Newer
Older
// This method is called from scanStartTag() to build up the list of
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
// XMLAttr objects that will be passed out in the start tag callout. We
// get the key/value pairs from the raw scan of explicitly provided attrs,
// which have not been normalized. And we get the element declaration from
// which we will get any defaulted or fixed attribute defs and add those
// in as well.
unsigned int
SGXMLScanner::buildAttList(const RefVectorOf<KVStringPair>& providedAttrs
, const unsigned int attCount
, XMLElementDecl* elemDecl
, RefVectorOf<XMLAttr>& toFill)
{
// Ask the element to clear the 'provided' flag on all of the att defs
// that it owns, and to return us a boolean indicating whether it has
// any defs.
const bool hasDefs = elemDecl->resetDefs();
// If there are no expliclitily provided attributes and there are no
// defined attributes for the element, the we don't have anything to do.
// So just return zero in this case.
if (!hasDefs && !attCount)
return 0;
// Keep up with how many attrs we end up with total
unsigned int retCount = 0;
// And get the current size of the output vector. This lets us use
// existing elements until we fill it, then start adding new ones.
const unsigned int curAttListSize = toFill.size();
// We need a buffer into which raw scanned attribute values will be
// normalized.
XMLBufBid bbNormal(&fBufMgr);
XMLBuffer& normBuf = bbNormal.getBuffer();
// Loop through our explicitly provided attributes, which are in the raw
// scanned form, and build up XMLAttr objects.
unsigned int index;
for (index = 0; index < attCount; index++)
{
const KVStringPair* curPair = providedAttrs.elementAt(index);
// We have to split the name into its prefix and name parts. Then
// we map the prefix to its URI.
const XMLCh* const namePtr = curPair->getKey();
ArrayJanitor<XMLCh> janName(0);
// use a stack-based buffer when possible.
XMLCh tempBuffer[100];
const int colonInd = XMLString::indexOf(namePtr, chColon);
const XMLCh* prefPtr = XMLUni::fgZeroLenString;
const XMLCh* suffPtr = XMLUni::fgZeroLenString;
if (colonInd != -1)
{
// We have to split the string, so make a copy.
if (XMLString::stringLen(namePtr) < sizeof(tempBuffer) / sizeof(tempBuffer[0]))
{
XMLString::copyString(tempBuffer, namePtr);
tempBuffer[colonInd] = chNull;
prefPtr = tempBuffer;
}
else
{
janName.reset(XMLString::replicate(namePtr, fMemoryManager), fMemoryManager);
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
janName[colonInd] = chNull;
prefPtr = janName.get();
}
suffPtr = prefPtr + colonInd + 1;
}
else
{
// No colon, so we just have a name with no prefix
suffPtr = namePtr;
}
// Map the prefix to a URI id. We tell him that we are mapping an
// attr prefix, so any xmlns attrs at this level will not affect it.
const unsigned int uriId = resolvePrefix(prefPtr, ElemStack::Mode_Attribute);
// If the uri comes back as the xmlns or xml URI or its just a name
// and that name is 'xmlns', then we handle it specially. So set a
// boolean flag that lets us quickly below know which we are dealing
// with.
const bool isNSAttr = (uriId == fXMLNSNamespaceId)
|| (uriId == fXMLNamespaceId)
|| XMLString::equals(suffPtr, XMLUni::fgXMLNSString)
|| XMLString::equals(getURIText(uriId), SchemaSymbols::fgURI_XSI);
// If its not a special case namespace attr of some sort, then we
// do normal checking and processing.
XMLAttDef::AttTypes attType;
if (!isNSAttr)
{
// Some checking for attribute wild card first (for schema)
bool laxThisOne = false;
bool skipThisOne = false;
XMLAttDef* attDefForWildCard = 0;
XMLAttDef* attDef = 0;
if (fGrammarType == Grammar::SchemaGrammarType) {
//retrieve the att def
attDef = ((SchemaElementDecl*)elemDecl)->getAttDef(suffPtr, uriId);
// if not found or faulted in - check for a matching wildcard attribute
// if no matching wildcard attribute, check (un)qualifed cases and flag
// appropriate errors
if (!attDef || (attDef->getCreateReason() == XMLAttDef::JustFaultIn)) {
SchemaAttDef* attWildCard = ((SchemaElementDecl*)elemDecl)->getAttWildCard();
if (attWildCard) {
//if schema, see if we should lax or skip the validation of this attribute
if (anyAttributeValidation(attWildCard, uriId, skipThisOne, laxThisOne)) {
SchemaGrammar* sGrammar = (SchemaGrammar*) fGrammarResolver->getGrammar(getURIText(uriId));
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
if (sGrammar && sGrammar->getGrammarType() == Grammar::SchemaGrammarType) {
RefHashTableOf<XMLAttDef>* attRegistry = sGrammar->getAttributeDeclRegistry();
if (attRegistry) {
attDefForWildCard = attRegistry->get(suffPtr);
}
}
}
}
else {
// not found, see if the attDef should be qualified or not
if (uriId == fEmptyNamespaceId) {
attDef = ((SchemaElementDecl*)elemDecl)->getAttDef(suffPtr, fURIStringPool->getId(fGrammar->getTargetNamespace()));
if (fValidate
&& attDef
&& attDef->getCreateReason() != XMLAttDef::JustFaultIn) {
// the attribute should be qualified
fValidator->emitError
(
XMLValid::AttributeNotQualified
, attDef->getFullName()
);
Gareth Reakes
committed
((SchemaAttDef *)(attDef))->setValidity(PSVIDefs::INVALID);
}
}
else {
attDef = ((SchemaElementDecl*)elemDecl)->getAttDef(suffPtr, fEmptyNamespaceId);
if (fValidate
&& attDef
&& attDef->getCreateReason() != XMLAttDef::JustFaultIn) {
// the attribute should be qualified
fValidator->emitError
(
XMLValid::AttributeNotUnQualified
, attDef->getFullName()
);
Gareth Reakes
committed
((SchemaAttDef *)(attDef))->setValidity(PSVIDefs::INVALID);
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
}
}
}
}
}
// Find this attribute within the parent element. We pass both
// the uriID/name and the raw QName buffer, since we don't know
// how the derived validator and its elements store attributes.
bool wasAdded = false;
if (!attDef) {
attDef = elemDecl->findAttr
(
curPair->getKey()
, uriId
, suffPtr
, prefPtr
, XMLElementDecl::AddIfNotFound
, wasAdded
);
}
Gareth Reakes
committed
if(!skipThisOne && fGrammarType == Grammar::SchemaGrammarType) {
//we may have set it to invalid already, but this is the first time we are guarenteed to have the attDef
if(((SchemaAttDef *)(attDef))->getValidity() != PSVIDefs::INVALID)
((SchemaAttDef *)(attDef))->setValidity(PSVIDefs::VALID);
Gareth Reakes
committed
((SchemaAttDef *)(attDef))->setValidationAttempted(PSVIDefs::FULL);
}
if (wasAdded)
{
// This is to tell the Validator that this attribute was
// faulted-in, was not an attribute in the attdef originally
attDef->setCreateReason(XMLAttDef::JustFaultIn);
}
bool errorCondition = fValidate && !attDefForWildCard &&
Gareth Reakes
committed
attDef->getCreateReason() == XMLAttDef::JustFaultIn && !attDef->getProvided();
if (errorCondition && !skipThisOne && !laxThisOne)
{
//
// Its not valid for this element, so issue an error if we are
// validating.
//
XMLBufBid bbMsg(&fBufMgr);
XMLBuffer& bufMsg = bbMsg.getBuffer();
if (uriId != fEmptyNamespaceId) {
XMLBufBid bbURI(&fBufMgr);
XMLBuffer& bufURI = bbURI.getBuffer();
getURIText(uriId, bufURI);
bufMsg.append(chOpenCurly);
bufMsg.append(bufURI.getRawBuffer());
bufMsg.append(chCloseCurly);
}
bufMsg.append(suffPtr);
fValidator->emitError
(
XMLValid::AttNotDefinedForElement
, bufMsg.getRawBuffer()
, elemDecl->getFullName()
);
Gareth Reakes
committed
((SchemaAttDef *)attDef)->setValidity(PSVIDefs::INVALID);
}
else if(errorCondition && laxThisOne) {
((SchemaAttDef *)(attDef))->setValidationAttempted(PSVIDefs::NONE);
((SchemaAttDef *)(attDef))->setValidity(PSVIDefs::UNKNOWN);
Gareth Reakes
committed
// If its already provided, then there are more than one of
// this attribute in this start tag, so emit an error.
if (attDef->getProvided())
{
emitError
(
XMLErrs::AttrAlreadyUsedInSTag
, attDef->getFullName()
, elemDecl->getFullName()
);
Gareth Reakes
committed
((SchemaAttDef *)(attDef))->setValidity(PSVIDefs::INVALID);
}
else
{
attDef->setProvided(true);
}
// Now normalize the raw value since we have the attribute type. We
// don't care about the return status here. If it failed, an error
// was issued, which is all we care about.
if (attDefForWildCard) {
Gareth Reakes
committed
((SchemaAttDef*)attDef)->setAnyDatatypeValidator(((SchemaAttDef*) attDefForWildCard)->getDatatypeValidator());
normalizeAttValue
(
attDefForWildCard
, curPair->getValue()
, normBuf
);
// If we found an attdef for this one, then lets validate it.
if (fNormalizeData)
{
DatatypeValidator* tempDV = ((SchemaAttDef*) attDefForWildCard)->getDatatypeValidator();
if (tempDV && tempDV->getWSFacet() != DatatypeValidator::PRESERVE)
{
// normalize the attribute according to schema whitespace facet
XMLBufBid bbtemp(&fBufMgr);
XMLBuffer& tempBuf = bbtemp.getBuffer();
((SchemaValidator*) fValidator)->normalizeWhiteSpace(tempDV, normBuf.getRawBuffer(), tempBuf);
normBuf.set(tempBuf.getRawBuffer());
}
}
if (fValidate && !skipThisOne) {
fValidator->validateAttrValue
(
attDefForWildCard
, normBuf.getRawBuffer()
, false
, elemDecl
);
}
// Save the type for later use
attType = attDefForWildCard->getType();
Gareth Reakes
committed
((SchemaElementDecl *)(elemDecl))->updateValidityFromAttribute((SchemaAttDef *)attDef);
DatatypeValidator* tempDV = ((SchemaAttDef*) attDefForWildCard)->getDatatypeValidator();
if(tempDV && tempDV->getType() == DatatypeValidator::Union)
((SchemaAttDef*)attDef)->setMembertypeValidator(((UnionDatatypeValidator *)tempDV)->getMemberTypeValidator());
}
else {
normalizeAttValue
(
attDef
, curPair->getValue()
, normBuf
);
// If we found an attdef for this one, then lets validate it.
if (attDef->getCreateReason() != XMLAttDef::JustFaultIn)
{
if (fNormalizeData && (fGrammarType == Grammar::SchemaGrammarType))
{
DatatypeValidator* tempDV = ((SchemaAttDef*) attDef)->getDatatypeValidator();
if (tempDV && tempDV->getWSFacet() != DatatypeValidator::PRESERVE)
{
// normalize the attribute according to schema whitespace facet
XMLBufBid bbtemp(&fBufMgr);
XMLBuffer& tempBuf = bbtemp.getBuffer();
((SchemaValidator*) fValidator)->normalizeWhiteSpace(tempDV, normBuf.getRawBuffer(), tempBuf);
normBuf.set(tempBuf.getRawBuffer());
}
}
if (fValidate && !skipThisOne)
{
fValidator->validateAttrValue
(
attDef
, normBuf.getRawBuffer()
, false
, elemDecl
);
}
}
// Save the type for later use
attType = attDef->getType();
Gareth Reakes
committed
((SchemaElementDecl *)(elemDecl))->updateValidityFromAttribute((SchemaAttDef *)attDef);
}
}
else
{
// Just normalize as CDATA
attType = XMLAttDef::CData;
normalizeAttRawValue
(
curPair->getKey()
, curPair->getValue()
, normBuf
);
}
Gareth Reakes
committed
// Add this attribute to the attribute list that we use to pass them
// to the handler. We reuse its existing elements but expand it as
// required.
XMLAttr* curAttr;
if (retCount >= curAttListSize)
{
curAttr = new (fMemoryManager) XMLAttr
(
uriId
, suffPtr
, prefPtr
, normBuf.getRawBuffer()
, attType
, true
, fMemoryManager
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
);
toFill.addElement(curAttr);
}
else
{
curAttr = toFill.elementAt(retCount);
curAttr->set
(
uriId
, suffPtr
, prefPtr
, normBuf.getRawBuffer()
, attType
);
curAttr->setSpecified(true);
}
// Bump the count of attrs in the list
retCount++;
}
// Now, if there are any attributes declared by this element, let's
// go through them and make sure that any required ones are provided,
// and fault in any fixed ones and defaulted ones that are not provided
// literally.
if (hasDefs)
{
// Check after all specified attrs are scanned
// (1) report error for REQUIRED attrs that are missing (V_TAGc)
// (2) add default attrs if missing (FIXED and NOT_FIXED)
XMLAttDefList& attDefList = elemDecl->getAttDefList();
while (attDefList.hasMoreElements())
{
// Get the current att def, for convenience and its def type
Gareth Reakes
committed
XMLAttDef *curDef = &attDefList.nextElement();
const XMLAttDef::DefAttTypes defType = curDef->getDefaultType();
Gareth Reakes
committed
if (!curDef->getProvided())
Gareth Reakes
committed
((SchemaAttDef *)curDef)->setValidationAttempted(PSVIDefs::FULL);
((SchemaAttDef *)curDef)->setValidity(PSVIDefs::VALID);
//the attributes is not provided
if (fValidate)
{
// If we are validating and its required, then an error
if ((defType == XMLAttDef::Required) ||
(defType == XMLAttDef::Required_And_Fixed) )
{
fValidator->emitError
(
XMLValid::RequiredAttrNotProvided
Gareth Reakes
committed
, curDef->getFullName()
Gareth Reakes
committed
((SchemaAttDef *)(curDef))->setValidity(PSVIDefs::INVALID);
}
else if ((defType == XMLAttDef::Default) ||
(defType == XMLAttDef::Fixed) )
{
Gareth Reakes
committed
if (fStandalone && curDef->isExternal())
{
// XML 1.0 Section 2.9
// Document is standalone, so attributes must not be defaulted.
Gareth Reakes
committed
fValidator->emitError(XMLValid::NoDefAttForStandalone, curDef->getFullName(), elemDecl->getFullName());
((SchemaAttDef *)(curDef))->setValidity(PSVIDefs::INVALID);
}
}
}
// Fault in the value if needed, and bump the att count.
// We have to
if ((defType == XMLAttDef::Default)
|| (defType == XMLAttDef::Fixed))
{
// Let the validator pass judgement on the attribute value
if (fValidate)
{
fValidator->validateAttrValue
(
Gareth Reakes
committed
curDef
, curDef->getValue()
, false
, elemDecl
);
}
XMLAttr* curAtt;
if (retCount >= curAttListSize)
{
curAtt = new (fMemoryManager) XMLAttr(fMemoryManager);
Gareth Reakes
committed
fValidator->faultInAttr(*curAtt, *curDef);
fAttrList->addElement(curAtt);
}
else
{
curAtt = fAttrList->elementAt(retCount);
Gareth Reakes
committed
fValidator->faultInAttr(*curAtt, *curDef);
}
// Indicate it was not explicitly specified and bump count
curAtt->setSpecified(false);
retCount++;
}
Gareth Reakes
committed
((SchemaElementDecl *)elemDecl)->updateValidityFromAttribute((SchemaAttDef *)curDef);
}
else
{
//attribute is provided
// (schema) report error for PROHIBITED attrs that are present (V_TAGc)
Gareth Reakes
committed
if (defType == XMLAttDef::Prohibited && fValidate) {
fValidator->emitError
(
XMLValid::ProhibitedAttributePresent
Gareth Reakes
committed
, curDef->getFullName()
((SchemaAttDef *)curDef)->setValidity(PSVIDefs::INVALID);
Gareth Reakes
committed
}
((SchemaElementDecl *)elemDecl)->updateValidityFromAttribute((SchemaAttDef *)curDef);
Gareth Reakes
committed
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
return retCount;
}
// This method will take a raw attribute value and normalize it according to
// the rules of the attribute type. It will put the resulting value into the
// passed buffer.
//
// This code assumes that escaped characters in the original value (via char
// refs) are prefixed by a 0xFFFF character. This is because some characters
// are legal if escaped only. And some escape chars are not subject to
// normalization rules.
bool SGXMLScanner::normalizeAttValue( const XMLAttDef* const attDef
, const XMLCh* const value
, XMLBuffer& toFill)
{
// A simple state value for a whitespace processing state machine
enum States
{
InWhitespace
, InContent
};
// Get the type and name
const XMLAttDef::AttTypes type = attDef->getType();
const XMLCh* const attrName = attDef->getFullName();
// Assume its going to go fine, and empty the target buffer in preperation
bool retVal = true;
toFill.reset();
// Get attribute def - to check to see if it's declared externally or not
bool isAttExternal = attDef->isExternal();
// Loop through the chars of the source value and normalize it according
// to the type.
States curState = InContent;
bool escaped;
bool firstNonWS = false;
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 (type == XMLAttDef::CData || type > XMLAttDef::Notation)
{
if (!escaped)
{
if ((nextCh == 0x09) || (nextCh == 0x0A) || (nextCh == 0x0D))
{
// Check Validity Constraint for Standalone document declaration
// XML 1.0, Section 2.9
if (fStandalone && fValidate && isAttExternal)
{
// Can't have a standalone document declaration of "yes" if attribute
// values are subject to normalisation
fValidator->emitError(XMLValid::NoAttNormForStandalone, attrName);
((SchemaAttDef *)attDef)->setValidity(PSVIDefs::INVALID);
}
nextCh = chSpace;
}
}
}
else
{
if (curState == InWhitespace)
{
{
if (firstNonWS)
toFill.append(chSpace);
curState = InContent;
firstNonWS = true;
}
else
{
srcPtr++;
continue;
}
}
else if (curState == InContent)
{
{
curState = InWhitespace;
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))
{
// Can't have a standalone document declaration of "yes" if attribute
// values are subject to normalisation
fValidator->emitError(XMLValid::NoAttNormForStandalone, attrName);
Gareth Reakes
committed
((SchemaAttDef *)attDef)->setValidity(PSVIDefs::INVALID);
}
}
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
((SchemaElementDecl *)fElemStack.topElement()->fThisElement)->updateValidityFromAttribute((SchemaAttDef *)attDef);
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
return retVal;
}
// This method will just normalize the input value as CDATA without
// any standalone checking.
bool SGXMLScanner::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.
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
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
SGXMLScanner::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
SGXMLScanner::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 SGXMLScanner::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);
fGrammar = fSchemaGrammar;
fGrammarType = Grammar::DTDGrammarType;
fRootGrammar = 0;
fValidator->setGrammar(fGrammar);
if (fValidatorFromUser) {
((SchemaValidator*) fValidator)->setErrorReporter(fErrorReporter);
((SchemaValidator*) fValidator)->setGrammarResolver(fGrammarResolver);
((SchemaValidator*) fValidator)->setExitOnFirstFatal(fExitOnFirstFatal);
}
// Reset validation
fValidate = (fValScheme == Val_Always) ? true : false;
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
// 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
fIDRefList->removeAll();
// 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;
fDoNamespaces = true;
fDoSchema = true;
// Reset the validators
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;
}
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
}
// 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 SGXMLScanner::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
const XMLCh* const rawBuf = toSend.getRawBuffer();
const 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 = topElem->fThisElement->getCharDataOpts();
if (charOpts == XMLElementDecl::NoCharData)
{
// They definitely cannot handle any type of char data
fValidator->emitError(XMLValid::NoCharDataInCM);
Gareth Reakes
committed
((SchemaElementDecl *)topElem->fThisElement)->setValidity(PSVIDefs::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)
{
// The normalized data can only be as large as the
// original size, so this will avoid allocating way
// too much or too little memory.
Khaled Noaman
committed
XMLBuffer toFill(len+1, fMemoryManager);
if (fNormalizeData)
{
DatatypeValidator* tempDV = ((SchemaElementDecl*) topElem->fThisElement)->getDatatypeValidator();
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, toFill.getRawBuffer(), tempBuf);
toFill.set(tempBuf.getRawBuffer());
}
}
// tell the schema validation about the character data for checkContent later
((SchemaValidator*) fValidator)->setDatatypeBuffer(toFill.getRawBuffer());
// call all active identity constraints
if (fMatcherStack->getMatcherCount())
fContent.append(toFill.getRawBuffer(), toFill.getLen());
if (fDocHandler)
fDocHandler->docCharacters(toFill.getRawBuffer(), toFill.getLen(), 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)
{
// The normalized data can only be as large as the
// original size, so this will avoid allocating way
// too much or too little memory.
Khaled Noaman
committed
XMLBuffer toFill(len+1, fMemoryManager);
if (fNormalizeData)
{
DatatypeValidator* tempDV = ((SchemaElementDecl*) topElem->fThisElement)->getDatatypeValidator();
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, toFill.getRawBuffer(), tempBuf);
toFill.set(tempBuf.getRawBuffer());
}
}
// tell the schema validation about the character data for checkContent later
((SchemaValidator*) fValidator)->setDatatypeBuffer(toFill.getRawBuffer());
// call all active identity constraints
if (fMatcherStack->getMatcherCount())
fContent.append(toFill.getRawBuffer(), toFill.getLen());
if (fDocHandler)
fDocHandler->docCharacters(toFill.getRawBuffer(), toFill.getLen(), false);
}
else
{
fValidator->emitError(XMLValid::NoCharDataInCM);
Gareth Reakes
committed
((SchemaElementDecl *)topElem->fThisElement)->setValidity(PSVIDefs::INVALID);
}
}
}
else
{
// call all active identity constraints
if (fMatcherStack->getMatcherCount())
fContent.append(toSend.getRawBuffer(), toSend.getLen());
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
// 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 SGXMLScanner::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