ENUMERATION) != 0 &&
(getEnumeration() != 0))
{
XMLCh* normContent = XMLString::replicate(content, manager);
ArrayJanitor<XMLCh> jan(normContent, manager);
normalizeContent(normContent, manager);
int i=0;
int enumLength = getEnumeration()->size();
for ( ; i < enumLength; i++)
{
if (XMLString::equals(normContent, getEnumeration()->elementAt(i)))
break;
}
if (i == enumLength)
ThrowXMLwithMemMgr1(InvalidDatatypeValueException, XMLExcepts::VALUE_NotIn_Enumeration, content, manager);
}
checkAdditionalFacet(content, manager);
}