fix assert logic

- Add null pointer check in XMLDocument::DeleteNode() method
- Add test case to verify null pointer handling
- Prevents segmentation fault when calling DeleteNode(nullptr)

.github/workflows/test.yml

@@ -6,15 +6,15 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        os: [ windows-2019, macos-latest, ubuntu-20.04 ]
+        os: [ windows-latest, macos-latest, ubuntu-latest ]
         cmake: [ 3.15, 3.x ]
         include:
-          - os: windows-2019
+          - os: windows-latest
             static_postfix: _static
             tree: tree /F
             CXX: cl
 
-          - os: ubuntu-20.04
+          - os: ubuntu-latest
             tree: tree
 
           - os: macos-latest
@@ -31,10 +31,10 @@ jobs:
     runs-on: ${{ matrix.os }}
     steps:
       # System set-up
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v4
       - uses: ilammy/msvc-dev-cmd@v1
-      - uses: seanmiddleditch/gha-setup-ninja@master
-      - uses: jwlawson/actions-setup-cmake@v1.8
+      - uses: seanmiddleditch/gha-setup-ninja@v4
+      - uses: jwlawson/actions-setup-cmake@v2
         with:
           cmake-version: ${{ matrix.cmake }}
 

CMakeLists.txt

@@ -1,5 +1,5 @@
 cmake_minimum_required(VERSION 3.15)
-project(tinyxml2 VERSION 10.0.0)
+project(tinyxml2 VERSION 11.0.0)
 
 include(CTest)
 option(tinyxml2_BUILD_TESTING "Build tests for tinyxml2" "${BUILD_TESTING}")
@@ -63,7 +63,9 @@ endif ()
 include(GNUInstallDirs)
 include(CMakePackageConfigHelpers)
 
-## Custom locations
+## Custom settings
+option(tinyxml2_INSTALL_PKGCONFIG "Create and install pkgconfig files" ON)
+
 set(tinyxml2_INSTALL_PKGCONFIGDIR "${CMAKE_INSTALL_LIBDIR}/pkgconfig"
     CACHE PATH "Directory for pkgconfig files")
 
@@ -121,10 +123,12 @@ install(
 
 ## pkg-config
 
-configure_file(cmake/tinyxml2.pc.in tinyxml2.pc.gen @ONLY)
-file(GENERATE OUTPUT tinyxml2.pc INPUT "${CMAKE_CURRENT_BINARY_DIR}/tinyxml2.pc.gen")
-install(
-    FILES "${CMAKE_CURRENT_BINARY_DIR}/tinyxml2.pc"
-    DESTINATION "${tinyxml2_INSTALL_PKGCONFIGDIR}"
-    COMPONENT tinyxml2_development
-)
+if (tinyxml2_INSTALL_PKGCONFIG)
+    configure_file(cmake/tinyxml2.pc.in tinyxml2.pc.gen @ONLY)
+    file(GENERATE OUTPUT tinyxml2.pc INPUT "${CMAKE_CURRENT_BINARY_DIR}/tinyxml2.pc.gen")
+    install(
+        FILES "${CMAKE_CURRENT_BINARY_DIR}/tinyxml2.pc"
+        DESTINATION "${tinyxml2_INSTALL_PKGCONFIGDIR}"
+        COMPONENT tinyxml2_development
+    )
+endif ()

contrib/html5-printer.cpp

@@ -7,7 +7,7 @@
 
 //  In HTML5, there are 16 so-called "void" elements.  "void elements" NEVER have
 //  inner content (but they MAY have attributes), and are assumed to be self-closing.
-//  An example of a self-closig HTML5 element is "<br/>" (line break)
+//  An example of a self-closing HTML5 element is "<br/>" (line break)
 //  All other elements are called "non-void" and MUST never self-close.
 //  Examples: "<div class='lolcats'></div>".
 

dox

@@ -38,7 +38,7 @@ PROJECT_NAME           = "TinyXML-2"
 # could be handy for archiving the generated documentation or if some version
 # control system is used.
 
-PROJECT_NUMBER = 10.0.0
+PROJECT_NUMBER = 11.0.0
 
 # Using the PROJECT_BRIEF tag one can provide an optional one line description
 # for a project that appears at the top of each page and should give viewer a

meson.build

@@ -22,7 +22,7 @@
 project(
     'tinyxml2',
     ['cpp'],
-    version : '10.0.0',
+    version : '11.0.0',
     meson_version : '>= 0.49.0',
 )
 

readme.md

@@ -59,7 +59,7 @@ browsers or have more complete XML needs, TinyXML-2 is not the parser for you.
 TinyXML-1 vs. TinyXML-2
 -----------------------
 
-TinyXML-2 long been the focus of all development. It is well tested 
+TinyXML-2 has long been the focus of all development. It is well tested 
 and should be used instead of TinyXML-1.
 
 TinyXML-2 uses a similar API to TinyXML-1 and the same
@@ -145,7 +145,7 @@ It essentially causes the XML to be parsed twice.
 
 For applications that need to know about text nodes that are composed entirely of 
 whitespace, PEDANTIC_WHITESPACE is available. PEDANTIC_WHITESPACE maintains all the
-whilespace between elements. 
+whitespace between elements. 
 
 PEDANTIC_WHITESPACE is a new mode and not as tested as the other whitespace modes.
 
@@ -178,7 +178,7 @@ will have the Value() of "Far & Away" when queried from the XMLText object,
 and will be written back to the XML stream/file as an ampersand.
 
 Additionally, any character can be specified by its Unicode code point:
-The syntax ` ` or ` ` are both to the non-breaking space character.
+The syntax ` ` or ` ` both refer to the non-breaking space character.
 This is called a 'numeric character reference'. Any numeric character reference
 that isn't one of the special entities above, will be read, but written as a
 regular code point. The output is correct, but the entity syntax isn't preserved.
@@ -270,7 +270,7 @@ tinyxml2.h files in your project and build with your other source code.
 
 There is also a CMake build included. CMake is the general build for TinyXML-2.
 
-(Additional build systems are costly to maintain, and tend to bit-rot. They are
+(Additional build systems are costly to maintain, and tend to become outdated. They are
 being removed over time.)
 
 Building TinyXML-2 - Using vcpkg

tinyxml2.cpp

@@ -24,7 +24,7 @@ distribution.
 #include "tinyxml2.h"
 
 #include <new>		// yes, this one new style header, is in the Android SDK.
-#if defined(ANDROID_NDK) || defined(__BORLANDC__) || defined(__QNXNTO__)
+#if defined(ANDROID_NDK) || defined(__BORLANDC__) || defined(__QNXNTO__) || defined(__CC_ARM)
 #   include <stddef.h>
 #   include <stdarg.h>
 #else
@@ -32,6 +32,27 @@ distribution.
 #   include <cstdarg>
 #endif
 
+// Handle fallthrough attribute for different compilers
+#ifndef __has_attribute
+#   define __has_attribute(x) 0
+#endif
+#ifndef __has_cpp_attribute
+#  define __has_cpp_attribute(x) 0
+#endif
+
+#if defined(_MSC_VER)
+#   define TIXML_FALLTHROUGH (void(0))
+#elif (__cplusplus >= 201703L && __has_cpp_attribute(fallthrough))
+#   define TIXML_FALLTHROUGH [[fallthrough]]
+#elif __has_cpp_attribute(clang::fallthrough)
+#   define TIXML_FALLTHROUGH [[clang::fallthrough]]
+#elif __has_attribute(fallthrough)
+#   define TIXML_FALLTHROUGH __attribute__((fallthrough))
+#else
+#   define TIXML_FALLTHROUGH (void(0))
+#endif
+
+
 #if defined(_MSC_VER) && (_MSC_VER >= 1400 ) && (!defined WINCE)
 	// Microsoft Visual Studio, version 2005 and higher. Not WinCE.
 	/*int _snprintf_s(
@@ -93,6 +114,9 @@ distribution.
 	#define TIXML_VSNPRINTF	vsnprintf
 	static inline int TIXML_VSCPRINTF( const char* format, va_list va )
 	{
+	    if (!format) {
+	        return 0;
+	    }
 		int len = vsnprintf( 0, 0, format, va );
 		TIXMLASSERT( len >= 0 );
 		return len;
@@ -106,14 +130,9 @@ distribution.
 #elif defined(__APPLE__) || defined(__FreeBSD__) || defined(__OpenBSD__) || defined(__NetBSD__) || defined(__DragonFly__) || defined(__CYGWIN__)
 	#define TIXML_FSEEK fseeko
 	#define TIXML_FTELL ftello
-#elif defined(__ANDROID__) 
-    #if __ANDROID_API__ > 24
-        #define TIXML_FSEEK fseeko64
-        #define TIXML_FTELL ftello64
-    #else
-        #define TIXML_FSEEK fseeko
-        #define TIXML_FTELL ftello
-    #endif
+#elif defined(__ANDROID__) && __ANDROID_API__ > 24
+	#define TIXML_FSEEK fseeko64
+	#define TIXML_FTELL ftello64
 #else
 	#define TIXML_FSEEK fseek
 	#define TIXML_FTELL ftell
@@ -239,13 +258,13 @@ char* StrPair::ParseName( char* p )
     if ( !p || !(*p) ) {
         return 0;
     }
-    if ( !XMLUtil::IsNameStartChar( (unsigned char) *p ) ) {
+    if ( !XMLUtil::IsNameStartChar( static_cast<unsigned char>(*p) ) ) {
         return 0;
     }
 
     char* const start = p;
     ++p;
-    while ( *p && XMLUtil::IsNameChar( (unsigned char) *p ) ) {
+    while ( *p && XMLUtil::IsNameChar( static_cast<unsigned char>(*p) ) ) {
         ++p;
     }
 
@@ -451,17 +470,17 @@ void XMLUtil::ConvertUTF32ToUTF8( unsigned long input, char* output, int* length
             --output;
             *output = static_cast<char>((input | BYTE_MARK) & BYTE_MASK);
             input >>= 6;
-            //fall through
+            TIXML_FALLTHROUGH;
         case 3:
             --output;
             *output = static_cast<char>((input | BYTE_MARK) & BYTE_MASK);
             input >>= 6;
-            //fall through
+            TIXML_FALLTHROUGH;
         case 2:
             --output;
             *output = static_cast<char>((input | BYTE_MARK) & BYTE_MASK);
             input >>= 6;
-            //fall through
+            TIXML_FALLTHROUGH;
         case 1:
             --output;
             *output = static_cast<char>(input | FIRST_BYTE_MARK[*length]);
@@ -472,102 +491,94 @@ void XMLUtil::ConvertUTF32ToUTF8( unsigned long input, char* output, int* length
 }
 
 
-const char* XMLUtil::GetCharacterRef( const char* p, char* value, int* length )
+const char* XMLUtil::GetCharacterRef(const char* p, char* value, int* length)
 {
-    // Presume an entity, and pull it out.
+    // Assume an entity, and pull it out.
     *length = 0;
 
-    if ( *(p+1) == '#' && *(p+2) ) {
-        unsigned long ucs = 0;
-        TIXMLASSERT( sizeof( ucs ) >= 4 );
+    static const uint32_t MAX_CODE_POINT = 0x10FFFF;
+
+    if (*(p + 1) == '#' && *(p + 2)) {
+        uint32_t ucs = 0;
         ptrdiff_t delta = 0;
-        unsigned mult = 1;
+        uint32_t mult = 1;
         static const char SEMICOLON = ';';
 
-        if ( *(p+2) == 'x' ) {
+        bool hex = false;
+        uint32_t radix = 10;
+        const char* q = 0;
+        char terminator = '#';
+
+        if (*(p + 2) == 'x') {
             // Hexadecimal.
-            const char* q = p+3;
-            if ( !(*q) ) {
-                return 0;
-            }
+            hex = true;
+            radix = 16;
+            terminator = 'x';
 
-            q = strchr( q, SEMICOLON );
-
-            if ( !q ) {
-                return 0;
-            }
-            TIXMLASSERT( *q == SEMICOLON );
-
-            delta = q-p;
-            --q;
-
-            while ( *q != 'x' ) {
-                unsigned int digit = 0;
-
-                if ( *q >= '0' && *q <= '9' ) {
-                    digit = *q - '0';
-                }
-                else if ( *q >= 'a' && *q <= 'f' ) {
-                    digit = *q - 'a' + 10;
-                }
-                else if ( *q >= 'A' && *q <= 'F' ) {
-                    digit = *q - 'A' + 10;
-                }
-                else {
-                    return 0;
-                }
-                TIXMLASSERT( digit < 16 );
-                TIXMLASSERT( digit == 0 || mult <= UINT_MAX / digit );
-                const unsigned int digitScaled = mult * digit;
-                TIXMLASSERT( ucs <= ULONG_MAX - digitScaled );
-                ucs += digitScaled;
-                TIXMLASSERT( mult <= UINT_MAX / 16 );
-                mult *= 16;
-                --q;
-            }
+            q = p + 3;
         }
         else {
             // Decimal.
-            const char* q = p+2;
-            if ( !(*q) ) {
+            q = p + 2;
+        }
+        if (!(*q)) {
+            return 0;
+        }
+
+        q = strchr(q, SEMICOLON);
+        if (!q) {
+            return 0;
+        }
+        TIXMLASSERT(*q == SEMICOLON);
+
+        delta = q - p;
+        --q;
+
+        while (*q != terminator) {
+            uint32_t digit = 0;
+
+            if (*q >= '0' && *q <= '9') {
+                digit = *q - '0';
+            }
+            else if (hex && (*q >= 'a' && *q <= 'f')) {
+                digit = *q - 'a' + 10;
+            }
+            else if (hex && (*q >= 'A' && *q <= 'F')) {
+                digit = *q - 'A' + 10;
+            }
+            else {
                 return 0;
             }
+            TIXMLASSERT(digit < radix);
 
-            q = strchr( q, SEMICOLON );
-
-            if ( !q ) {
-                return 0;
+            const unsigned int digitScaled = mult * digit;
+            ucs += digitScaled;
+            mult *= radix;       
+            
+            // Security check: could a value exist that is out of range?
+            // Easily; limit to the MAX_CODE_POINT, which also allows for a
+            // bunch of leading zeroes.
+            if (mult > MAX_CODE_POINT) {
+                mult = MAX_CODE_POINT;
             }
-            TIXMLASSERT( *q == SEMICOLON );
-
-            delta = q-p;
             --q;
-
-            while ( *q != '#' ) {
-                if ( *q >= '0' && *q <= '9' ) {
-                    const unsigned int digit = *q - '0';
-                    TIXMLASSERT( digit < 10 );
-                    TIXMLASSERT( digit == 0 || mult <= UINT_MAX / digit );
-                    const unsigned int digitScaled = mult * digit;
-                    TIXMLASSERT( ucs <= ULONG_MAX - digitScaled );
-                    ucs += digitScaled;
-                }
-                else {
-                    return 0;
-                }
-                TIXMLASSERT( mult <= UINT_MAX / 10 );
-                mult *= 10;
-                --q;
-            }
+        }
+        // Out of range:
+        if (ucs > MAX_CODE_POINT) {
+            return 0;
         }
         // convert the UCS to UTF-8
-        ConvertUTF32ToUTF8( ucs, value, length );
+        ConvertUTF32ToUTF8(ucs, value, length);
+		if (length == 0) {
+            // If length is 0, there was an error. (Security? Bad input?)
+            // Fail safely.
+			return 0;
+		}
         return p + delta + 1;
     }
-    return p+1;
+    return p + 1;
 }
 
-
 void XMLUtil::ToStr( int v, char* buffer, int bufferSize )
 {
     TIXML_SNPRINTF( buffer, bufferSize, "%d", v );
@@ -610,7 +621,7 @@ void XMLUtil::ToStr( int64_t v, char* buffer, int bufferSize )
 void XMLUtil::ToStr( uint64_t v, char* buffer, int bufferSize )
 {
     // horrible syntax trick to make the compiler happy about %llu
-    TIXML_SNPRINTF(buffer, bufferSize, "%llu", (long long)v);
+    TIXML_SNPRINTF(buffer, bufferSize, "%llu", static_cast<unsigned long long>(v));
 }
 
 bool XMLUtil::ToInt(const char* str, int* value)
@@ -705,7 +716,7 @@ bool XMLUtil::ToInt64(const char* str, int64_t* value)
 bool XMLUtil::ToUnsigned64(const char* str, uint64_t* value) {
     unsigned long long v = 0;	// horrible syntax trick to make the compiler happy about %llu
     if(TIXML_SSCANF(str, IsPrefixHex(str) ? "%llx" : "%llu", &v) == 1) {
-        *value = (uint64_t)v;
+        *value = static_cast<uint64_t>(v);
         return true;
     }
     return false;
@@ -823,7 +834,17 @@ XMLNode::XMLNode( XMLDocument* doc ) :
 
 XMLNode::~XMLNode()
 {
-    DeleteChildren();
+    // Fast path: this node is dying, so maintaining _firstChild/_lastChild and
+    // sibling _prev/_next links is unnecessary. Only _parent must be zeroed to
+    // satisfy the MarkInUse assertion inside DeleteNode.
+    XMLNode *currentChild = _firstChild;
+    while (currentChild != NULL) {
+        XMLNode *next = currentChild->_next;
+        currentChild->_parent = 0;
+        DeleteNode(currentChild);
+        currentChild = next;
+    }
+
     if ( _parent ) {
         _parent->Unlink( this );
     }
@@ -1982,7 +2003,7 @@ char* XMLElement::ParseAttributes( char* p, int* curLineNumPtr )
         }
 
         // attribute.
-        if (XMLUtil::IsNameStartChar( (unsigned char) *p ) ) {
+        if (XMLUtil::IsNameStartChar( static_cast<unsigned char>(*p) ) ) {
             XMLAttribute* attrib = CreateAttribute();
             TIXMLASSERT( attrib );
             attrib->_parseLineNum = _document->_parseCurLineNum;
@@ -2226,7 +2247,7 @@ void XMLDocument::MarkInUse(const XMLNode* const node)
 	TIXMLASSERT(node);
 	TIXMLASSERT(node->_parent == 0);
 
-	for (int i = 0; i < _unlinked.Size(); ++i) {
+	for (size_t i = 0; i < _unlinked.Size(); ++i) {
 		if (node == _unlinked[i]) {
 			_unlinked.SwapRemove(i);
 			break;
@@ -2336,9 +2357,10 @@ static FILE* callfopen( const char* filepath, const char* mode )
     return fp;
 }
 
-void XMLDocument::DeleteNode( XMLNode* node )	{
-    TIXMLASSERT( node );
-    TIXMLASSERT(node->_document == this );
+void XMLDocument::DeleteNode( XMLNode* node )	{   
+    if(node == 0) {
+        return; // check for null pointer
+    }
     if (node->_parent) {
         node->_parent->DeleteChild( node );
     }
@@ -2509,7 +2531,7 @@ void XMLDocument::ClearError() {
 
 void XMLDocument::SetError( XMLError error, int lineNum, const char* format, ... )
 {
-    TIXMLASSERT( error >= 0 && error < XML_ERROR_COUNT );
+    TIXMLASSERT(error >= 0 && error < XML_ERROR_COUNT);
     _errorID = error;
     _errorLineNum = lineNum;
 	_errorStr.Reset();
@@ -2518,7 +2540,8 @@ void XMLDocument::SetError( XMLError error, int lineNum, const char* format, ...
     char* buffer = new char[BUFFER_SIZE];
 
     TIXMLASSERT(sizeof(error) <= sizeof(int));
-    TIXML_SNPRINTF(buffer, BUFFER_SIZE, "Error=%s ErrorID=%d (0x%x) Line number=%d", ErrorIDToName(error), int(error), int(error), lineNum);
+    TIXML_SNPRINTF(buffer, BUFFER_SIZE, "Error=%s ErrorID=%d (0x%x) Line number=%d",
+        ErrorIDToName(error), static_cast<int>(error), static_cast<unsigned int>(error), lineNum);
 
 	if (format) {
 		size_t len = strlen(buffer);
@@ -2589,7 +2612,7 @@ void XMLDocument::PopDepth()
 	--_parsingDepth;
 }
 
-XMLPrinter::XMLPrinter( FILE* file, bool compact, int depth ) :
+XMLPrinter::XMLPrinter( FILE* file, bool compact, int depth, EscapeAposCharsInAttributes aposInAttributes ) :
     _elementJustOpened( false ),
     _stack(),
     _firstElement( true ),
@@ -2606,9 +2629,11 @@ XMLPrinter::XMLPrinter( FILE* file, bool compact, int depth ) :
     }
     for( int i=0; i<NUM_ENTITIES; ++i ) {
         const char entityValue = entities[i].value;
-        const unsigned char flagIndex = static_cast<unsigned char>(entityValue);
-        TIXMLASSERT( flagIndex < ENTITY_RANGE );
-        _entityFlag[flagIndex] = true;
+        if ((aposInAttributes == ESCAPE_APOS_CHARS_IN_ATTRIBUTES) || (entityValue != SINGLE_QUOTE)) {
+            const unsigned char flagIndex = static_cast<unsigned char>(entityValue);
+            TIXMLASSERT( flagIndex < ENTITY_RANGE );
+            _entityFlag[flagIndex] = true;
+        }
     }
     _restrictedEntityFlag[static_cast<unsigned char>('&')] = true;
     _restrictedEntityFlag[static_cast<unsigned char>('<')] = true;
@@ -2645,7 +2670,7 @@ void XMLPrinter::Write( const char* data, size_t size )
         fwrite ( data , sizeof(char), size, _fp);
     }
     else {
-        char* p = _buffer.PushArr( static_cast<int>(size) ) - 1;   // back up over the null terminator.
+        char* p = _buffer.PushArr( size ) - 1;   // back up over the null terminator.
         memcpy( p, data, size );
         p[size] = 0;
     }

tinyxml2.h

@@ -42,14 +42,6 @@ distribution.
 #endif
 #include <stdint.h>
 
-/*
-	gcc:
-        g++ -Wall -DTINYXML2_DEBUG tinyxml2.cpp xmltest.cpp -o gccxmltest.exe
-
-    Formatting, Artistic Style:
-        AStyle.exe --style=1tbs --indent-switches --break-closing-brackets --indent-preprocessor tinyxml2.cpp tinyxml2.h
-*/
-
 #if defined( _DEBUG ) || defined (__DEBUG__)
 #   ifndef TINYXML2_DEBUG
 #       define TINYXML2_DEBUG
@@ -96,11 +88,11 @@ distribution.
 /* Versioning, past 1.0.14:
 	http://semver.org/
 */
-static const int TIXML2_MAJOR_VERSION = 10;
+static const int TIXML2_MAJOR_VERSION = 11;
 static const int TIXML2_MINOR_VERSION = 0;
 static const int TIXML2_PATCH_VERSION = 0;
 
-#define TINYXML2_MAJOR_VERSION 10
+#define TINYXML2_MAJOR_VERSION 11
 #define TINYXML2_MINOR_VERSION 0
 #define TINYXML2_PATCH_VERSION 0
 
@@ -199,7 +191,7 @@ private:
 	Has a small initial memory pool, so that low or no usage will not
 	cause a call to new/delete
 */
-template <class T, int INITIAL_SIZE>
+template <class T, size_t INITIAL_SIZE>
 class DynArray
 {
 public:
@@ -227,9 +219,8 @@ public:
         ++_size;
     }
 
-    T* PushArr( int count ) {
-        TIXMLASSERT( count >= 0 );
-        TIXMLASSERT( _size <= INT_MAX - count );
+    T* PushArr( size_t count ) {
+        TIXMLASSERT( _size <= SIZE_MAX - count );
         EnsureCapacity( _size+count );
         T* ret = &_mem[_size];
         _size += count;
@@ -242,7 +233,7 @@ public:
         return _mem[_size];
     }
 
-    void PopArr( int count ) {
+    void PopArr( size_t count ) {
         TIXMLASSERT( _size >= count );
         _size -= count;
     }
@@ -251,13 +242,13 @@ public:
         return _size == 0;
     }
 
-    T& operator[](int i)				{
-        TIXMLASSERT( i>= 0 && i < _size );
+    T& operator[](size_t i) {
+        TIXMLASSERT( i < _size );
         return _mem[i];
     }
 
-    const T& operator[](int i) const	{
-        TIXMLASSERT( i>= 0 && i < _size );
+    const T& operator[](size_t i) const {
+        TIXMLASSERT( i < _size );
         return _mem[i];
     }
 
@@ -266,18 +257,17 @@ public:
         return _mem[ _size - 1];
     }
 
-    int Size() const					{
-        TIXMLASSERT( _size >= 0 );
+    size_t Size() const {
         return _size;
     }
 
-    int Capacity() const				{
+    size_t Capacity() const {
         TIXMLASSERT( _allocated >= INITIAL_SIZE );
         return _allocated;
     }
 
-	void SwapRemove(int i) {
-		TIXMLASSERT(i >= 0 && i < _size);
+	void SwapRemove(size_t i) {
+		TIXMLASSERT(i < _size);
 		TIXMLASSERT(_size > 0);
 		_mem[i] = _mem[_size - 1];
 		--_size;
@@ -297,14 +287,14 @@ private:
     DynArray( const DynArray& ); // not supported
     void operator=( const DynArray& ); // not supported
 
-    void EnsureCapacity( int cap ) {
+    void EnsureCapacity( size_t cap ) {
         TIXMLASSERT( cap > 0 );
         if ( cap > _allocated ) {
-            TIXMLASSERT( cap <= INT_MAX / 2 );
-            const int newAllocated = cap * 2;
-            T* newMem = new T[static_cast<unsigned int>(newAllocated)];
+            TIXMLASSERT( cap <= SIZE_MAX / 2 / sizeof(T));
+            const size_t newAllocated = cap * 2;
+            T* newMem = new T[newAllocated];
             TIXMLASSERT( newAllocated >= _size );
-            memcpy( newMem, _mem, sizeof(T)*static_cast<size_t>(_size) );	// warning: not using constructors, only works for PODs
+            memcpy( newMem, _mem, sizeof(T) * _size );	// warning: not using constructors, only works for PODs
             if ( _mem != _pool ) {
                 delete [] _mem;
             }
@@ -314,9 +304,9 @@ private:
     }
 
     T*  _mem;
-    T   _pool[static_cast<size_t>(INITIAL_SIZE)];
-    int _allocated;		// objects allocated
-    int _size;			// number objects in use
+    T   _pool[INITIAL_SIZE];
+    size_t _allocated;		// objects allocated
+    size_t _size;			// number objects in use
 };
 
 
@@ -330,7 +320,7 @@ public:
     MemPool() {}
     virtual ~MemPool() {}
 
-    virtual int ItemSize() const = 0;
+    virtual size_t ItemSize() const = 0;
     virtual void* Alloc() = 0;
     virtual void Free( void* ) = 0;
     virtual void SetTracked() = 0;
@@ -340,7 +330,7 @@ public:
 /*
 	Template child class to create pools of the correct type.
 */
-template< int ITEM_SIZE >
+template< size_t ITEM_SIZE >
 class MemPoolT : public MemPool
 {
 public:
@@ -362,10 +352,10 @@ public:
         _nUntracked = 0;
     }
 
-    virtual int ItemSize() const override{
+    virtual size_t ItemSize() const override {
         return ITEM_SIZE;
     }
-    int CurrentAllocs() const		{
+    size_t CurrentAllocs() const {
         return _currentAllocs;
     }
 
@@ -376,7 +366,7 @@ public:
             _blockPtrs.Push( block );
 
             Item* blockItems = block->items;
-            for( int i = 0; i < ITEMS_PER_BLOCK - 1; ++i ) {
+            for( size_t i = 0; i < ITEMS_PER_BLOCK - 1; ++i ) {
                 blockItems[i].next = &(blockItems[i + 1]);
             }
             blockItems[ITEMS_PER_BLOCK - 1].next = 0;
@@ -417,7 +407,7 @@ public:
         --_nUntracked;
     }
 
-    int Untracked() const {
+    size_t Untracked() const {
         return _nUntracked;
     }
 
@@ -448,10 +438,10 @@ private:
     DynArray< Block*, 10 > _blockPtrs;
     Item* _root;
 
-    int _currentAllocs;
-    int _nAllocs;
-    int _maxAllocs;
-    int _nUntracked;
+    size_t _currentAllocs;
+    size_t _nAllocs;
+    size_t _maxAllocs;
+    size_t _nUntracked;
 };
 
 
@@ -892,7 +882,7 @@ public:
 
 		If the 'target' is null, then the nodes will
 		be allocated in the current document. If 'target'
-        is specified, the memory will be allocated is the
+        is specified, the memory will be allocated in the
         specified XMLDocument.
 
 		NOTE: This is probably not the correct tool to
@@ -1981,11 +1971,11 @@ private:
 	void PushDepth();
 	void PopDepth();
 
-    template<class NodeType, int PoolElementSize>
+    template<class NodeType, size_t PoolElementSize>
     NodeType* CreateUnlinkedNode( MemPoolT<PoolElementSize>& pool );
 };
 
-template<class NodeType, int PoolElementSize>
+template<class NodeType, size_t PoolElementSize>
 inline NodeType* XMLDocument::CreateUnlinkedNode( MemPoolT<PoolElementSize>& pool )
 {
     TIXMLASSERT( sizeof( NodeType ) == PoolElementSize );
@@ -2241,13 +2231,18 @@ private:
 class TINYXML2_LIB XMLPrinter : public XMLVisitor
 {
 public:
+    enum EscapeAposCharsInAttributes {
+        ESCAPE_APOS_CHARS_IN_ATTRIBUTES,
+        DONT_ESCAPE_APOS_CHARS_IN_ATTRIBUTES
+    };
+
     /** Construct the printer. If the FILE* is specified,
     	this will print to the FILE. Else it will print
     	to memory, and the result is available in CStr().
     	If 'compact' is set to true, then output is created
     	with only required whitespace and newlines.
     */
-    XMLPrinter( FILE* file=0, bool compact = false, int depth = 0 );
+    XMLPrinter( FILE* file=0, bool compact = false, int depth = 0, EscapeAposCharsInAttributes aposInAttributes = ESCAPE_APOS_CHARS_IN_ATTRIBUTES );
     virtual ~XMLPrinter()	{}
 
     /** If streaming, write the BOM and declaration. */
@@ -2315,7 +2310,7 @@ public:
     	of the XML file in memory. (Note the size returned
     	includes the terminating null.)
     */
-    int CStrSize() const {
+    size_t CStrSize() const {
         return _buffer.Size();
     }
     /**
@@ -2375,7 +2370,7 @@ private:
 };
 
 
-}	// tinyxml2
+} // namespace tinyxml2
 
 #if defined(_MSC_VER)
 #   pragma warning(pop)

xmltest.cpp

@@ -324,9 +324,9 @@ int main( int argc, const char ** argv )
 
 		printf( "Test file '%s' loaded. ErrorID=%d\n", argv[1], errorID );
 		if ( !errorID ) {
-			printf( "Load time=%u\n",   (unsigned)(loadTime - startTime) );
-			printf( "Delete time=%u\n", (unsigned)(deleteTime - loadTime) );
-			printf( "Total time=%u\n",  (unsigned)(deleteTime - startTime) );
+			printf( "Load time=%u\n",   static_cast<unsigned>(loadTime - startTime) );
+			printf( "Delete time=%u\n", static_cast<unsigned>(deleteTime - loadTime) );
+			printf( "Total time=%u\n",  static_cast<unsigned>(deleteTime - startTime) );
 		}
 		exit(0);
 	}
@@ -595,8 +595,8 @@ int main( int argc, const char ** argv )
 
 		result = ele->QueryDoubleAttribute( "attr0", &dVal );
 		XMLTest( "Query attribute: int as double", XML_SUCCESS, result);
-		XMLTest( "Query attribute: int as double", 1, (int)dVal );
-		XMLTest( "Query attribute: int as double", 1, (int)ele->DoubleAttribute("attr0"));
+		XMLTest( "Query attribute: int as double", 1, static_cast<int>(dVal) );
+		XMLTest( "Query attribute: int as double", 1, static_cast<int>(ele->DoubleAttribute("attr0")));
 
 		result = ele->QueryDoubleAttribute( "attr1", &dVal );
 		XMLTest( "Query attribute: double as double", XML_SUCCESS, result);
@@ -648,17 +648,17 @@ int main( int argc, const char ** argv )
 
 		{
 			XMLError queryResult = ele->QueryAttribute( "int", &iVal2 );
-			XMLTest( "Query int attribute generic", (int)XML_SUCCESS, queryResult);
+			XMLTest( "Query int attribute generic", static_cast<int>(XML_SUCCESS), queryResult);
 		}
 		{
 			XMLError queryResult = ele->QueryAttribute( "double", &dVal2 );
-			XMLTest( "Query double attribute generic", (int)XML_SUCCESS, queryResult);
+			XMLTest( "Query double attribute generic", static_cast<int>(XML_SUCCESS), queryResult);
 		}
 
 		XMLTest( "Attribute match test", "strValue", ele->Attribute( "str", "strValue" ) );
 		XMLTest( "Attribute round trip. c-string.", "strValue", cStr );
 		XMLTest( "Attribute round trip. int.", 1, iVal );
-		XMLTest( "Attribute round trip. double.", -1, (int)dVal );
+		XMLTest( "Attribute round trip. double.", -1, static_cast<int>(dVal) );
 		XMLTest( "Alternate query", true, iVal == iVal2 );
 		XMLTest( "Alternate query", true, dVal == dVal2 );
 		XMLTest( "Alternate query", true, iVal == ele->IntAttribute("int") );
@@ -675,7 +675,7 @@ int main( int argc, const char ** argv )
 		const unsigned char correctValue[] = {	0xd1U, 0x86U, 0xd0U, 0xb5U, 0xd0U, 0xbdU, 0xd0U, 0xbdU,
 												0xd0U, 0xbeU, 0xd1U, 0x81U, 0xd1U, 0x82U, 0xd1U, 0x8cU, 0 };
 
-		XMLTest( "UTF-8: Russian value.", (const char*)correctValue, element->Attribute( "value" ) );
+		XMLTest( "UTF-8: Russian value.", reinterpret_cast<const char*>(correctValue), element->Attribute( "value" ) );
 
 		const unsigned char russianElementName[] = {	0xd0U, 0xa0U, 0xd1U, 0x83U,
 														0xd1U, 0x81U, 0xd1U, 0x81U,
@@ -683,7 +683,7 @@ int main( int argc, const char ** argv )
 														0xd0U, 0xb9U, 0 };
 		const char russianText[] = "<\xD0\xB8\xD0\xBC\xD0\xB5\xD0\xB5\xD1\x82>";
 
-		XMLText* text = doc.FirstChildElement( "document" )->FirstChildElement( (const char*) russianElementName )->FirstChild()->ToText();
+		XMLText* text = doc.FirstChildElement( "document" )->FirstChildElement( reinterpret_cast<const char*>(russianElementName) )->FirstChild()->ToText();
 		XMLTest( "UTF-8: Browsing russian element name.",
 				 russianText,
 				 text->Value() );
@@ -824,7 +824,7 @@ int main( int argc, const char ** argv )
 			{
 				int v = 0;
 				XMLError queryResult = element->QueryAttribute("attrib", &v);
-				XMLTest("Attribute: int", (int)XML_SUCCESS, queryResult, true);
+				XMLTest("Attribute: int", static_cast<int>(XML_SUCCESS), queryResult, true);
 				XMLTest("Attribute: int", -100, v, true);
 			}
 			XMLTest("Attribute: int", -100, element->IntAttribute("attrib"), true);
@@ -840,7 +840,7 @@ int main( int argc, const char ** argv )
 			{
 				unsigned v = 0;
 				XMLError queryResult = element->QueryAttribute("attrib", &v);
-				XMLTest("Attribute: unsigned", (int)XML_SUCCESS, queryResult, true);
+				XMLTest("Attribute: unsigned", static_cast<int>(XML_SUCCESS), queryResult, true);
 				XMLTest("Attribute: unsigned", unsigned(100), v, true);
 			}
 			{
@@ -864,7 +864,7 @@ int main( int argc, const char ** argv )
 			{
 				int64_t v = 0;
 				XMLError queryResult = element->QueryAttribute("attrib", &v);
-				XMLTest("Attribute: int64_t", (int)XML_SUCCESS, queryResult, true);
+				XMLTest("Attribute: int64_t", static_cast<int>(XML_SUCCESS), queryResult, true);
 				XMLTest("Attribute: int64_t", BIG, v, true);
 			}
 			XMLTest("Attribute: int64_t", BIG, element->Int64Attribute("attrib"), true);
@@ -880,7 +880,7 @@ int main( int argc, const char ** argv )
             {
                 uint64_t v = 0;
 				XMLError queryResult = element->QueryAttribute("attrib", &v);
-                XMLTest("Attribute: uint64_t", (int)XML_SUCCESS, queryResult, true);
+                XMLTest("Attribute: uint64_t", static_cast<int>(XML_SUCCESS), queryResult, true);
                 XMLTest("Attribute: uint64_t", BIG_POS, v, true);
             }
             XMLTest("Attribute: uint64_t", BIG_POS, element->Unsigned64Attribute("attrib"), true);
@@ -896,7 +896,7 @@ int main( int argc, const char ** argv )
 			{
 				bool v = false;
 				XMLError queryResult = element->QueryAttribute("attrib", &v);
-				XMLTest("Attribute: bool", (int)XML_SUCCESS, queryResult, true);
+				XMLTest("Attribute: bool", static_cast<int>(XML_SUCCESS), queryResult, true);
 				XMLTest("Attribute: bool", true, v, true);
 			}
 			XMLTest("Attribute: bool", true, element->BoolAttribute("attrib"), true);
@@ -924,7 +924,7 @@ int main( int argc, const char ** argv )
 			{
 				double v = 0;
 				XMLError queryResult = element->QueryAttribute("attrib", &v);
-				XMLTest("Attribute: bool", (int)XML_SUCCESS, queryResult, true);
+				XMLTest("Attribute: bool", static_cast<int>(XML_SUCCESS), queryResult, true);
 				XMLTest("Attribute: double", 100.0, v, true);
 			}
 			XMLTest("Attribute: double", 100.0, element->DoubleAttribute("attrib"), true);
@@ -940,7 +940,7 @@ int main( int argc, const char ** argv )
 			{
 				float v = 0;
 				XMLError queryResult = element->QueryAttribute("attrib", &v);
-				XMLTest("Attribute: float", (int)XML_SUCCESS, queryResult, true);
+				XMLTest("Attribute: float", static_cast<int>(XML_SUCCESS), queryResult, true);
 				XMLTest("Attribute: float", 100.0f, v, true);
 			}
 			XMLTest("Attribute: float", 100.0f, element->FloatAttribute("attrib"), true);
@@ -1005,7 +1005,7 @@ int main( int argc, const char ** argv )
 			FILE* fp1 = fopen("resources/out/printer_1.xml", "w");
 			XMLPrinter printer(fp1);
 
-			printer.PushDeclaration("version = '1.0' enconding = 'utf-8'");
+			printer.PushDeclaration("version = '1.0' encoding = 'utf-8'");
 
 			printer.OpenElement("foo");
 			printer.PushAttribute("attrib-text", "text");
@@ -1068,15 +1068,15 @@ int main( int argc, const char ** argv )
 
 			unsigned  unsigned_value;
 			unsigned_value = root->FirstChildElement("unsigned")->UnsignedText();
-			XMLTest("PushText( unsigned value ) test", (unsigned)12, unsigned_value);
+			XMLTest("PushText( unsigned value ) test", static_cast<unsigned>(12), unsigned_value);
 
 			int64_t  int64_t_value;
 			int64_t_value = root->FirstChildElement("int64_t")->Int64Text();
-			XMLTest("PushText( int64_t value ) test", (int64_t) 13, int64_t_value);
+			XMLTest("PushText( int64_t value ) test", static_cast<int64_t>(13), int64_t_value);
 
 			uint64_t uint64_t_value;
 			uint64_t_value = root->FirstChildElement("uint64_t")->Unsigned64Text();
-			XMLTest("PushText( uint64_t value ) test", (uint64_t) 14, uint64_t_value);
+			XMLTest("PushText( uint64_t value ) test", static_cast<uint64_t>(14), uint64_t_value);
 
 			float  float_value;
 			float_value = root->FirstChildElement("float")->FloatText();
@@ -1096,7 +1096,7 @@ int main( int argc, const char ** argv )
 
 			const XMLDeclaration* declaration = cdoc.FirstChild()->ToDeclaration();
 			const char* declaration_value = declaration->Value();
-			XMLTest("PushDeclaration() test", "version = '1.0' enconding = 'utf-8'", declaration_value);
+			XMLTest("PushDeclaration() test", "version = '1.0' encoding = 'utf-8'", declaration_value);
 		}
 	}
 
@@ -1397,7 +1397,7 @@ int main( int argc, const char ** argv )
 		buf[61] = 0;
 
 		XMLDocument doc;
-		doc.Parse( (const char*)buf);
+		doc.Parse( reinterpret_cast<const char*>(buf) );
 		XMLTest( "Broken CDATA", true, doc.Error() );
 	}
 
@@ -1642,7 +1642,7 @@ int main( int argc, const char ** argv )
 
 		static const char* result  = "\xef\xbb\xbf<?xml version=\"1.0\" encoding=\"UTF-8\"?>";
 		XMLTest( "BOM and default declaration", result, printer.CStr(), false );
-		XMLTest( "CStrSize", 42, printer.CStrSize(), false );
+		XMLTest( "CStrSize", true, printer.CStrSize() == 42, false );
 	}
 	{
 		const char* xml = "<ipxml ws='1'><info bla=' /></ipxml>";
@@ -1753,7 +1753,7 @@ int main( int argc, const char ** argv )
 			unsigned unsignedValue = 0;
 			XMLError queryResult = pointElement->FirstChildElement( "y" )->QueryUnsignedText( &unsignedValue );
 			XMLTest( "QueryUnsignedText result", XML_SUCCESS, queryResult, false );
-			XMLTest( "QueryUnsignedText", (unsigned)1, unsignedValue, false );
+			XMLTest( "QueryUnsignedText", static_cast<unsigned>(1), unsignedValue, false );
 		}
 
 		{
@@ -2025,7 +2025,13 @@ int main( int argc, const char ** argv )
 		XMLTest("Parse nested elements with pedantic whitespace", false, doc.Error());
 		XMLTest("Pedantic whitespace", true, 0 == doc.RootElement()->FirstChildElement()->GetText());
 	}
-
+	//Check the robustness of the DeleteNode function in handling null pointers.
+	{
+		XMLDocument doc;
+		doc.DeleteNode(nullptr);
+		XMLTest("DeleteNode with null pointer", true, doc.Error() == XML_SUCCESS);
+	}
+		
 	// Check sample xml can be parsed with pedantic mode
 	{
 		XMLDocument doc(true, PEDANTIC_WHITESPACE);
@@ -2639,7 +2645,7 @@ int main( int argc, const char ** argv )
             "<?xml version=\"1.0\"?>\n"					// 2 DecL
             "<root>\r\n"								// 3 Element
             "\n"										// 4
-            "text contining new line \n"				// 5 Text
+            "text containing new line \n"				// 5 Text
             " and also containing crlf \r\n"			// 6
             "<sub><![CDATA[\n"							// 7 Element Text
             "cdata containing new line \n"				// 8
@@ -2666,6 +2672,70 @@ int main( int argc, const char ** argv )
     	doc.PrintError();
     }
 
+	// ---------- CVE-2024-50615 -----------
+	{
+		const char* xml = "<Hello value='12&#65;34' value2='56&#x42;78'>Text</Hello>";
+		XMLDocument doc;
+		doc.Parse(xml);
+		const char* value = doc.FirstChildElement()->Attribute("value");
+		const char* value2 = doc.FirstChildElement()->Attribute("value2");
+		XMLTest("Test attribute encode", false, doc.Error());
+		XMLTest("Test decimal value", value, "12A34");
+		XMLTest("Test hex encode", value2, "56B78");
+	}
+
+	{
+		const char* xml = "<Hello value='&#ABC9000000065;' value2='&#xffffffff;' value3='&#5000000000;' value4='&#x00000045;' value5='&#x000000000000000021;'>Text</Hello>";
+		XMLDocument doc;
+		doc.Parse(xml);
+		const char* value = doc.FirstChildElement()->Attribute("value");
+		const char* value2 = doc.FirstChildElement()->Attribute("value2");
+		const char* value3 = doc.FirstChildElement()->Attribute("value3");
+		const char* value4 = doc.FirstChildElement()->Attribute("value4");
+		const char* value5 = doc.FirstChildElement()->Attribute("value5");
+		XMLTest("Test attribute encode", false, doc.Error());
+		XMLTest("Test attribute encode too long value", value, "&#ABC9000000065;"); // test long value
+		XMLTest("Test attribute encode out of unicode range", value2, "&#xffffffff;"); // out of unicode range
+		XMLTest("Test attribute encode out of int max value", value3, "&#5000000000;"); // out of int max value
+		XMLTest("Test attribute encode with a Hex value", value4, "E"); // hex value in unicode value
+		XMLTest("Test attribute encode with a Hex value", value5, "!"); // hex value in unicode value
+	}
+
+	// ---------- XMLPrinter Apos Escaping ------
+	{
+		const char* testText = "text containing a ' character";
+		XMLDocument doc;
+		XMLElement* element = doc.NewElement( "element" );
+		doc.InsertEndChild( element );
+		element->SetAttribute( "attrib", testText );
+		{
+			XMLPrinter defaultPrinter;
+			doc.Print( &defaultPrinter );
+			const char* defaultOutput = defaultPrinter.CStr();
+			const bool foundTextWithUnescapedApos = (strstr(defaultOutput, testText) != nullptr); 
+			XMLTest("Default XMLPrinter should escape ' characters", false, foundTextWithUnescapedApos);
+			{
+				XMLDocument parsingDoc;
+				parsingDoc.Parse(defaultOutput);
+				const XMLAttribute* attrib = parsingDoc.FirstChildElement("element")->FindAttribute("attrib");
+				XMLTest("Default XMLPrinter should output parsable xml", testText, attrib->Value(), true);
+			}
+		}
+		{
+			XMLPrinter customPrinter(0, false, 0, XMLPrinter::DONT_ESCAPE_APOS_CHARS_IN_ATTRIBUTES);
+			doc.Print( &customPrinter );
+			const char* customOutput = customPrinter.CStr();
+			const bool foundTextWithUnescapedApos = (strstr(customOutput, testText) != nullptr); 
+			XMLTest("Custom XMLPrinter should not escape ' characters", true, foundTextWithUnescapedApos);
+			{
+				XMLDocument parsingDoc;
+				parsingDoc.Parse(customOutput);
+				const XMLAttribute* attrib = parsingDoc.FirstChildElement("element")->FindAttribute("attrib");
+				XMLTest("Custom XMLPrinter should output parsable xml", testText, attrib->Value(), true);
+			}
+		}
+	}
+	
     // ----------- Performance tracking --------------
 	{
 #if defined( _MSC_VER )
@@ -2715,9 +2785,9 @@ int main( int argc, const char ** argv )
 #endif
 
 #if defined( _MSC_VER )
-		const double duration = 1000.0 * (double)(end - start) / ((double)freq * (double)COUNT);
+		const double duration = 1000.0 * static_cast<double>(end - start) / (static_cast<double>(freq) * static_cast<double>(COUNT));
 #else
-		const double duration = (double)(cend - cstart) / (double)COUNT;
+		const double duration = static_cast<double>(cend - cstart) / static_cast<double>(COUNT);
 #endif
 		printf("\nParsing dream.xml (%s): %.3f milli-seconds\n", note, duration);
 	}