-/* 28bcd8b1ba7eb595d82822908257fd9c3589b4243e3c922d0369f35bfcd7b506 (2.7.3+)
+/* fab937ab8b186d7d296013669c332e6dfce2f99567882cff1f8eb24223c524a7 (2.7.4+)
__ __ _
___\ \/ /_ __ __ _| |_
/ _ \\ /| '_ \ / _` | __|
Copyright (c) 2002-2016 Karl Waclawek <karl@waclawek.net>
Copyright (c) 2005-2009 Steven Solie <steven@solie.ca>
Copyright (c) 2016 Eric Rahm <erahm@mozilla.com>
- Copyright (c) 2016-2025 Sebastian Pipping <sebastian@pipping.org>
+ Copyright (c) 2016-2026 Sebastian Pipping <sebastian@pipping.org>
Copyright (c) 2016 Gaurav <g.gupta@samsung.com>
Copyright (c) 2016 Thomas Beutlich <tc@tbeu.de>
Copyright (c) 2016 Gustavo Grieco <gustavo.grieco@imag.fr>
Copyright (c) 2024-2025 Berkay Eren Ürün <berkay.ueruen@siemens.com>
Copyright (c) 2024 Hanno Böck <hanno@gentoo.org>
Copyright (c) 2025 Matthew Fernandez <matthew.fernandez@gmail.com>
+ Copyright (c) 2025 Atrem Borovik <polzovatellllk@gmail.com>
+ Copyright (c) 2025 Alfonso Gregory <gfunni234@gmail.com>
+ Copyright (c) 2026 Rosen Penev <rosenp@gmail.com>
Licensed under the MIT license:
Permission is hereby granted, free of charge, to any person obtaining
#include <limits.h> /* INT_MAX, UINT_MAX */
#include <stdio.h> /* fprintf */
#include <stdlib.h> /* getenv, rand_s */
-#include <stdint.h> /* uintptr_t */
+#include <stdint.h> /* SIZE_MAX, uintptr_t */
#include <math.h> /* isnan */
#ifdef _WIN32
# endif /* defined(GRND_NONBLOCK) */
#endif /* defined(HAVE_GETRANDOM) || defined(HAVE_SYSCALL_GETRANDOM) */
-#if defined(HAVE_LIBBSD) \
- && (defined(HAVE_ARC4RANDOM_BUF) || defined(HAVE_ARC4RANDOM))
-# include <bsd/stdlib.h>
-#endif
-
#if defined(_WIN32) && ! defined(LOAD_LIBRARY_SEARCH_SYSTEM32)
# define LOAD_LIBRARY_SEARCH_SYSTEM32 0x00000800
#endif
* Linux >=3.17 + glibc (including <2.25) (syscall SYS_getrandom): HAVE_SYSCALL_GETRANDOM, \
* BSD / macOS >=10.7 / glibc >=2.36 (arc4random_buf): HAVE_ARC4RANDOM_BUF, \
* BSD / macOS (including <10.7) / glibc >=2.36 (arc4random): HAVE_ARC4RANDOM, \
- * libbsd (arc4random_buf): HAVE_ARC4RANDOM_BUF + HAVE_LIBBSD, \
- * libbsd (arc4random): HAVE_ARC4RANDOM + HAVE_LIBBSD, \
* Linux (including <3.17) / BSD / macOS (including <10.7) / Solaris >=8 (/dev/urandom): XML_DEV_URANDOM, \
* Windows >=Vista (rand_s): _WIN32. \
\
const char *rawName; /* tagName in the original encoding */
int rawNameLength;
TAG_NAME name; /* tagName in the API encoding */
- char *buf; /* buffer for name components */
- char *bufEnd; /* end of the buffer */
+ union {
+ char *raw; /* for byte-level access (rawName storage) */
+ XML_Char *str; /* for character-level access (converted name) */
+ } buf; /* buffer for name components */
+ char *bufEnd; /* end of the buffer */
BINDING *bindings;
} TAG;
typedef struct block {
struct block *next;
int size;
- XML_Char s[1];
+ XML_Char s[];
} BLOCK;
typedef struct {
# endif /* ! defined(_WIN32) && defined(XML_DEV_URANDOM) */
/* .. and self-made low quality for backup: */
+ entropy = gather_time_entropy();
+# if ! defined(__wasi__)
/* Process ID is 0 bits entropy if attacker has local access */
- entropy = gather_time_entropy() ^ getpid();
+ entropy ^= getpid();
+# endif
/* Factors are 2^31-1 and 2^61-1 (Mersenne primes M31 and M61) */
if (sizeof(unsigned long) == 4) {
XML_ExternalEntityRefHandler oldExternalEntityRefHandler;
XML_SkippedEntityHandler oldSkippedEntityHandler;
XML_UnknownEncodingHandler oldUnknownEncodingHandler;
+ void *oldUnknownEncodingHandlerData;
XML_ElementDeclHandler oldElementDeclHandler;
XML_AttlistDeclHandler oldAttlistDeclHandler;
XML_EntityDeclHandler oldEntityDeclHandler;
oldExternalEntityRefHandler = parser->m_externalEntityRefHandler;
oldSkippedEntityHandler = parser->m_skippedEntityHandler;
oldUnknownEncodingHandler = parser->m_unknownEncodingHandler;
+ oldUnknownEncodingHandlerData = parser->m_unknownEncodingHandlerData;
oldElementDeclHandler = parser->m_elementDeclHandler;
oldAttlistDeclHandler = parser->m_attlistDeclHandler;
oldEntityDeclHandler = parser->m_entityDeclHandler;
parser->m_externalEntityRefHandler = oldExternalEntityRefHandler;
parser->m_skippedEntityHandler = oldSkippedEntityHandler;
parser->m_unknownEncodingHandler = oldUnknownEncodingHandler;
+ parser->m_unknownEncodingHandlerData = oldUnknownEncodingHandlerData;
parser->m_elementDeclHandler = oldElementDeclHandler;
parser->m_attlistDeclHandler = oldAttlistDeclHandler;
parser->m_entityDeclHandler = oldEntityDeclHandler;
}
p = tagList;
tagList = tagList->parent;
- FREE(parser, p->buf);
+ FREE(parser, p->buf.raw);
destroyBindings(p->bindings, parser);
FREE(parser, p);
}
// NOTE: We are avoiding MALLOC(..) here to leave limiting
// the input size to the application using Expat.
newBuf = parser->m_mem.malloc_fcn(bufferSize);
- if (newBuf == 0) {
+ if (newBuf == NULL) {
parser->m_errorCode = XML_ERROR_NO_MEMORY;
return NULL;
}
size_t bufSize;
size_t nameLen = sizeof(XML_Char) * (tag->name.strLen + 1);
size_t rawNameLen;
- char *rawNameBuf = tag->buf + nameLen;
+ char *rawNameBuf = tag->buf.raw + nameLen;
/* Stop if already stored. Since m_tagStack is a stack, we can stop
at the first entry that has already been copied; everything
below it in the stack is already been accounted for in a
if (rawNameLen > (size_t)INT_MAX - nameLen)
return XML_FALSE;
bufSize = nameLen + rawNameLen;
- if (bufSize > (size_t)(tag->bufEnd - tag->buf)) {
- char *temp = REALLOC(parser, tag->buf, bufSize);
+ if (bufSize > (size_t)(tag->bufEnd - tag->buf.raw)) {
+ char *temp = REALLOC(parser, tag->buf.raw, bufSize);
if (temp == NULL)
return XML_FALSE;
- /* if tag->name.str points to tag->buf (only when namespace
+ /* if tag->name.str points to tag->buf.str (only when namespace
processing is off) then we have to update it
*/
- if (tag->name.str == (XML_Char *)tag->buf)
+ if (tag->name.str == tag->buf.str)
tag->name.str = (XML_Char *)temp;
/* if tag->name.localPart is set (when namespace processing is on)
then update it as well, since it will always point into tag->buf
*/
if (tag->name.localPart)
tag->name.localPart
- = (XML_Char *)temp + (tag->name.localPart - (XML_Char *)tag->buf);
- tag->buf = temp;
+ = (XML_Char *)temp + (tag->name.localPart - tag->buf.str);
+ tag->buf.raw = temp;
tag->bufEnd = temp + bufSize;
rawNameBuf = temp + nameLen;
}
tag = MALLOC(parser, sizeof(TAG));
if (! tag)
return XML_ERROR_NO_MEMORY;
- tag->buf = MALLOC(parser, INIT_TAG_BUF_SIZE);
- if (! tag->buf) {
+ tag->buf.raw = MALLOC(parser, INIT_TAG_BUF_SIZE);
+ if (! tag->buf.raw) {
FREE(parser, tag);
return XML_ERROR_NO_MEMORY;
}
- tag->bufEnd = tag->buf + INIT_TAG_BUF_SIZE;
+ tag->bufEnd = tag->buf.raw + INIT_TAG_BUF_SIZE;
}
tag->bindings = NULL;
tag->parent = parser->m_tagStack;
{
const char *rawNameEnd = tag->rawName + tag->rawNameLength;
const char *fromPtr = tag->rawName;
- toPtr = (XML_Char *)tag->buf;
+ toPtr = tag->buf.str;
for (;;) {
- int bufSize;
int convLen;
const enum XML_Convert_Result convert_res
= XmlConvert(enc, &fromPtr, rawNameEnd, (ICHAR **)&toPtr,
(ICHAR *)tag->bufEnd - 1);
- convLen = (int)(toPtr - (XML_Char *)tag->buf);
+ convLen = (int)(toPtr - tag->buf.str);
if ((fromPtr >= rawNameEnd)
|| (convert_res == XML_CONVERT_INPUT_INCOMPLETE)) {
tag->name.strLen = convLen;
break;
}
- bufSize = (int)(tag->bufEnd - tag->buf) << 1;
+ if (SIZE_MAX / 2 < (size_t)(tag->bufEnd - tag->buf.raw))
+ return XML_ERROR_NO_MEMORY;
+ const size_t bufSize = (size_t)(tag->bufEnd - tag->buf.raw) * 2;
{
- char *temp = REALLOC(parser, tag->buf, bufSize);
+ char *temp = REALLOC(parser, tag->buf.raw, bufSize);
if (temp == NULL)
return XML_ERROR_NO_MEMORY;
- tag->buf = temp;
+ tag->buf.raw = temp;
tag->bufEnd = temp + bufSize;
toPtr = (XML_Char *)temp + convLen;
}
}
}
- tag->name.str = (XML_Char *)tag->buf;
+ tag->name.str = tag->buf.str;
*toPtr = XML_T('\0');
result
= storeAtts(parser, enc, s, &(tag->name), &(tag->bindings), account);
* from -Wtype-limits on platforms where
* sizeof(unsigned int) < sizeof(size_t), e.g. on x86_64. */
#if UINT_MAX >= SIZE_MAX
- if ((unsigned)parser->m_attsSize > (size_t)(-1) / sizeof(ATTRIBUTE)) {
+ if ((unsigned)parser->m_attsSize > SIZE_MAX / sizeof(ATTRIBUTE)) {
parser->m_attsSize = oldAttsSize;
return XML_ERROR_NO_MEMORY;
}
* from -Wtype-limits on platforms where
* sizeof(unsigned int) < sizeof(size_t), e.g. on x86_64. */
# if UINT_MAX >= SIZE_MAX
- if ((unsigned)parser->m_attsSize > (size_t)(-1) / sizeof(XML_AttrInfo)) {
+ if ((unsigned)parser->m_attsSize > SIZE_MAX / sizeof(XML_AttrInfo)) {
parser->m_attsSize = oldAttsSize;
return XML_ERROR_NO_MEMORY;
}
* from -Wtype-limits on platforms where
* sizeof(unsigned int) < sizeof(size_t), e.g. on x86_64. */
#if UINT_MAX >= SIZE_MAX
- if (nsAttsSize > (size_t)(-1) / sizeof(NS_ATT)) {
+ if (nsAttsSize > SIZE_MAX / sizeof(NS_ATT)) {
/* Restore actual size of memory in m_nsAtts */
parser->m_nsAttsPower = oldNsAttsPower;
return XML_ERROR_NO_MEMORY;
* from -Wtype-limits on platforms where
* sizeof(unsigned int) < sizeof(size_t), e.g. on x86_64. */
#if UINT_MAX >= SIZE_MAX
- if ((unsigned)(n + EXPAND_SPARE) > (size_t)(-1) / sizeof(XML_Char)) {
+ if ((unsigned)(n + EXPAND_SPARE) > SIZE_MAX / sizeof(XML_Char)) {
return XML_ERROR_NO_MEMORY;
}
#endif
* from -Wtype-limits on platforms where
* sizeof(unsigned int) < sizeof(size_t), e.g. on x86_64. */
#if UINT_MAX >= SIZE_MAX
- if ((unsigned)(len + EXPAND_SPARE) > (size_t)(-1) / sizeof(XML_Char)) {
+ if ((unsigned)(len + EXPAND_SPARE) > SIZE_MAX / sizeof(XML_Char)) {
return XML_ERROR_NO_MEMORY;
}
#endif
* from -Wtype-limits on platforms where
* sizeof(unsigned int) < sizeof(size_t), e.g. on x86_64. */
#if UINT_MAX >= SIZE_MAX
- if ((unsigned)(len + EXPAND_SPARE) > (size_t)(-1) / sizeof(XML_Char)) {
+ if ((unsigned)(len + EXPAND_SPARE) > SIZE_MAX / sizeof(XML_Char)) {
return XML_ERROR_NO_MEMORY;
}
#endif
* from -Wtype-limits on platforms where
* sizeof(unsigned int) < sizeof(size_t), e.g. on x86_64. */
#if UINT_MAX >= SIZE_MAX
- if (parser->m_groupSize > (size_t)(-1) / sizeof(int)) {
+ if (parser->m_groupSize > SIZE_MAX / sizeof(int)) {
+ parser->m_groupSize /= 2;
return XML_ERROR_NO_MEMORY;
}
#endif
int *const new_scaff_index = REALLOC(
parser, dtd->scaffIndex, parser->m_groupSize * sizeof(int));
- if (new_scaff_index == NULL)
+ if (new_scaff_index == NULL) {
+ parser->m_groupSize /= 2;
return XML_ERROR_NO_MEMORY;
+ }
dtd->scaffIndex = new_scaff_index;
}
} else {
* from -Wtype-limits on platforms where
* sizeof(unsigned int) < sizeof(size_t), e.g. on x86_64. */
#if UINT_MAX >= SIZE_MAX
- if ((unsigned)count > (size_t)(-1) / sizeof(DEFAULT_ATTRIBUTE)) {
+ if ((unsigned)count > SIZE_MAX / sizeof(DEFAULT_ATTRIBUTE)) {
return 0;
}
#endif
* from -Wtype-limits on platforms where
* sizeof(int) < sizeof(size_t), e.g. on x86_64. */
#if UINT_MAX >= SIZE_MAX
- if ((size_t)oldE->nDefaultAtts
- > ((size_t)(-1) / sizeof(DEFAULT_ATTRIBUTE))) {
+ if ((size_t)oldE->nDefaultAtts > SIZE_MAX / sizeof(DEFAULT_ATTRIBUTE)) {
return 0;
}
#endif
unsigned long newMask = (unsigned long)newSize - 1;
/* Detect and prevent integer overflow */
- if (newSize > (size_t)(-1) / sizeof(NAMED *)) {
+ if (newSize > SIZE_MAX / sizeof(NAMED *)) {
return NULL;
}
static XML_Bool FASTCALL
poolGrow(STRING_POOL *pool) {
if (pool->freeBlocks) {
- if (pool->start == 0) {
+ if (pool->start == NULL) {
pool->blocks = pool->freeBlocks;
pool->freeBlocks = pool->freeBlocks->next;
pool->blocks->next = NULL;
* from -Wtype-limits on platforms where
* sizeof(unsigned int) < sizeof(size_t), e.g. on x86_64. */
#if UINT_MAX >= SIZE_MAX
- if (parser->m_groupSize > ((size_t)(-1) / sizeof(int))) {
+ if (parser->m_groupSize > SIZE_MAX / sizeof(int)) {
return -1;
}
#endif
* from -Wtype-limits on platforms where
* sizeof(unsigned int) < sizeof(size_t), e.g. on x86_64. */
#if UINT_MAX >= SIZE_MAX
- if (dtd->scaffSize > (size_t)(-1) / 2u / sizeof(CONTENT_SCAFFOLD)) {
+ if (dtd->scaffSize > SIZE_MAX / 2u / sizeof(CONTENT_SCAFFOLD)) {
return -1;
}
#endif
* from -Wtype-limits on platforms where
* sizeof(unsigned int) < sizeof(size_t), e.g. on x86_64. */
#if UINT_MAX >= SIZE_MAX
- if (dtd->scaffCount > (size_t)(-1) / sizeof(XML_Content)) {
+ if (dtd->scaffCount > SIZE_MAX / sizeof(XML_Content)) {
return NULL;
}
- if (dtd->contentStringLen > (size_t)(-1) / sizeof(XML_Char)) {
+ if (dtd->contentStringLen > SIZE_MAX / sizeof(XML_Char)) {
return NULL;
}
#endif
if (dtd->scaffCount * sizeof(XML_Content)
- > (size_t)(-1) - dtd->contentStringLen * sizeof(XML_Char)) {
+ > SIZE_MAX - dtd->contentStringLen * sizeof(XML_Char)) {
return NULL;
}