Merge pull request #539 from libexpat/prevent-more-integer-overflows

[CVE-2022-22822 to CVE-2022-22827] lib: Prevent more integer overflows
diff --git a/.github/workflows/linux.yml b/.github/workflows/linux.yml
index 951ec74..138d05c 100644
--- a/.github/workflows/linux.yml
+++ b/.github/workflows/linux.yml
@@ -71,6 +71,8 @@
           - MODE: qa-sh
             FLAT_ENV: CC=gcc CXX=g++ LD=ld QA_PROCESSOR=gcov CMAKE_ARGS=-D_EXPAT_M32=ON
           - MODE: qa-sh
+            FLAT_ENV: CC=gcc CXX=g++ LD=ld QA_PROCESSOR=gcov CMAKE_ARGS="-D_EXPAT_M32=ON -DEXPAT_ATTR_INFO=ON"
+          - MODE: qa-sh
             FLAT_ENV: CC=i686-w64-mingw32-gcc CXX=i686-w64-mingw32-g++ LD=i686-w64-mingw32-ld QA_PROCESSOR=gcov CMAKE_ARGS="-DCMAKE_SYSTEM_NAME=Windows -DWIN32=ON -DMINGW=ON -DEXPAT_ATTR_INFO=ON"
           - MODE: qa-sh
             FLAT_ENV: CC=i686-w64-mingw32-gcc CXX=i686-w64-mingw32-g++ LD=i686-w64-mingw32-ld QA_PROCESSOR=gcov CMAKE_ARGS="-DCMAKE_SYSTEM_NAME=Windows -DWIN32=ON -DMINGW=ON -DEXPAT_ATTR_INFO=ON -DEXPAT_CHAR_TYPE=wchar_t"
diff --git a/expat/Changes b/expat/Changes
index 98d4f53..d035bad 100644
--- a/expat/Changes
+++ b/expat/Changes
@@ -20,6 +20,16 @@
                     on variable m_groupSize in function doProlog leading
                     to realloc acting as free.
                     Impact is denial of service or more.
+            #539  CVE-2022-22822 to CVE-2022-22827 -- Prevent integer overflows
+                    near memory allocation at multiple places.  Mitre assigned
+                    a dedicated CVE for each involved internal C function:
+                    - CVE-2022-22822 for function addBinding
+                    - CVE-2022-22823 for function build_model
+                    - CVE-2022-22824 for function defineAttribute
+                    - CVE-2022-22825 for function lookup
+                    - CVE-2022-22826 for function nextScaffoldPart
+                    - CVE-2022-22827 for function storeAtts
+                    Impact is denial of service or more.
 
         Other changes:
             #535  CMake: Make call to file(GENERATE [..]) work for CMake <3.19
diff --git a/expat/lib/xmlparse.c b/expat/lib/xmlparse.c
index 8f24312..575e73e 100644
--- a/expat/lib/xmlparse.c
+++ b/expat/lib/xmlparse.c
@@ -3261,13 +3261,38 @@
 
   /* 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) {
@@ -3276,6 +3301,17 @@
     }
     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) {
@@ -3610,9 +3646,31 @@
   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;
@@ -3708,6 +3766,21 @@
   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)
@@ -3720,6 +3793,21 @@
     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) {
@@ -6141,7 +6229,24 @@
       }
     } 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)
@@ -6792,8 +6897,20 @@
     /* 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)
@@ -7143,6 +7260,20 @@
   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)
@@ -7212,8 +7343,26 @@
   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)