diff --git a/src/internal/ReaderMgr.hpp b/src/internal/ReaderMgr.hpp
index 617a03ec3c22c21ec81d177d1ee7b4eba2e9a883..18ca8227923c60be717c4662056b701797af4f4a 100644
--- a/src/internal/ReaderMgr.hpp
+++ b/src/internal/ReaderMgr.hpp
@@ -1,37 +1,37 @@
/*
* The Apache Software License, Version 1.1
- *
+ *
* Copyright (c) 1999-2000 The Apache Software Foundation. All rights
* reserved.
- *
+ *
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
- *
+ *
* 1. Redistributions of source code must retain the above copyright
- * notice, this list of conditions and the following disclaimer.
- *
+ * notice, this list of conditions and the following disclaimer.
+ *
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
- *
+ *
* 3. The end-user documentation included with the redistribution,
- * if any, must include the following acknowledgment:
+ * if any, must include the following acknowledgment:
* "This product includes software developed by the
* Apache Software Foundation (http://www.apache.org/)."
* Alternately, this acknowledgment may appear in the software itself,
* if and wherever such third-party acknowledgments normally appear.
- *
+ *
* 4. The names "Xerces" and "Apache Software Foundation" must
* not be used to endorse or promote products derived from this
- * software without prior written permission. For written
+ * software without prior written permission. For written
* permission, please contact apache\@apache.org.
- *
+ *
* 5. Products derived from this software may not be called "Apache",
* nor may "Apache" appear in their name, without prior written
* permission of the Apache Software Foundation.
- *
+ *
* THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
* WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
@@ -45,7 +45,7 @@
* OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
* ====================================================================
- *
+ *
* This software consists of voluntary contributions made by many
* individuals on behalf of the Apache Software Foundation, and was
* originally based on software copyright (c) 1999, International
@@ -56,6 +56,9 @@
/*
* $Log$
+ * Revision 1.13 2001/07/12 18:50:08 tng
+ * Some performance modification regarding standalone check and xml decl check.
+ *
* Revision 1.12 2000/09/09 00:18:18 andyh
* Reordered member variables in ThrowEOEJanitor. Patch submitted
* by Kirk Wylie.
@@ -187,6 +190,7 @@ public :
void skipQuotedString(const XMLCh quoteCh);
XMLCh skipUntilIn(const XMLCh* const listToSkip);
XMLCh skipUntilInOrWS(const XMLCh* const listToSkip);
+ bool peekString(const XMLCh* const toPeek);
// -----------------------------------------------------------------------
@@ -405,6 +409,11 @@ inline void ReaderMgr::skipPastChar(const XMLCh toSkipPast)
}
}
+inline bool ReaderMgr::peekString(const XMLCh* const toPeek)
+{
+ return fCurReader->peekString(toPeek);
+}
+
inline void ReaderMgr::setEntityHandler(XMLEntityHandler* const newHandler)
{
fEntityHandler = newHandler;
diff --git a/src/internal/XMLReader.cpp b/src/internal/XMLReader.cpp
index 1ea28b2b6404661df7be5ab6dc8e21ab866f94f3..7862f3a440297ae081d285b2190de5f94cc63436 100644
--- a/src/internal/XMLReader.cpp
+++ b/src/internal/XMLReader.cpp
@@ -1136,6 +1136,50 @@ bool XMLReader::skippedString(const XMLCh* const toSkip)
return true;
}
+//
+// This is just to peek if the next coming buffer
+// matches the string toPeek.
+// Similar to skippedString, but just the fCharIndex and fCurCol are not updated
+//
+bool XMLReader::peekString(const XMLCh* const toPeek)
+{
+ // Get the length of the string to skip
+ const unsigned int srcLen = XMLString::stringLen(toPeek);
+
+ //
+ // See if the current reader has enough chars to test against this
+ // string. If not, then ask it to reload its buffer. If that does not
+ // get us enough, then it cannot match.
+ //
+ // NOTE: This works because strings never have to cross a reader! And
+ // a string to skip will never have a new line in it, so we will never
+ // miss adjusting the current line.
+ //
+ unsigned int charsLeft = charsLeftInBuffer();
+ while (charsLeft < srcLen)
+ {
+ refreshCharBuffer();
+ unsigned int t = charsLeftInBuffer();
+ if (t == charsLeft) // if the refreshCharBuf() did not add anything new
+ return false; // give up and return.
+ charsLeft = t;
+ }
+
+
+
+
+ //
+ // Ok, now we now that the current reader has enough chars in its
+ // buffer and that its index is back at zero. So we can do a quick and
+ // dirty comparison straight to its buffer with no requirement to unget
+ // if it fails.
+ //
+ if (XMLString::compareNString(&fCharBuf[fCharIndex], toPeek, srcLen))
+ return false;
+
+ return true;
+}
+
// ---------------------------------------------------------------------------
// XMLReader: Setter methods (most are inlined)
diff --git a/src/internal/XMLReader.hpp b/src/internal/XMLReader.hpp
index eb93d74fb03f313e3385c0214e810ea89e39b02e..c967435ce7dab6d7ff924de63886f49b30871d34 100644
--- a/src/internal/XMLReader.hpp
+++ b/src/internal/XMLReader.hpp
@@ -56,6 +56,9 @@
/*
* $Log$
+ * Revision 1.17 2001/07/12 18:50:13 tng
+ * Some performance modification regarding standalone check and xml decl check.
+ *
* Revision 1.16 2001/05/11 13:26:17 tng
* Copyright update.
*
@@ -273,6 +276,7 @@ public:
bool skippedChar(const XMLCh toSkip);
bool skippedSpace();
bool skippedString(const XMLCh* const toSkip);
+ bool peekString(const XMLCh* const toPeek);
// -----------------------------------------------------------------------
diff --git a/src/internal/XMLScanner.cpp b/src/internal/XMLScanner.cpp
index a9e47827f64d8a149671792fc030a6dbce532e4f..4f57c94d11f69cf31d1863fdf9c963055ded5c80 100644
--- a/src/internal/XMLScanner.cpp
+++ b/src/internal/XMLScanner.cpp
@@ -2377,7 +2377,7 @@ bool XMLScanner::scanStartTag(bool& gotData)
// char refs expanded.
//
fReaderMgr.skipPastSpaces();
- if (!scanAttValue(attDef->getFullName(), fAttValueBuf, attDef->getType()))
+ if (!scanAttValue(attDef, fAttValueBuf))
{
static const XMLCh tmpList[] =
{
@@ -2512,7 +2512,8 @@ bool XMLScanner::scanStartTag(bool& gotData)
// top again.
//
emitError(XMLErrs::ExpectedAttrName);
- scanAttValue(XMLUni::fgZeroLenString, fAttValueBuf, XMLAttDef::CData);
+ fReaderMgr.getNextChar();
+ fReaderMgr.skipQuotedString(nextCh);
fReaderMgr.skipPastSpaces();
continue;
}
@@ -3420,58 +3421,3 @@ XMLScanner::resolveQName( const XMLCh* const qName
return uriId;
}
-bool XMLScanner::checkXMLDecl(bool startWithAngle) {
- //
- // [23] XMLDecl ::= '<?xml' VersionInfo EncodingDecl? SDDecl? S? '?>'
- // [24] VersionInfo ::= S 'version' Eq ("'" VersionNum "'" | '"' VersionNum '"')
- //
- // [3] S ::= (#x20 | #x9 | #xD | #xA)+
- //
-
- if (startWithAngle) {
- if (fReaderMgr.skippedString(XMLUni::fgXMLDeclStringSpace)
- || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringHTab)
- || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringLF)
- || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringCR))
- {
- return true;
- }
- else if (fReaderMgr.skippedString(XMLUni::fgXMLDeclStringSpaceU)
- || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringHTabU)
- || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringLFU)
- || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringCRU))
- {
- //
- // Just in case, check for upper case. If found, issue
- // an error, but keep going.
- //
- emitError(XMLErrs::XMLDeclMustBeLowerCase);
- return true;
- }
- }
- else {
- if (fReaderMgr.skippedString(XMLUni::fgXMLStringSpace)
- || fReaderMgr.skippedString(XMLUni::fgXMLStringHTab)
- || fReaderMgr.skippedString(XMLUni::fgXMLStringLF)
- || fReaderMgr.skippedString(XMLUni::fgXMLStringCR))
- {
- return true;
- }
- else if (fReaderMgr.skippedString(XMLUni::fgXMLStringSpaceU)
- || fReaderMgr.skippedString(XMLUni::fgXMLStringHTabU)
- || fReaderMgr.skippedString(XMLUni::fgXMLStringLFU)
- || fReaderMgr.skippedString(XMLUni::fgXMLStringCRU))
- {
- //
- // Just in case, check for upper case. If found, issue
- // an error, but keep going.
- //
- emitError(XMLErrs::XMLDeclMustBeLowerCase);
- return true;
- }
- }
-
- return false;
-}
-
-
diff --git a/src/internal/XMLScanner.hpp b/src/internal/XMLScanner.hpp
index 06befa032ce58b422c1e5cead054291c4b969b4c..bef1130e9e676e381f55e7f7fd865e2bfddabcad 100644
--- a/src/internal/XMLScanner.hpp
+++ b/src/internal/XMLScanner.hpp
@@ -56,6 +56,9 @@
/*
* $Log$
+ * Revision 1.26 2001/07/12 18:50:17 tng
+ * Some performance modification regarding standalone check and xml decl check.
+ *
* Revision 1.25 2001/07/10 21:09:31 tng
* Give proper error messsage when scanning external id.
*
@@ -519,10 +522,15 @@ private :
void checkIDRefs();
bool isLegalToken(const XMLPScanToken& toCheck);
bool normalizeAttValue
+ (
+ const XMLAttDef* const attDef
+ , const XMLCh* const value
+ , XMLBuffer& toFill
+ );
+ bool normalizeAttRawValue
(
const XMLCh* const attrName
, const XMLCh* const value
- , const XMLAttDef::AttTypes type
, XMLBuffer& toFill
);
@@ -569,9 +577,8 @@ private :
);
bool scanAttValue
(
- const XMLCh* const attrName
+ const XMLAttDef* const attDef
, XMLBuffer& toFill
- , const XMLAttDef::AttTypes type
);
void scanCDSection();
void scanCharData(XMLBuffer& toToUse);
@@ -1116,4 +1123,62 @@ inline void XMLScanner::setDoValidation(const bool validate)
else
fValScheme = Val_Never;
}
+
+inline bool XMLScanner::checkXMLDecl(bool startWithAngle) {
+ //
+ // [23] XMLDecl ::= '<?xml' VersionInfo EncodingDecl? SDDecl? S? '?>'
+ // [24] VersionInfo ::= S 'version' Eq ("'" VersionNum "'" | '"' VersionNum '"')
+ //
+ // [3] S ::= (#x20 | #x9 | #xD | #xA)+
+ //
+
+ if (startWithAngle) {
+ if (fReaderMgr.peekString(XMLUni::fgXMLDeclString)) {
+ if (fReaderMgr.skippedString(XMLUni::fgXMLDeclStringSpace)
+ || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringHTab)
+ || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringLF)
+ || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringCR))
+ {
+ return true;
+ }
+ else if (fReaderMgr.skippedString(XMLUni::fgXMLDeclStringSpaceU)
+ || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringHTabU)
+ || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringLFU)
+ || fReaderMgr.skippedString(XMLUni::fgXMLDeclStringCRU))
+ {
+ //
+ // Just in case, check for upper case. If found, issue
+ // an error, but keep going.
+ //
+ emitError(XMLErrs::XMLDeclMustBeLowerCase);
+ return true;
+ }
+ }
+ }
+ else {
+ if (fReaderMgr.peekString(XMLUni::fgXMLString)) {
+ if (fReaderMgr.skippedString(XMLUni::fgXMLStringSpace)
+ || fReaderMgr.skippedString(XMLUni::fgXMLStringHTab)
+ || fReaderMgr.skippedString(XMLUni::fgXMLStringLF)
+ || fReaderMgr.skippedString(XMLUni::fgXMLStringCR))
+ {
+ return true;
+ }
+ else if (fReaderMgr.skippedString(XMLUni::fgXMLStringSpaceU)
+ || fReaderMgr.skippedString(XMLUni::fgXMLStringHTabU)
+ || fReaderMgr.skippedString(XMLUni::fgXMLStringLFU)
+ || fReaderMgr.skippedString(XMLUni::fgXMLStringCRU))
+ {
+ //
+ // Just in case, check for upper case. If found, issue
+ // an error, but keep going.
+ //
+ emitError(XMLErrs::XMLDeclMustBeLowerCase);
+ return true;
+ }
+ }
+ }
+
+ return false;
+}
#endif
diff --git a/src/internal/XMLScanner2.cpp b/src/internal/XMLScanner2.cpp
index 285df72a24ab927934716c52b8ca8491eeec8014..157726b480ed8684fd8c1dcdafbd6faeed6dbc5f 100644
--- a/src/internal/XMLScanner2.cpp
+++ b/src/internal/XMLScanner2.cpp
@@ -313,9 +313,8 @@ XMLScanner::buildAttList(const RefVectorOf<KVStringPair>& providedAttrs
//
normalizeAttValue
(
- curPair->getKey()
+ attDef
, curPair->getValue()
- , attDef->getType()
, normBuf
);
@@ -350,11 +349,10 @@ XMLScanner::buildAttList(const RefVectorOf<KVStringPair>& providedAttrs
{
// Just normalize as CDATA
attType = XMLAttDef::CData;
- normalizeAttValue
+ normalizeAttRawValue
(
curPair->getKey()
, curPair->getValue()
- , XMLAttDef::CData
, normBuf
);
}
@@ -546,9 +544,8 @@ bool XMLScanner::isLegalToken(const XMLPScanToken& toCheck)
// are legal if escaped only. And some escape chars are not subject to
// normalization rules.
//
-bool XMLScanner::normalizeAttValue( const XMLCh* const attrName
+bool XMLScanner::normalizeAttValue( const XMLAttDef* const attDef
, const XMLCh* const value
- , const XMLAttDef::AttTypes type
, XMLBuffer& toFill)
{
// A simple state value for a whitespace processing state machine
@@ -558,6 +555,10 @@ bool XMLScanner::normalizeAttValue( const XMLCh* const attrName
, 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();
@@ -565,13 +566,7 @@ bool XMLScanner::normalizeAttValue( const XMLCh* const attrName
//
// Get attribute def - to check to see if it's declared externally or not
//
- bool added = false;
- bool isAttExternal = false;
- const ElemStack::StackElem* topElem = fElemStack.topElement();
- if (topElem && topElem->fThisElement) {
- const XMLAttDef* attDef = topElem->fThisElement->findAttr(attrName, 0, 0, 0, XMLElementDecl::FailIfNotFound, added);
- isAttExternal = (attDef) ? attDef->isExternal() : false;
- }
+ bool isAttExternal = attDef->isExternal();
//
// Loop through the chars of the source value and normalize it according
@@ -639,18 +634,6 @@ bool XMLScanner::normalizeAttValue( const XMLCh* const attrName
}
else
{
- //
- // 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);
- }
srcPtr++;
continue;
}
@@ -661,13 +644,13 @@ bool XMLScanner::normalizeAttValue( const XMLCh* const attrName
{
curState = InWhitespace;
srcPtr++;
- if (!firstNonWS || (nextCh != chSpace))
+ //
+ // Check Validity Constraint for Standalone document declaration
+ // XML 1.0, Section 2.9
+ //
+ if (fStandalone && fValidate && isAttExternal)
{
- //
- // Check Validity Constraint for Standalone document declaration
- // XML 1.0, Section 2.9
- //
- if (fStandalone && fValidate && isAttExternal)
+ if (!firstNonWS || (nextCh != chSpace) || (fReaderMgr.lookingAtSpace()))
{
//
// Can't have a standalone document declaration of "yes" if attribute
@@ -691,6 +674,76 @@ bool XMLScanner::normalizeAttValue( const XMLCh* const attrName
return retVal;
}
+//
+// This method will just normalize the input value as CDATA without
+// any standalone checking.
+//
+bool XMLScanner::normalizeAttRawValue( const XMLCh* const attrName
+ , const XMLCh* const value
+ , XMLBuffer& toFill)
+{
+ // A simple state value for a whitespace processing state machine
+ enum States
+ {
+ InWhitespace
+ , InContent
+ };
+
+ // 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.
+ //
+ 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 (!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.
+ //
+ if (XMLReader::isWhitespace(nextCh))
+ 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
XMLScanner::resolvePrefix( const XMLCh* const prefix
, const ElemStack::MapModes mode)
@@ -1096,7 +1149,7 @@ void XMLScanner::updateNSMap(const XMLCh* const attrName
// care about the return value. An error was issued for the error, which
// is all we care about here.
//
- normalizeAttValue(attrName, attrValue, XMLAttDef::CData, normalBuf);
+ normalizeAttRawValue(attrName, attrValue, normalBuf);
//
// Ok, we have to get the unique id for the attribute value, which is the
@@ -1566,9 +1619,8 @@ bool XMLScanner::basicAttrValueScan(const XMLCh* const attrName, XMLBuffer& toFi
}
-bool XMLScanner::scanAttValue( const XMLCh* const attrName
- , XMLBuffer& toFill
- , const XMLAttDef::AttTypes type)
+bool XMLScanner::scanAttValue( const XMLAttDef* const attDef
+ , XMLBuffer& toFill)
{
enum States
{
@@ -1576,6 +1628,9 @@ bool XMLScanner::scanAttValue( const XMLCh* const attrName
, InContent
};
+ // Get the type and name
+ const XMLAttDef::AttTypes type = attDef->getType();
+ const XMLCh* const attrName = attDef->getFullName();
// Reset the target buffer
toFill.reset();
@@ -1594,13 +1649,7 @@ bool XMLScanner::scanAttValue( const XMLCh* const attrName
//
// Get attribute def - to check to see if it's declared externally or not
//
- bool added = false;
- bool isAttExternal = false;
- const ElemStack::StackElem* topElem = fElemStack.topElement();
- if (topElem && topElem->fThisElement) {
- const XMLAttDef* attDef = topElem->fThisElement->findAttr(attrName, 0, 0, 0, XMLElementDecl::FailIfNotFound, added);
- isAttExternal = (attDef) ? attDef->isExternal() : false;
- }
+ bool isAttExternal = attDef->isExternal();
//
// Loop until we get the attribute value. Note that we use a double
@@ -1765,18 +1814,6 @@ bool XMLScanner::scanAttValue( const XMLCh* const attrName
}
else
{
- //
- // 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);
- }
continue;
}
}
@@ -1785,13 +1822,13 @@ bool XMLScanner::scanAttValue( const XMLCh* const attrName
if (XMLReader::isWhitespace(nextCh))
{
curState = InWhitespace;
- if (!firstNonWS || (nextCh != chSpace))
+ //
+ // Check Validity Constraint for Standalone document declaration
+ // XML 1.0, Section 2.9
+ //
+ if (fStandalone && fValidate && isAttExternal)
{
- //
- // Check Validity Constraint for Standalone document declaration
- // XML 1.0, Section 2.9
- //
- if (fStandalone && fValidate && isAttExternal)
+ if (!firstNonWS || (nextCh != chSpace) || (fReaderMgr.lookingAtSpace()))
{
//
// Can't have a standalone document declaration of "yes" if attribute
diff --git a/src/util/XMLUni.cpp b/src/util/XMLUni.cpp
index a9190c2a0a68cd74b40a1ddd4df8e448853df3d1..724c8c2d07be0ad9eaa0b6cf0f45bd644f02e35c 100644
--- a/src/util/XMLUni.cpp
+++ b/src/util/XMLUni.cpp
@@ -509,6 +509,11 @@ const XMLCh XMLUni::fgXMLStringLFU[] =
};
//<?xml{S}
+const XMLCh XMLUni::fgXMLDeclString[] =
+{
+ chOpenAngle, chQuestion, chLatin_x, chLatin_m, chLatin_l, chNull
+};
+
const XMLCh XMLUni::fgXMLDeclStringSpace[] =
{
chOpenAngle, chQuestion, chLatin_x, chLatin_m, chLatin_l, chSpace, chNull
diff --git a/src/util/XMLUni.hpp b/src/util/XMLUni.hpp
index 02f8b7edc904719a7e12bf216410fa990f1ad00f..88c4a91f6354f504a867a2fc0ae47456f0b56358 100644
--- a/src/util/XMLUni.hpp
+++ b/src/util/XMLUni.hpp
@@ -164,6 +164,7 @@ public :
static const XMLCh fgXMLStringHTabU[];
static const XMLCh fgXMLStringCRU[];
static const XMLCh fgXMLStringLFU[];
+ static const XMLCh fgXMLDeclString[];
static const XMLCh fgXMLDeclStringSpace[];
static const XMLCh fgXMLDeclStringHTab[];
static const XMLCh fgXMLDeclStringLF[];