Newer
Older
srcPtr++;
// Check Validity Constraint for Standalone document declaration
// XML 1.0, Section 2.9
if (fStandalone && fValidate && isAttExternal)
{
if (!firstNonWS || (nextCh != chSpace) || (!*srcPtr) || fReaderMgr.getCurrentReader()->isWhitespace(*srcPtr))
Gareth Reakes
committed
// Can't have a standalone document declaration of "yes" if attribute
// values are subject to normalisation
fValidator->emitError(XMLValid::NoAttNormForStandalone, attName);
if(fGrammarType == Grammar::SchemaGrammarType && attDef) {
Gareth Reakes
committed
((SchemaAttDef *)(attDef))->setValidity(PSVIDefs::INVALID);
if (getPSVIHandler())
{
// REVISIT:
// PSVIAttribute->setValidity(PSVIItem::VALIDITY_INVALID);
}
Gareth Reakes
committed
}
}
}
continue;
}
firstNonWS = true;
}
}
// Add this char to the target buffer
toFill.append(nextCh);
// And move up to the next character in the source
srcPtr++;
}
Gareth Reakes
committed
if(fGrammarType == Grammar::SchemaGrammarType && attDef)
Gareth Reakes
committed
((SchemaElementDecl *)fElemStack.topElement()->fThisElement)->updateValidityFromAttribute((SchemaAttDef *)attDef);
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
return retVal;
}
// This method will just normalize the input value as CDATA without
// any standalone checking.
bool IGXMLScanner::normalizeAttRawValue( const XMLCh* const attrName
, const XMLCh* const value
, XMLBuffer& toFill)
{
// Assume its going to go fine, and empty the target buffer in preperation
bool retVal = true;
toFill.reset();
// Loop through the chars of the source value and normalize it according
// to the type.
bool escaped;
XMLCh nextCh;
const XMLCh* srcPtr = value;
while (*srcPtr)
{
// Get the next character from the source. We have to watch for
// escaped characters (which are indicated by a 0xFFFF value followed
// by the char that was escaped.)
nextCh = *srcPtr;
escaped = (nextCh == 0xFFFF);
if (escaped)
nextCh = *++srcPtr;
// If its not escaped, then make sure its not a < character, which is
// not allowed in attribute values.
if (!escaped && (*srcPtr == chOpenAngle))
{
emitError(XMLErrs::BracketInAttrValue, attrName);
retVal = false;
}
if (!escaped)
{
// NOTE: Yes this is a little redundant in that a 0x20 is
// replaced with an 0x20. But its faster to do this (I think)
// than checking for 9, A, and D separately.
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
nextCh = chSpace;
}
// Add this char to the target buffer
toFill.append(nextCh);
// And move up to the next character in the source
srcPtr++;
}
return retVal;
}
unsigned int
IGXMLScanner::resolvePrefix( const XMLCh* const prefix
, const ElemStack::MapModes mode)
{
// Watch for the special namespace prefixes. We always map these to
// special URIs. 'xml' gets mapped to the official URI that its defined
// to map to by the NS spec. xmlns gets mapped to a special place holder
// URI that we define (so that it maps to something checkable.)
if (XMLString::equals(prefix, XMLUni::fgXMLNSString))
return fXMLNSNamespaceId;
else if (XMLString::equals(prefix, XMLUni::fgXMLString))
return fXMLNamespaceId;
// Ask the element stack to search up itself for a mapping for the
// passed prefix.
bool unknown;
unsigned int uriId = fElemStack.mapPrefixToURI(prefix, mode, unknown);
// If it was unknown, then the URI was faked in but we have to issue an error
if (unknown)
emitError(XMLErrs::UnknownPrefix, prefix);
return uriId;
}
unsigned int
IGXMLScanner::resolvePrefix( const XMLCh* const prefix
, XMLBuffer& bufToFill
, const ElemStack::MapModes mode)
{
// Watch for the special namespace prefixes. We always map these to
// special URIs. 'xml' gets mapped to the official URI that its defined
// to map to by the NS spec. xmlns gets mapped to a special place holder
// URI that we define (so that it maps to something checkable.)
if (XMLString::equals(prefix, XMLUni::fgXMLNSString))
return fXMLNSNamespaceId;
else if (XMLString::equals(prefix, XMLUni::fgXMLString))
return fXMLNamespaceId;
// Ask the element stack to search up itself for a mapping for the
// passed prefix.
bool unknown;
unsigned int uriId = fElemStack.mapPrefixToURI(prefix, mode, unknown);
// If it was unknown, then the URI was faked in but we have to issue an error
if (unknown)
emitError(XMLErrs::UnknownPrefix, prefix);
getURIText(uriId,bufToFill);
return uriId;
}
// This method will reset the scanner data structures, and related plugged
// in stuff, for a new scan session. We get the input source for the primary
// XML entity, create the reader for it, and push it on the stack so that
// upon successful return from here we are ready to go.
void IGXMLScanner::scanReset(const InputSource& src)
{
// This call implicitly tells us that we are going to reuse the scanner
// if it was previously used. So tell the validator to reset itself.
//
// But, if the fUseCacheGrammar flag is set, then don't reset it.
//
// NOTE: The ReaderMgr is flushed on the way out, because that is
// required to insure that files are closed.
fGrammarResolver->cacheGrammarFromParse(fToCacheGrammar);
fGrammarResolver->useCachedGrammarInParse(fUseCachedGrammar);
fDTDGrammar = (DTDGrammar*) fGrammarResolver->getGrammar(XMLUni::fgDTDEntityString);
if (!fDTDGrammar) {
fDTDGrammar = new (fGrammarPoolMemoryManager) DTDGrammar(fGrammarPoolMemoryManager);
fGrammarResolver->putGrammar(fDTDGrammar);
}
else
fDTDGrammar->reset();
fGrammar = fDTDGrammar;
fGrammarType = fGrammar->getGrammarType();
fRootGrammar = 0;
if (fValidatorFromUser) {
if (fValidator->handlesDTD())
fValidator->setGrammar(fGrammar);
else if (fValidator->handlesSchema()) {
((SchemaValidator*) fValidator)->setErrorReporter(fErrorReporter);
((SchemaValidator*) fValidator)->setGrammarResolver(fGrammarResolver);
((SchemaValidator*) fValidator)->setExitOnFirstFatal(fExitOnFirstFatal);
}
}
else {
// set fValidator as fDTDValidator
fValidator = fDTDValidator;
fValidator->setGrammar(fGrammar);
}
// Reset validation
fValidate = (fValScheme == Val_Always) ? true : false;
// And for all installed handlers, send reset events. This gives them
// a chance to flush any cached data.
if (fDocHandler)
fDocHandler->resetDocument();
if (fEntityHandler)
fEntityHandler->resetEntities();
if (fErrorReporter)
fErrorReporter->resetErrors();
// Clear out the id reference list
fMemoryManager->deallocate(fRootElemName);//delete [] fRootElemName;
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
fRootElemName = 0;
// Reset IdentityConstraints
fValueStoreCache->startDocument();
fMatcherStack->clear();
// Reset the element stack, and give it the latest ids for the special
// URIs it has to know about.
fElemStack.reset
(
fEmptyNamespaceId
, fUnknownNamespaceId
, fXMLNamespaceId
, fXMLNSNamespaceId
);
if (!fSchemaNamespaceId)
fSchemaNamespaceId = fURIStringPool->addOrFind(SchemaSymbols::fgURI_XSI);
// Reset some status flags
fInException = false;
fStandalone = false;
fErrorCount = 0;
fHasNoDTD = true;
fSeeXsi = false;
// Reset the validators
fDTDValidator->reset();
fDTDValidator->setErrorReporter(fErrorReporter);
fSchemaValidator->reset();
fSchemaValidator->setErrorReporter(fErrorReporter);
fSchemaValidator->setExitOnFirstFatal(fExitOnFirstFatal);
fSchemaValidator->setGrammarResolver(fGrammarResolver);
if (fValidatorFromUser)
fValidator->reset();
// Handle the creation of the XML reader object for this input source.
// This will provide us with transcoding and basic lexing services.
XMLReader* newReader = fReaderMgr.createReader
(
src
, true
, XMLReader::RefFrom_NonLiteral
, XMLReader::Type_General
, XMLReader::Source_External
, fCalculateSrcOfs
);
if (!newReader) {
if (src.getIssueFatalErrorIfNotFound())
ThrowXML1(RuntimeException, XMLExcepts::Scan_CouldNotOpenSource, src.getSystemId());
else
ThrowXML1(RuntimeException, XMLExcepts::Scan_CouldNotOpenSource_Warning, src.getSystemId());
}
// Push this read onto the reader manager
fReaderMgr.pushReader(newReader, 0);
// and reset security-related things if necessary:
if(fSecurityManager != 0)
{
fEntityExpansionLimit = fSecurityManager->getEntityExpansionLimit();
fEntityExpansionCount = 0;
}
fElemCount = 0;
if(fUIntPoolRowTotal >= 32)
{ // 8 KB tied up with validating attributes...
fAttDefRegistry->removeAll();
fUndeclaredAttrRegistry->removeAll();
fUndeclaredAttrRegistryNS->removeAll();
recreateUIntPool();
}
else
{
// note that this will implicitly reset the values of the hashtables,
// though their buckets will still be tied up
resetUIntPool();
}
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
}
// This method is called between markup in content. It scans for character
// data that is sent to the document handler. It watches for any markup
// characters that would indicate that the character data has ended. It also
// handles expansion of general and character entities.
//
// sendData() is a local static helper for this method which handles some
// code that must be done in three different places here.
void IGXMLScanner::sendCharData(XMLBuffer& toSend)
{
// If no data in the buffer, then nothing to do
if (toSend.isEmpty())
return;
// We do different things according to whether we are validating or
// not. If not, its always just characters; else, it depends on the
// current element's content model.
if (fValidate)
{
// Get the raw data we need for the callback
XMLCh* rawBuf = toSend.getRawBuffer();
unsigned int len = toSend.getLen();
// And see if the current element is a 'Children' style content model
const ElemStack::StackElem* topElem = fElemStack.topElement();
// Get the character data opts for the current element
XMLElementDecl::CharDataOpts charOpts = XMLElementDecl::AllCharData;
if(fGrammar->getGrammarType() == Grammar::SchemaGrammarType)
{
// And see if the current element is a 'Children' style content model
ComplexTypeInfo *currType = ((SchemaValidator*)fValidator)->getCurrentTypeInfo();
if(currType)
{
SchemaElementDecl::ModelTypes modelType = (SchemaElementDecl::ModelTypes) currType->getContentType();
if(modelType == SchemaElementDecl::Children)
charOpts = XMLElementDecl::SpacesOk;
else if(modelType == SchemaElementDecl::Empty)
charOpts = XMLElementDecl::NoCharData;
}
} else // DTD grammar
charOpts = topElem->fThisElement->getCharDataOpts();
if (charOpts == XMLElementDecl::NoCharData)
{
// They definitely cannot handle any type of char data
fValidator->emitError(XMLValid::NoCharDataInCM);
if(fGrammarType == Grammar::SchemaGrammarType)
Gareth Reakes
committed
((SchemaElementDecl *)topElem->fThisElement)->setValidity(PSVIDefs::INVALID);
if (getPSVIHandler())
{
// REVISIT:
// PSVIElement->setValidity(PSVIItem::VALIDITY_INVALID);
}
}
{
// Its all spaces. So, if they can take spaces, then send it
// as ignorable whitespace. If they can handle any char data
// send it as characters.
if (charOpts == XMLElementDecl::SpacesOk) {
if (fDocHandler)
fDocHandler->ignorableWhitespace(rawBuf, len, false);
}
else if (charOpts == XMLElementDecl::AllCharData)
{
if (fGrammarType != Grammar::SchemaGrammarType)
{
if (fDocHandler)
fDocHandler->docCharacters(rawBuf, len, false);
}
else
{
SchemaValidator *schemaValidator = (SchemaValidator *)fValidator;
DatatypeValidator* tempDV = ((SchemaValidator*) fValidator)->getCurrentDatatypeValidator();
if (tempDV && tempDV->getWSFacet() != DatatypeValidator::PRESERVE)
{
// normalize the character according to schema whitespace facet
XMLBufBid bbtemp(&fBufMgr);
XMLBuffer& tempBuf = bbtemp.getBuffer();
((SchemaValidator*) fValidator)->normalizeWhiteSpace(tempDV, rawBuf, tempBuf);
rawBuf = tempBuf.getRawBuffer();
len = tempBuf.getLen();
}
}
// tell the schema validation about the character data for checkContent later
schemaValidator->setDatatypeBuffer(rawBuf);
if (fMatcherStack->getMatcherCount())
fContent.append(rawBuf, len);
fDocHandler->docCharacters(rawBuf, len, false);
}
}
}
else
{
// If they can take any char data, then send it. Otherwise, they
// can only handle whitespace and can't handle this stuff so
// issue an error.
if (charOpts == XMLElementDecl::AllCharData)
{
if (fGrammarType != Grammar::SchemaGrammarType)
{
if (fDocHandler)
fDocHandler->docCharacters(rawBuf, len, false);
}
else
{
SchemaValidator *schemaValidator = (SchemaValidator*)fValidator;
DatatypeValidator* tempDV = ((SchemaValidator*) fValidator)->getCurrentDatatypeValidator();
if (tempDV && tempDV->getWSFacet() != DatatypeValidator::PRESERVE)
{
// normalize the character according to schema whitespace facet
XMLBufBid bbtemp(&fBufMgr);
XMLBuffer& tempBuf = bbtemp.getBuffer();
((SchemaValidator*) fValidator)->normalizeWhiteSpace(tempDV, rawBuf, tempBuf);
rawBuf = tempBuf.getRawBuffer();
len = tempBuf.getLen();
}
}
// tell the schema validation about the character data for checkContent later
schemaValidator->setDatatypeBuffer(rawBuf);
if (fMatcherStack->getMatcherCount())
fContent.append(rawBuf, len);
fDocHandler->docCharacters(rawBuf, len, false);
}
}
else
{
fValidator->emitError(XMLValid::NoCharDataInCM);
if(fGrammarType == Grammar::SchemaGrammarType)
Gareth Reakes
committed
((SchemaElementDecl *)topElem->fThisElement)->setValidity(PSVIDefs::INVALID);
if (getPSVIHandler())
{
// REVISIT:
// PSVIAttribute->setValidity(PSVIItem::VALIDITY_INVALID);
}
}
}
}
}
else
{
// call all active identity constraints
if (fGrammarType == Grammar::SchemaGrammarType) {
if (fMatcherStack->getMatcherCount())
fContent.append(toSend.getRawBuffer(), toSend.getLen());
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
}
// Always assume its just char data if not validating
if (fDocHandler)
fDocHandler->docCharacters(toSend.getRawBuffer(), toSend.getLen(), false);
}
// Reset buffer
toSend.reset();
}
// This method is called with a key/value string pair that represents an
// xmlns="yyy" or xmlns:xxx="yyy" attribute. This method will update the
// current top of the element stack based on this data. We know that when
// we get here, that it is one of these forms, so we don't bother confirming
// it.
//
// But we have to ensure
// 1. xxx is not xmlns
// 2. if xxx is xml, then yyy must match XMLUni::fgXMLURIName, and vice versa
// 3. yyy is not XMLUni::fgXMLNSURIName
// 4. if xxx is not null, then yyy cannot be an empty string.
void IGXMLScanner::updateNSMap(const XMLCh* const attrName
, const XMLCh* const attrValue)
{
// We need a buffer to normalize the attribute value into
XMLBufBid bbNormal(&fBufMgr);
XMLBuffer& normalBuf = bbNormal.getBuffer();
// Normalize the value into the passed buffer. In this case, we don't
// care about the return value. An error was issued for the error, which
// is all we care about here.
normalizeAttRawValue(attrName, attrValue, normalBuf);
XMLCh* namespaceURI = normalBuf.getRawBuffer();
// We either have the default prefix (""), or we point it into the attr
// name parameter. Note that the xmlns is not the prefix we care about
// here. To us, the 'prefix' is really the local part of the attrName
// parameter.
//
// Check 1. xxx is not xmlns
// 2. if xxx is xml, then yyy must match XMLUni::fgXMLURIName, and vice versa
// 3. yyy is not XMLUni::fgXMLNSURIName
// 4. if xxx is not null, then yyy cannot be an empty string.
const XMLCh* prefPtr = XMLUni::fgZeroLenString;
const int colonOfs = XMLString::indexOf(attrName, chColon);
if (colonOfs != -1) {
prefPtr = &attrName[colonOfs + 1];
if (XMLString::equals(prefPtr, XMLUni::fgXMLNSString))
emitError(XMLErrs::NoUseOfxmlnsAsPrefix);
else if (XMLString::equals(prefPtr, XMLUni::fgXMLString)) {
if (!XMLString::equals(namespaceURI, XMLUni::fgXMLURIName))
emitError(XMLErrs::PrefixXMLNotMatchXMLURI);
}
Neil Graham
committed
if (!namespaceURI)
emitError(XMLErrs::NoEmptyStrNamespace, attrName);
else if(!*namespaceURI && fXMLVersion == XMLReader::XMLV1_0)
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
emitError(XMLErrs::NoEmptyStrNamespace, attrName);
}
if (XMLString::equals(namespaceURI, XMLUni::fgXMLNSURIName))
emitError(XMLErrs::NoUseOfxmlnsURI);
else if (XMLString::equals(namespaceURI, XMLUni::fgXMLURIName)) {
if (!XMLString::equals(prefPtr, XMLUni::fgXMLString))
emitError(XMLErrs::XMLURINotMatchXMLPrefix);
}
// Ok, we have to get the unique id for the attribute value, which is the
// URI that this value should be mapped to. The validator has the
// namespace string pool, so we ask him to find or add this new one. Then
// we ask the element stack to add this prefix to URI Id mapping.
fElemStack.addPrefix
(
prefPtr
, fURIStringPool->addOrFind(namespaceURI)
);
}
void IGXMLScanner::scanRawAttrListforNameSpaces(const RefVectorOf<KVStringPair>* theRawAttrList, int attCount)
{
// Make an initial pass through the list and find any xmlns attributes or
// schema attributes.
// When we find one, send it off to be used to update the element stack's
// namespace mappings.
int index = 0;
for (index = 0; index < attCount; index++)
{
// each attribute has the prefix:suffix="value"
const KVStringPair* curPair = fRawAttrList->elementAt(index);
const XMLCh* rawPtr = curPair->getKey();
// If either the key begins with "xmlns:" or its just plain
// "xmlns", then use it to update the map.
if (!XMLString::compareNString(rawPtr, XMLUni::fgXMLNSColonString, 6)
|| XMLString::equals(rawPtr, XMLUni::fgXMLNSString))
{
const XMLCh* valuePtr = curPair->getValue();
updateNSMap(rawPtr, valuePtr);
// if the schema URI is seen in the the valuePtr, set the boolean seeXsi
if (XMLString::equals(valuePtr, SchemaSymbols::fgURI_XSI)) {
fSeeXsi = true;
}
}
}
// walk through the list again to deal with "xsi:...."
if (fDoSchema && fSeeXsi)
{
// Schema Xsi Type yyyy (e.g. xsi:type="yyyyy")
XMLBufBid bbXsi(&fBufMgr);
XMLBuffer& fXsiType = bbXsi.getBuffer();
for (index = 0; index < attCount; index++)
{
// each attribute has the prefix:suffix="value"
const KVStringPair* curPair = fRawAttrList->elementAt(index);
const XMLCh* rawPtr = curPair->getKey();
const XMLCh* prefPtr = XMLUni::fgZeroLenString;
int colonInd = XMLString::indexOf(rawPtr, chColon);
if (colonInd != -1) {
fURIBuf.set(rawPtr, colonInd);
prefPtr = fURIBuf.getRawBuffer();
}
// if schema URI has been seen, scan for the schema location and uri
// and resolve the schema grammar; or scan for schema type
if (resolvePrefix(prefPtr, ElemStack::Mode_Attribute) == fSchemaNamespaceId) {
const XMLCh* valuePtr = curPair->getValue();
const XMLCh* suffPtr = &rawPtr[colonInd + 1];
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
if (XMLString::equals(suffPtr, SchemaSymbols::fgXSI_SCHEMALOCACTION))
parseSchemaLocation(valuePtr);
else if (XMLString::equals(suffPtr, SchemaSymbols::fgXSI_NONAMESPACESCHEMALOCACTION))
resolveSchemaGrammar(valuePtr, XMLUni::fgZeroLenString);
if (XMLString::equals(suffPtr, SchemaSymbols::fgXSI_TYPE)) {
fXsiType.set(valuePtr);
}
else if (XMLString::equals(suffPtr, SchemaSymbols::fgATT_NILL)
&& fValidator && fValidator->handlesSchema()
&& XMLString::equals(valuePtr, SchemaSymbols::fgATTVAL_TRUE)) {
((SchemaValidator*)fValidator)->setNillable(true);
}
}
}
if (fValidator && fValidator->handlesSchema()) {
if (!fXsiType.isEmpty()) {
int colonPos = -1;
unsigned int uriId = resolveQName (
fXsiType.getRawBuffer()
, fPrefixBuf
, ElemStack::Mode_Element
, colonPos
);
((SchemaValidator*)fValidator)->setXsiType(fPrefixBuf.getRawBuffer(), fXsiType.getRawBuffer() + colonPos + 1, uriId);
}
}
}
}
void IGXMLScanner::parseSchemaLocation(const XMLCh* const schemaLocationStr)
{
XMLCh* locStr = XMLString::replicate(schemaLocationStr, fMemoryManager);
ArrayJanitor<XMLCh> janLoc(locStr, fMemoryManager);
processSchemaLocation(locStr);
unsigned int size = fLocationPairs->size();
if (size % 2 != 0 ) {
emitError(XMLErrs::BadSchemaLocation);
} else {
for(unsigned int i=0; i<size; i=i+2) {
resolveSchemaGrammar(fLocationPairs->elementAt(i+1), fLocationPairs->elementAt(i));
}
}
}
void IGXMLScanner::resolveSchemaGrammar(const XMLCh* const loc, const XMLCh* const uri) {
Grammar* grammar = fGrammarResolver->getGrammar(uri);
if (!grammar || grammar->getGrammarType() == Grammar::DTDGrammarType) {
XSDDOMParser parser(0, fMemoryManager, 0);
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
parser.setValidationScheme(XercesDOMParser::Val_Never);
parser.setDoNamespaces(true);
parser.setUserEntityHandler(fEntityHandler);
parser.setUserErrorReporter(fErrorReporter);
// Create a buffer for expanding the system id
XMLBufBid bbSys(&fBufMgr);
XMLBuffer& expSysId = bbSys.getBuffer();
XMLBuffer& normalizedSysId = bbSys.getBuffer();
normalizeURI(loc, normalizedSysId);
// Allow the entity handler to expand the system id if they choose
// to do so.
InputSource* srcToFill = 0;
const XMLCh* normalizedURI = normalizedSysId.getRawBuffer();
if (fEntityHandler)
{
if (!fEntityHandler->expandSystemId(normalizedURI, expSysId))
expSysId.set(normalizedURI);
ReaderMgr::LastExtEntityInfo lastInfo;
fReaderMgr.getLastExtEntityInfo(lastInfo);
XMLResourceIdentifier resourceIdentifier(XMLResourceIdentifier::SchemaGrammar,
expSysId.getRawBuffer(), uri, XMLUni::fgZeroLenString, lastInfo.systemId);
srcToFill = fEntityHandler->resolveEntity(&resourceIdentifier);
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
}
else
{
expSysId.set(normalizedURI);
}
// If they didn't create a source via the entity handler, then we
// have to create one on our own.
if (!srcToFill)
{
ReaderMgr::LastExtEntityInfo lastInfo;
fReaderMgr.getLastExtEntityInfo(lastInfo);
try
{
XMLURL urlTmp(lastInfo.systemId, expSysId.getRawBuffer());
if (urlTmp.isRelative())
{
ThrowXML
(
MalformedURLException
, XMLExcepts::URL_NoProtocolPresent
);
}
else {
if (fStandardUriConformant && urlTmp.hasInvalidChar())
ThrowXML(MalformedURLException, XMLExcepts::URL_MalformedURL);
Khaled Noaman
committed
srcToFill = new (fMemoryManager) URLInputSource(urlTmp, fMemoryManager);
}
catch(const MalformedURLException& e)
// Its not a URL, so lets assume its a local file name if non-standard uri is allowed
if (!fStandardUriConformant)
srcToFill = new (fMemoryManager) LocalFileInputSource
(
lastInfo.systemId
, expSysId.getRawBuffer()
Khaled Noaman
committed
, fMemoryManager
);
else
throw e;
}
}
// Put a janitor on the input source
Janitor<InputSource> janSrc(srcToFill);
// Should just issue warning if the schema is not found
bool flag = srcToFill->getIssueFatalErrorIfNotFound();
srcToFill->setIssueFatalErrorIfNotFound(false);
parser.parse(*srcToFill);
// Reset the InputSource
srcToFill->setIssueFatalErrorIfNotFound(flag);
if (parser.getSawFatal() && fExitOnFirstFatal)
emitError(XMLErrs::SchemaScanFatalError);
DOMDocument* document = parser.getDocument(); //Our Grammar
if (document != 0) {
DOMElement* root = document->getDocumentElement();// This is what we pass to TraverserSchema
if (root != 0)
{
const XMLCh* newUri = root->getAttribute(SchemaSymbols::fgATT_TARGETNAMESPACE);
if (!XMLString::equals(newUri, uri)) {
if (fValidate || fValScheme == Val_Auto) {
fValidator->emitError(XMLValid::WrongTargetNamespace, loc, uri);
Gareth Reakes
committed
}
grammar = fGrammarResolver->getGrammar(newUri);
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
}
if (!grammar || grammar->getGrammarType() == Grammar::DTDGrammarType) {
// Since we have seen a grammar, set our validation flag
// at this point if the validation scheme is auto
if (fValScheme == Val_Auto && !fValidate) {
fValidate = true;
fElemStack.setValidationFlag(fValidate);
}
// we have seen a schema, so set up the fValidator as fSchemaValidator
if (!fValidator->handlesSchema())
{
if (fValidatorFromUser) {
// the fValidator is from user
ThrowXML(RuntimeException, XMLExcepts::Gen_NoSchemaValidator);
}
else {
fValidator = fSchemaValidator;
}
}
grammar = new (fGrammarPoolMemoryManager) SchemaGrammar(fGrammarPoolMemoryManager);
XMLSchemaDescription* gramDesc = (XMLSchemaDescription*) grammar->getGrammarDescription();
gramDesc->setContextType(XMLSchemaDescription::CONTEXT_PREPARSE);
gramDesc->setLocationHints(srcToFill->getSystemId());
TraverseSchema traverseSchema
(
root
, fURIStringPool
, (SchemaGrammar*) grammar
, fGrammarResolver
, this
, srcToFill->getSystemId()
, fEntityHandler
, fErrorReporter
, fMemoryManager
);
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
if (fGrammarType == Grammar::DTDGrammarType) {
fGrammar = grammar;
fGrammarType = Grammar::SchemaGrammarType;
fValidator->setGrammar(fGrammar);
}
if (fValidate) {
// validate the Schema scan so far
fValidator->preContentValidation(false);
}
}
}
}
}
else {
// Since we have seen a grammar, set our validation flag
// at this point if the validation scheme is auto
if (fValScheme == Val_Auto && !fValidate) {
fValidate = true;
fElemStack.setValidationFlag(fValidate);
}
// we have seen a schema, so set up the fValidator as fSchemaValidator
if (!fValidator->handlesSchema())
{
if (fValidatorFromUser) {
// the fValidator is from user
ThrowXML(RuntimeException, XMLExcepts::Gen_NoSchemaValidator);
}
else {
fValidator = fSchemaValidator;
}
}
if (fGrammarType == Grammar::DTDGrammarType) {
fGrammar = grammar;
fGrammarType = Grammar::SchemaGrammarType;
fValidator->setGrammar(fGrammar);
}
}
}
InputSource* IGXMLScanner::resolveSystemId(const XMLCh* const sysId)
{
// Create a buffer for expanding the system id
XMLBufBid bbSys(&fBufMgr);
XMLBuffer& expSysId = bbSys.getBuffer();
// Allow the entity handler to expand the system id if they choose
// to do so.
InputSource* srcToFill = 0;
if (fEntityHandler)
{
if (!fEntityHandler->expandSystemId(sysId, expSysId))
expSysId.set(sysId);
ReaderMgr::LastExtEntityInfo lastInfo;
fReaderMgr.getLastExtEntityInfo(lastInfo);
XMLResourceIdentifier resourceIdentifier(XMLResourceIdentifier::ExternalEntity,
expSysId.getRawBuffer(), 0, XMLUni::fgZeroLenString, lastInfo.systemId);
srcToFill = fEntityHandler->resolveEntity(&resourceIdentifier);
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
}
else
{
expSysId.set(sysId);
}
// If they didn't create a source via the entity handler, then we
// have to create one on our own.
if (!srcToFill)
{
ReaderMgr::LastExtEntityInfo lastInfo;
fReaderMgr.getLastExtEntityInfo(lastInfo);
try
{
XMLURL urlTmp(lastInfo.systemId, expSysId.getRawBuffer());
if (urlTmp.isRelative())
{
ThrowXML
(
MalformedURLException
, XMLExcepts::URL_NoProtocolPresent
);
}
else {
if (fStandardUriConformant && urlTmp.hasInvalidChar())
ThrowXML(MalformedURLException, XMLExcepts::URL_MalformedURL);
Khaled Noaman
committed
srcToFill = new (fMemoryManager) URLInputSource(urlTmp, fMemoryManager);
}
catch(const MalformedURLException& e)
// Its not a URL, so lets assume its a local file name if non-standard uri is allowed
if (!fStandardUriConformant)
srcToFill = new (fMemoryManager) LocalFileInputSource
(
lastInfo.systemId
, expSysId.getRawBuffer()
Khaled Noaman
committed
, fMemoryManager
);
else
throw e;
}
}
return srcToFill;
}
// ---------------------------------------------------------------------------
// IGXMLScanner: Private grammar preparsing methods
// ---------------------------------------------------------------------------
Grammar* IGXMLScanner::loadXMLSchemaGrammar(const InputSource& src,
const bool toCache)
{
// Reset the validators
fSchemaValidator->reset();
fSchemaValidator->setErrorReporter(fErrorReporter);
fSchemaValidator->setExitOnFirstFatal(fExitOnFirstFatal);
fSchemaValidator->setGrammarResolver(fGrammarResolver);
if (fValidatorFromUser)
fValidator->reset();
if (!fValidator->handlesSchema()) {
if (fValidatorFromUser && fValidate)
ThrowXML(RuntimeException, XMLExcepts::Gen_NoSchemaValidator);
else {
fValidator = fSchemaValidator;
}
}
XSDDOMParser parser(0, fMemoryManager, 0);
parser.setValidationScheme(XercesDOMParser::Val_Never);
parser.setDoNamespaces(true);
parser.setUserEntityHandler(fEntityHandler);
parser.setUserErrorReporter(fErrorReporter);
// Should just issue warning if the schema is not found
bool flag = src.getIssueFatalErrorIfNotFound();
((InputSource&) src).setIssueFatalErrorIfNotFound(false);
parser.parse(src);
// Reset the InputSource
((InputSource&) src).setIssueFatalErrorIfNotFound(flag);
if (parser.getSawFatal() && fExitOnFirstFatal)
emitError(XMLErrs::SchemaScanFatalError);
DOMDocument* document = parser.getDocument(); //Our Grammar
if (document != 0) {
DOMElement* root = document->getDocumentElement();// This is what we pass to TraverserSchema
if (root != 0)
{
SchemaGrammar* grammar = new (fGrammarPoolMemoryManager) SchemaGrammar(fGrammarPoolMemoryManager);
XMLSchemaDescription* gramDesc = (XMLSchemaDescription*) grammar->getGrammarDescription();
gramDesc->setContextType(XMLSchemaDescription::CONTEXT_PREPARSE);
gramDesc->setLocationHints(src.getSystemId());
TraverseSchema traverseSchema
(
root
, fURIStringPool
, (SchemaGrammar*) grammar
, fGrammarResolver
, this
, src.getSystemId()
, fEntityHandler
, fErrorReporter
, fMemoryManager
);
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
if (fValidate) {
// validate the Schema scan so far
fValidator->setGrammar(grammar);
fValidator->preContentValidation(false, true);
}
if (toCache) {
fGrammarResolver->cacheGrammars();
}
return grammar;
}
}
return 0;
}
// ---------------------------------------------------------------------------
// IGXMLScanner: Private parsing methods
// ---------------------------------------------------------------------------
// This method is called to do a raw scan of an attribute value. It does not
// do normalization (since we don't know their types yet.) It just scans the
// value and does entity expansion.
//
// End of entity's must be dealt with here. During DTD scan, they can come
// from external entities. During content, they can come from any entity.
// We just eat the end of entity and continue with our scan until we come
// to the closing quote. If an unterminated value causes us to go through