Merge branch 'upstream-expat' into update-expat

* upstream-expat:
  expat 2022-01-16 (57c7da69)

7 files changed, 214 insertions, 19 deletions

diff --git a/Utilities/cmexpat/ConfigureChecks.cmake b/Utilities/cmexpat/ConfigureChecks.cmake
index 4da252c332..341bab7b9a 100644
--- a/Utilities/cmexpat/ConfigureChecks.cmake
+++ b/Utilities/cmexpat/ConfigureChecks.cmake
@@ -2,6 +2,7 @@ include(CheckCCompilerFlag)
 include(CheckCSourceCompiles)
 include(CheckIncludeFile)
 include(CheckIncludeFiles)
+include(CheckLibraryExists)
 include(CheckSymbolExists)
 include(TestBigEndian)
 
diff --git a/Utilities/cmexpat/README.md b/Utilities/cmexpat/README.md
index 251dc8a19b..6fdd614871 100644
--- a/Utilities/cmexpat/README.md
+++ b/Utilities/cmexpat/README.md
@@ -5,7 +5,7 @@
 [![Downloads GitHub](https://img.shields.io/github/downloads/libexpat/libexpat/total?label=Downloads%20GitHub)](https://github.com/libexpat/libexpat/releases)
 
 
-# Expat, Release 2.4.1
+# Expat, Release 2.4.3
 
 This is Expat, a C library for parsing XML, started by
 [James Clark](https://en.wikipedia.org/wiki/James_Clark_%28programmer%29) in 1997.
diff --git a/Utilities/cmexpat/lib/expat.h b/Utilities/cmexpat/lib/expat.h
index b7d6d35480..0f021e25de 100644
--- a/Utilities/cmexpat/lib/expat.h
+++ b/Utilities/cmexpat/lib/expat.h
@@ -11,7 +11,7 @@
    Copyright (c) 2000-2005 Fred L. Drake, Jr. <fdrake@users.sourceforge.net>
    Copyright (c) 2001-2002 Greg Stein <gstein@users.sourceforge.net>
    Copyright (c) 2002-2016 Karl Waclawek <karl@waclawek.net>
-   Copyright (c) 2016-2021 Sebastian Pipping <sebastian@pipping.org>
+   Copyright (c) 2016-2022 Sebastian Pipping <sebastian@pipping.org>
    Copyright (c) 2016      Cristian Rodríguez <crrodriguez@opensuse.org>
    Copyright (c) 2016      Thomas Beutlich <tc@tbeu.de>
    Copyright (c) 2017      Rhodri James <rhodri@wildebeest.org.uk>
@@ -1041,7 +1041,7 @@ XML_SetBillionLaughsAttackProtectionActivationThreshold(
 */
 #define XML_MAJOR_VERSION 2
 #define XML_MINOR_VERSION 4
-#define XML_MICRO_VERSION 1
+#define XML_MICRO_VERSION 3
 
 #ifdef __cplusplus
 }
diff --git a/Utilities/cmexpat/lib/xmlparse.c b/Utilities/cmexpat/lib/xmlparse.c
index 5ba56eaea6..b2f5fc6bee 100644
--- a/Utilities/cmexpat/lib/xmlparse.c
+++ b/Utilities/cmexpat/lib/xmlparse.c
@@ -1,4 +1,4 @@
-/* 8539b9040d9d901366a62560a064af7cb99811335784b363abc039c5b0ebc416 (2.4.1+)
+/* 9ca2a2fedc35bcb13ba9a134ba5e173020bc2ff5f5a311abf742cec7da1ff26a (2.4.3+)
                             __  __            _
                          ___\ \/ /_ __   __ _| |_
                         / _ \\  /| '_ \ / _` | __|
@@ -13,7 +13,7 @@
    Copyright (c) 2002-2016 Karl Waclawek <karl@waclawek.net>
    Copyright (c) 2005-2009 Steven Solie <ssolie@users.sourceforge.net>
    Copyright (c) 2016      Eric Rahm <erahm@mozilla.com>
-   Copyright (c) 2016-2021 Sebastian Pipping <sebastian@pipping.org>
+   Copyright (c) 2016-2022 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>
@@ -32,6 +32,7 @@
    Copyright (c) 2019      David Loffredo <loffredo@steptools.com>
    Copyright (c) 2019-2020 Ben Wagner <bungeman@chromium.org>
    Copyright (c) 2019      Vadim Zeitlin <vadim@zeitlins.org>
+   Copyright (c) 2021      Dong-hee Na <donghee.na@python.org>
    Licensed under the MIT license:
 
    Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -54,6 +55,10 @@
    USE OR OTHER DEALINGS IN THE SOFTWARE.
 */
 
+#define XML_BUILDING_EXPAT 1
+
+#include <expat_config.h>
+
 #if ! defined(_GNU_SOURCE)
 #  define _GNU_SOURCE 1 /* syscall prototype */
 #endif
@@ -84,14 +89,10 @@
 #  include <errno.h>
 #endif
 
-#define XML_BUILDING_EXPAT 1
-
 #ifdef _WIN32
 #  include "winconfig.h"
 #endif
 
-#include <expat_config.h>
-
 #include "ascii.h"
 #include "expat.h"
 #include "siphash.h"
@@ -3260,13 +3261,38 @@ storeAtts(XML_Parser parser, const ENCODING *enc, const char *attStr,
 
   /* get the attributes from the tokenizer */
   n = XmlGetAttributes(enc, attStr, parser->m_attsSize, parser->m_atts);
+
+  /* Detect and prevent integer overflow */
+  if (n > INT_MAX - nDefaultAtts) {
+    return XML_ERROR_NO_MEMORY;
+  }
+
   if (n + nDefaultAtts > parser->m_attsSize) {
     int oldAttsSize = parser->m_attsSize;
     ATTRIBUTE *temp;
 #ifdef XML_ATTR_INFO
     XML_AttrInfo *temp2;
 #endif
+
+    /* Detect and prevent integer overflow */
+    if ((nDefaultAtts > INT_MAX - INIT_ATTS_SIZE)
+        || (n > INT_MAX - (nDefaultAtts + INIT_ATTS_SIZE))) {
+      return XML_ERROR_NO_MEMORY;
+    }
+
     parser->m_attsSize = n + nDefaultAtts + INIT_ATTS_SIZE;
+
+    /* Detect and prevent integer overflow.
+     * The preprocessor guard addresses the "always false" warning
+     * 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)) {
+      parser->m_attsSize = oldAttsSize;
+      return XML_ERROR_NO_MEMORY;
+    }
+#endif
+
     temp = (ATTRIBUTE *)REALLOC(parser, (void *)parser->m_atts,
                                 parser->m_attsSize * sizeof(ATTRIBUTE));
     if (temp == NULL) {
@@ -3275,6 +3301,17 @@ storeAtts(XML_Parser parser, const ENCODING *enc, const char *attStr,
     }
     parser->m_atts = temp;
 #ifdef XML_ATTR_INFO
+    /* Detect and prevent integer overflow.
+     * The preprocessor guard addresses the "always false" warning
+     * 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)) {
+      parser->m_attsSize = oldAttsSize;
+      return XML_ERROR_NO_MEMORY;
+    }
+#  endif
+
     temp2 = (XML_AttrInfo *)REALLOC(parser, (void *)parser->m_attInfo,
                                     parser->m_attsSize * sizeof(XML_AttrInfo));
     if (temp2 == NULL) {
@@ -3413,7 +3450,13 @@ storeAtts(XML_Parser parser, const ENCODING *enc, const char *attStr,
   if (nPrefixes) {
     int j; /* hash table index */
     unsigned long version = parser->m_nsAttsVersion;
-    int nsAttsSize = (int)1 << parser->m_nsAttsPower;
+
+    /* Detect and prevent invalid shift */
+    if (parser->m_nsAttsPower >= sizeof(unsigned int) * 8 /* bits per byte */) {
+      return XML_ERROR_NO_MEMORY;
+    }
+
+    unsigned int nsAttsSize = 1u << parser->m_nsAttsPower;
     unsigned char oldNsAttsPower = parser->m_nsAttsPower;
     /* size of hash table must be at least 2 * (# of prefixed attributes) */
     if ((nPrefixes << 1)
@@ -3424,7 +3467,28 @@ storeAtts(XML_Parser parser, const ENCODING *enc, const char *attStr,
         ;
       if (parser->m_nsAttsPower < 3)
         parser->m_nsAttsPower = 3;
-      nsAttsSize = (int)1 << parser->m_nsAttsPower;
+
+      /* Detect and prevent invalid shift */
+      if (parser->m_nsAttsPower >= sizeof(nsAttsSize) * 8 /* bits per byte */) {
+        /* Restore actual size of memory in m_nsAtts */
+        parser->m_nsAttsPower = oldNsAttsPower;
+        return XML_ERROR_NO_MEMORY;
+      }
+
+      nsAttsSize = 1u << parser->m_nsAttsPower;
+
+      /* Detect and prevent integer overflow.
+       * The preprocessor guard addresses the "always false" warning
+       * 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)) {
+        /* Restore actual size of memory in m_nsAtts */
+        parser->m_nsAttsPower = oldNsAttsPower;
+        return XML_ERROR_NO_MEMORY;
+      }
+#endif
+
       temp = (NS_ATT *)REALLOC(parser, parser->m_nsAtts,
                                nsAttsSize * sizeof(NS_ATT));
       if (! temp) {
@@ -3582,9 +3646,31 @@ storeAtts(XML_Parser parser, const ENCODING *enc, const char *attStr,
   tagNamePtr->prefixLen = prefixLen;
   for (i = 0; localPart[i++];)
     ; /* i includes null terminator */
+
+  /* Detect and prevent integer overflow */
+  if (binding->uriLen > INT_MAX - prefixLen
+      || i > INT_MAX - (binding->uriLen + prefixLen)) {
+    return XML_ERROR_NO_MEMORY;
+  }
+
   n = i + binding->uriLen + prefixLen;
   if (n > binding->uriAlloc) {
     TAG *p;
+
+    /* Detect and prevent integer overflow */
+    if (n > INT_MAX - EXPAND_SPARE) {
+      return XML_ERROR_NO_MEMORY;
+    }
+    /* Detect and prevent integer overflow.
+     * The preprocessor guard addresses the "always false" warning
+     * 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)) {
+      return XML_ERROR_NO_MEMORY;
+    }
+#endif
+
     uri = (XML_Char *)MALLOC(parser, (n + EXPAND_SPARE) * sizeof(XML_Char));
     if (! uri)
       return XML_ERROR_NO_MEMORY;
@@ -3680,6 +3766,21 @@ addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId,
   if (parser->m_freeBindingList) {
     b = parser->m_freeBindingList;
     if (len > b->uriAlloc) {
+      /* Detect and prevent integer overflow */
+      if (len > INT_MAX - EXPAND_SPARE) {
+        return XML_ERROR_NO_MEMORY;
+      }
+
+      /* Detect and prevent integer overflow.
+       * The preprocessor guard addresses the "always false" warning
+       * 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)) {
+        return XML_ERROR_NO_MEMORY;
+      }
+#endif
+
       XML_Char *temp = (XML_Char *)REALLOC(
           parser, b->uri, sizeof(XML_Char) * (len + EXPAND_SPARE));
       if (temp == NULL)
@@ -3692,6 +3793,21 @@ addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId,
     b = (BINDING *)MALLOC(parser, sizeof(BINDING));
     if (! b)
       return XML_ERROR_NO_MEMORY;
+
+    /* Detect and prevent integer overflow */
+    if (len > INT_MAX - EXPAND_SPARE) {
+      return XML_ERROR_NO_MEMORY;
+    }
+    /* Detect and prevent integer overflow.
+     * The preprocessor guard addresses the "always false" warning
+     * 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)) {
+      return XML_ERROR_NO_MEMORY;
+    }
+#endif
+
     b->uri
         = (XML_Char *)MALLOC(parser, sizeof(XML_Char) * (len + EXPAND_SPARE));
     if (! b->uri) {
@@ -5018,6 +5134,11 @@ doProlog(XML_Parser parser, const ENCODING *enc, const char *s, const char *end,
       if (parser->m_prologState.level >= parser->m_groupSize) {
         if (parser->m_groupSize) {
           {
+            /* Detect and prevent integer overflow */
+            if (parser->m_groupSize > (unsigned int)(-1) / 2u) {
+              return XML_ERROR_NO_MEMORY;
+            }
+
             char *const new_connector = (char *)REALLOC(
                 parser, parser->m_groupConnector, parser->m_groupSize *= 2);
             if (new_connector == NULL) {
@@ -5028,6 +5149,16 @@ doProlog(XML_Parser parser, const ENCODING *enc, const char *s, const char *end,
           }
 
           if (dtd->scaffIndex) {
+            /* Detect and prevent integer overflow.
+             * The preprocessor guard addresses the "always false" warning
+             * 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)) {
+              return XML_ERROR_NO_MEMORY;
+            }
+#endif
+
             int *const new_scaff_index = (int *)REALLOC(
                 parser, dtd->scaffIndex, parser->m_groupSize * sizeof(int));
             if (new_scaff_index == NULL)
@@ -6098,7 +6229,24 @@ defineAttribute(ELEMENT_TYPE *type, ATTRIBUTE_ID *attId, XML_Bool isCdata,
       }
     } else {
       DEFAULT_ATTRIBUTE *temp;
+
+      /* Detect and prevent integer overflow */
+      if (type->allocDefaultAtts > INT_MAX / 2) {
+        return 0;
+      }
+
       int count = type->allocDefaultAtts * 2;
+
+      /* Detect and prevent integer overflow.
+       * The preprocessor guard addresses the "always false" warning
+       * 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)) {
+        return 0;
+      }
+#endif
+
       temp = (DEFAULT_ATTRIBUTE *)REALLOC(parser, type->defaultAtts,
                                           (count * sizeof(DEFAULT_ATTRIBUTE)));
       if (temp == NULL)
@@ -6749,8 +6897,20 @@ lookup(XML_Parser parser, HASH_TABLE *table, KEY name, size_t createSize) {
     /* check for overflow (table is half full) */
     if (table->used >> (table->power - 1)) {
       unsigned char newPower = table->power + 1;
+
+      /* Detect and prevent invalid shift */
+      if (newPower >= sizeof(unsigned long) * 8 /* bits per byte */) {
+        return NULL;
+      }
+
       size_t newSize = (size_t)1 << newPower;
       unsigned long newMask = (unsigned long)newSize - 1;
+
+      /* Detect and prevent integer overflow */
+      if (newSize > (size_t)(-1) / sizeof(NAMED *)) {
+        return NULL;
+      }
+
       size_t tsize = newSize * sizeof(NAMED *);
       NAMED **newV = (NAMED **)table->mem->malloc_fcn(tsize);
       if (! newV)
@@ -7100,6 +7260,20 @@ nextScaffoldPart(XML_Parser parser) {
   if (dtd->scaffCount >= dtd->scaffSize) {
     CONTENT_SCAFFOLD *temp;
     if (dtd->scaffold) {
+      /* Detect and prevent integer overflow */
+      if (dtd->scaffSize > UINT_MAX / 2u) {
+        return -1;
+      }
+      /* Detect and prevent integer overflow.
+       * The preprocessor guard addresses the "always false" warning
+       * 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)) {
+        return -1;
+      }
+#endif
+
       temp = (CONTENT_SCAFFOLD *)REALLOC(
           parser, dtd->scaffold, dtd->scaffSize * 2 * sizeof(CONTENT_SCAFFOLD));
       if (temp == NULL)
@@ -7169,8 +7343,26 @@ build_model(XML_Parser parser) {
   XML_Content *ret;
   XML_Content *cpos;
   XML_Char *str;
-  int allocsize = (dtd->scaffCount * sizeof(XML_Content)
-                   + (dtd->contentStringLen * sizeof(XML_Char)));
+
+  /* Detect and prevent integer overflow.
+   * The preprocessor guard addresses the "always false" warning
+   * 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)) {
+    return NULL;
+  }
+  if (dtd->contentStringLen > (size_t)(-1) / sizeof(XML_Char)) {
+    return NULL;
+  }
+#endif
+  if (dtd->scaffCount * sizeof(XML_Content)
+      > (size_t)(-1) - dtd->contentStringLen * sizeof(XML_Char)) {
+    return NULL;
+  }
+
+  const size_t allocsize = (dtd->scaffCount * sizeof(XML_Content)
+                            + (dtd->contentStringLen * sizeof(XML_Char)));
 
   ret = (XML_Content *)MALLOC(parser, allocsize);
   if (! ret)
diff --git a/Utilities/cmexpat/lib/xmlrole.c b/Utilities/cmexpat/lib/xmlrole.c
index 08173b0fd5..77746ee42d 100644
--- a/Utilities/cmexpat/lib/xmlrole.c
+++ b/Utilities/cmexpat/lib/xmlrole.c
@@ -15,6 +15,7 @@
    Copyright (c) 2016-2021 Sebastian Pipping <sebastian@pipping.org>
    Copyright (c) 2017      Rhodri James <rhodri@wildebeest.org.uk>
    Copyright (c) 2019      David Loffredo <loffredo@steptools.com>
+   Copyright (c) 2021      Dong-hee Na <donghee.na@python.org>
    Licensed under the MIT license:
 
    Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -37,14 +38,14 @@
    USE OR OTHER DEALINGS IN THE SOFTWARE.
 */
 
+#include <expat_config.h>
+
 #include <stddef.h>
 
 #ifdef _WIN32
 #  include "winconfig.h"
 #endif
 
-#include <expat_config.h>
-
 #include "expat_external.h"
 #include "internal.h"
 #include "xmlrole.h"
diff --git a/Utilities/cmexpat/lib/xmltok.c b/Utilities/cmexpat/lib/xmltok.c
index f2b6b40606..502ca1adc3 100644
--- a/Utilities/cmexpat/lib/xmltok.c
+++ b/Utilities/cmexpat/lib/xmltok.c
@@ -20,6 +20,7 @@
    Copyright (c) 2017      Benbuck Nason <bnason@netflix.com>
    Copyright (c) 2017      José Gutiérrez de la Concha <jose@zeroc.com>
    Copyright (c) 2019      David Loffredo <loffredo@steptools.com>
+   Copyright (c) 2021      Dong-hee Na <donghee.na@python.org>
    Licensed under the MIT license:
 
    Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -42,6 +43,8 @@
    USE OR OTHER DEALINGS IN THE SOFTWARE.
 */
 
+#include <expat_config.h>
+
 #include <stddef.h>
 #include <string.h> /* memcpy */
 #include <stdbool.h>
@@ -50,8 +53,6 @@
 #  include "winconfig.h"
 #endif
 
-#include <expat_config.h>
-
 #include "expat_external.h"
 #include "internal.h"
 #include "xmltok.h"
diff --git a/Utilities/cmexpat/lib/xmltok_ns.c b/Utilities/cmexpat/lib/xmltok_ns.c
index 5fd8392235..fbdd3e3c7b 100644
--- a/Utilities/cmexpat/lib/xmltok_ns.c
+++ b/Utilities/cmexpat/lib/xmltok_ns.c
@@ -11,7 +11,7 @@
    Copyright (c) 2002      Greg Stein <gstein@users.sourceforge.net>
    Copyright (c) 2002      Fred L. Drake, Jr. <fdrake@users.sourceforge.net>
    Copyright (c) 2002-2006 Karl Waclawek <karl@waclawek.net>
-   Copyright (c) 2017      Sebastian Pipping <sebastian@pipping.org>
+   Copyright (c) 2017-2021 Sebastian Pipping <sebastian@pipping.org>
    Licensed under the MIT license:
 
    Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -93,7 +93,7 @@ NS(XmlInitEncoding)(INIT_ENCODING *p, const ENCODING **encPtr,
 static const ENCODING *
 NS(findEncoding)(const ENCODING *enc, const char *ptr, const char *end) {
 #  define ENCODING_MAX 128
-  char buf[ENCODING_MAX];
+  char buf[ENCODING_MAX] = "";
   char *p = buf;
   int i;
   XmlUtf8Convert(enc, &ptr, end, &p, p + ENCODING_MAX - 1);