gnu: expat: Fix CVE-2016-0718. Improve fix for CVE-2015-1283.

* gnu/packages/patches/expat-CVE-2015-1283-refix.patch,
gnu/packages/patches/expat-CVE-2016-0718.patch: New files.
* gnu/local.mk (dist_patch_DATA): Add them.
* gnu/packages/xml.scm (expat)[replacement]: New field.
(expat/fixed): New variable.
[source]: Use new patches.

2 files changed, 803 insertions, 0 deletions

diff --git a/gnu/packages/patches/expat-CVE-2015-1283-refix.patch b/gnu/packages/patches/expat-CVE-2015-1283-refix.patch
new file mode 100644
index 0000000000..af5e3bcc3e
--- /dev/null
+++ b/gnu/packages/patches/expat-CVE-2015-1283-refix.patch
@@ -0,0 +1,42 @@
+Update previous fix for CVE-2015-1283 to not rely on undefined behavior.
+
+Copied from Debian, as found in Debian package version 2.1.0-6+deb8u2.
+
+https://sources.debian.net/src/expat/2.1.0-6%2Bdeb8u2/debian/patches/CVE-2015-1283-refix.patch/
+
+From 29a11774d8ebbafe8418b4a5ffb4cc1160b194a1 Mon Sep 17 00:00:00 2001
+From: Pascal Cuoq <cuoq@trust-in-soft.com>
+Date: Sun, 15 May 2016 09:05:46 +0200
+Subject: [PATCH] Avoid relying on undefined behavior in CVE-2015-1283 fix.
+
+---
+ expat/lib/xmlparse.c | 6 ++++--
+ 1 file changed, 4 insertions(+), 2 deletions(-)
+
+diff --git a/lib/xmlparse.c b/lib/xmlparse.c
+index 13e080d..cdb12ef 100644
+--- a/lib/xmlparse.c
++++ b/lib/xmlparse.c
+@@ -1695,7 +1695,8 @@ XML_GetBuffer(XML_Parser parser, int len
+   }
+ 
+   if (len > bufferLim - bufferEnd) {
+-    int neededSize = len + (int)(bufferEnd - bufferPtr);
++    /* Do not invoke signed arithmetic overflow: */
++    int neededSize = (int) ((unsigned)len + (unsigned)(bufferEnd - bufferPtr));
+ /* BEGIN MOZILLA CHANGE (sanity check neededSize) */
+     if (neededSize < 0) {
+       errorCode = XML_ERROR_NO_MEMORY;
+@@ -1729,7 +1730,8 @@ XML_GetBuffer(XML_Parser parser, int len
+       if (bufferSize == 0)
+         bufferSize = INIT_BUFFER_SIZE;
+       do {
+-        bufferSize *= 2;
++        /* Do not invoke signed arithmetic overflow: */
++        bufferSize = (int) (2U * (unsigned) bufferSize);
+ /* BEGIN MOZILLA CHANGE (prevent infinite loop on overflow) */
+       } while (bufferSize < neededSize && bufferSize > 0);
+ /* END MOZILLA CHANGE */
+-- 
+2.8.2
+
diff --git a/gnu/packages/patches/expat-CVE-2016-0718.patch b/gnu/packages/patches/expat-CVE-2016-0718.patch
new file mode 100644
index 0000000000..22436c20cc
--- /dev/null
+++ b/gnu/packages/patches/expat-CVE-2016-0718.patch
@@ -0,0 +1,761 @@
+Fix CVE-2016-0718.
+
+Copied from Debian, as found in Debian package version 2.1.0-6+deb8u2.
+
+https://sources.debian.net/src/expat/2.1.0-6%2Bdeb8u2/debian/patches/CVE-2016-0718-v2-2-1.patch/
+
+From cdfcb1b5c95e93b00ae9e9d25708b4a3bee72c15 Mon Sep 17 00:00:00 2001
+From: Sebastian Pipping <sebastian@pipping.org>
+Date: Mon, 2 May 2016 00:02:44 +0200
+Subject: [PATCH] Address CVE-2016-0718 (/patch/ version 2.2.1)
+
+* Out of bounds memory access when doing text conversion on malformed input
+* Integer overflow related to memory allocation
+
+Reported by Gustavo Grieco
+
+Patch credits go to
+* Christian Heimes
+* Karl Waclawek
+* Gustavo Grieco
+* Sebastian Pipping
+* Pascal Cuoq
+---
+ expat/lib/xmlparse.c    |  34 +++++++++-----
+ expat/lib/xmltok.c      | 115 +++++++++++++++++++++++++++++++++++-------------
+ expat/lib/xmltok.h      |  10 ++++-
+ expat/lib/xmltok_impl.c |  62 +++++++++++++-------------
+ 4 files changed, 146 insertions(+), 75 deletions(-)
+
+diff --git a/lib/xmlparse.c b/lib/xmlparse.c
+index e308c79..13e080d 100644
+--- a/lib/xmlparse.c
++++ b/lib/xmlparse.c
+@@ -2436,11 +2436,11 @@ doContent(XML_Parser parser,
+           for (;;) {
+             int bufSize;
+             int convLen;
+-            XmlConvert(enc,
++            const enum XML_Convert_Result convert_res = XmlConvert(enc,
+                        &fromPtr, rawNameEnd,
+                        (ICHAR **)&toPtr, (ICHAR *)tag->bufEnd - 1);
+             convLen = (int)(toPtr - (XML_Char *)tag->buf);
+-            if (fromPtr == rawNameEnd) {
++            if ((convert_res == XML_CONVERT_COMPLETED) || (convert_res == XML_CONVERT_INPUT_INCOMPLETE)) {
+               tag->name.strLen = convLen;
+               break;
+             }
+@@ -2661,11 +2661,11 @@ doContent(XML_Parser parser,
+           if (MUST_CONVERT(enc, s)) {
+             for (;;) {
+               ICHAR *dataPtr = (ICHAR *)dataBuf;
+-              XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd);
++              const enum XML_Convert_Result convert_res = XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd);
+               *eventEndPP = s;
+               charDataHandler(handlerArg, dataBuf,
+                               (int)(dataPtr - (ICHAR *)dataBuf));
+-              if (s == next)
++              if ((convert_res == XML_CONVERT_COMPLETED) || (convert_res == XML_CONVERT_INPUT_INCOMPLETE))
+                 break;
+               *eventPP = s;
+             }
+@@ -3269,11 +3269,11 @@ doCdataSection(XML_Parser parser,
+           if (MUST_CONVERT(enc, s)) {
+             for (;;) {
+               ICHAR *dataPtr = (ICHAR *)dataBuf;
+-              XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd);
++              const enum XML_Convert_Result convert_res = XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd);
+               *eventEndPP = next;
+               charDataHandler(handlerArg, dataBuf,
+                               (int)(dataPtr - (ICHAR *)dataBuf));
+-              if (s == next)
++              if ((convert_res == XML_CONVERT_COMPLETED) || (convert_res == XML_CONVERT_INPUT_INCOMPLETE))
+                 break;
+               *eventPP = s;
+             }
+@@ -5350,6 +5350,7 @@ reportDefault(XML_Parser parser, const ENCODING *enc,
+               const char *s, const char *end)
+ {
+   if (MUST_CONVERT(enc, s)) {
++    enum XML_Convert_Result convert_res;
+     const char **eventPP;
+     const char **eventEndPP;
+     if (enc == encoding) {
+@@ -5362,11 +5363,11 @@ reportDefault(XML_Parser parser, const ENCODING *enc,
+     }
+     do {
+       ICHAR *dataPtr = (ICHAR *)dataBuf;
+-      XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd);
++      convert_res = XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd);
+       *eventEndPP = s;
+       defaultHandler(handlerArg, dataBuf, (int)(dataPtr - (ICHAR *)dataBuf));
+       *eventPP = s;
+-    } while (s != end);
++    } while ((convert_res != XML_CONVERT_COMPLETED) && (convert_res != XML_CONVERT_INPUT_INCOMPLETE));
+   }
+   else
+     defaultHandler(handlerArg, (XML_Char *)s, (int)((XML_Char *)end - (XML_Char *)s));
+@@ -6169,8 +6170,8 @@ poolAppend(STRING_POOL *pool, const ENCODING *enc,
+   if (!pool->ptr && !poolGrow(pool))
+     return NULL;
+   for (;;) {
+-    XmlConvert(enc, &ptr, end, (ICHAR **)&(pool->ptr), (ICHAR *)pool->end);
+-    if (ptr == end)
++    const enum XML_Convert_Result convert_res = XmlConvert(enc, &ptr, end, (ICHAR **)&(pool->ptr), (ICHAR *)pool->end);
++    if ((convert_res == XML_CONVERT_COMPLETED) || (convert_res == XML_CONVERT_INPUT_INCOMPLETE))
+       break;
+     if (!poolGrow(pool))
+       return NULL;
+@@ -6254,8 +6255,13 @@ poolGrow(STRING_POOL *pool)
+     }
+   }
+   if (pool->blocks && pool->start == pool->blocks->s) {
+-    int blockSize = (int)(pool->end - pool->start)*2;
+-    BLOCK *temp = (BLOCK *)
++    BLOCK *temp;
++    int blockSize = (int)((unsigned)(pool->end - pool->start)*2U);
++
++    if (blockSize < 0)
++      return XML_FALSE;
++
++    temp = (BLOCK *)
+       pool->mem->realloc_fcn(pool->blocks,
+                              (offsetof(BLOCK, s)
+                               + blockSize * sizeof(XML_Char)));
+@@ -6270,6 +6276,10 @@ poolGrow(STRING_POOL *pool)
+   else {
+     BLOCK *tem;
+     int blockSize = (int)(pool->end - pool->start);
++
++    if (blockSize < 0)
++      return XML_FALSE;
++
+     if (blockSize < INIT_BLOCK_SIZE)
+       blockSize = INIT_BLOCK_SIZE;
+     else
+diff --git a/lib/xmltok.c b/lib/xmltok.c
+index bf09dfc..cb98ce1 100644
+--- a/lib/xmltok.c
++++ b/lib/xmltok.c
+@@ -318,39 +318,55 @@ enum {  /* UTF8_cvalN is value of masked first byte of N byte sequence */
+   UTF8_cval4 = 0xf0
+ };
+ 
+-static void PTRCALL
++static enum XML_Convert_Result PTRCALL
+ utf8_toUtf8(const ENCODING *enc,
+             const char **fromP, const char *fromLim,
+             char **toP, const char *toLim)
+ {
++  enum XML_Convert_Result res = XML_CONVERT_COMPLETED;
+   char *to;
+   const char *from;
+   if (fromLim - *fromP > toLim - *toP) {
+     /* Avoid copying partial characters. */
++    res = XML_CONVERT_OUTPUT_EXHAUSTED;
+     for (fromLim = *fromP + (toLim - *toP); fromLim > *fromP; fromLim--)
+       if (((unsigned char)fromLim[-1] & 0xc0) != 0x80)
+         break;
+   }
+-  for (to = *toP, from = *fromP; from != fromLim; from++, to++)
++  for (to = *toP, from = *fromP; (from < fromLim) && (to < toLim); from++, to++)
+     *to = *from;
+   *fromP = from;
+   *toP = to;
++
++  if ((to == toLim) && (from < fromLim))
++    return XML_CONVERT_OUTPUT_EXHAUSTED;
++  else
++    return res;
+ }
+ 
+-static void PTRCALL
++static enum XML_Convert_Result PTRCALL
+ utf8_toUtf16(const ENCODING *enc,
+              const char **fromP, const char *fromLim,
+              unsigned short **toP, const unsigned short *toLim)
+ {
++  enum XML_Convert_Result res = XML_CONVERT_COMPLETED;
+   unsigned short *to = *toP;
+   const char *from = *fromP;
+-  while (from != fromLim && to != toLim) {
++  while (from < fromLim && to < toLim) {
+     switch (((struct normal_encoding *)enc)->type[(unsigned char)*from]) {
+     case BT_LEAD2:
++      if (fromLim - from < 2) {
++        res = XML_CONVERT_INPUT_INCOMPLETE;
++        break;
++      }
+       *to++ = (unsigned short)(((from[0] & 0x1f) << 6) | (from[1] & 0x3f));
+       from += 2;
+       break;
+     case BT_LEAD3:
++      if (fromLim - from < 3) {
++        res = XML_CONVERT_INPUT_INCOMPLETE;
++        break;
++      }
+       *to++ = (unsigned short)(((from[0] & 0xf) << 12)
+                                | ((from[1] & 0x3f) << 6) | (from[2] & 0x3f));
+       from += 3;
+@@ -358,8 +374,14 @@ utf8_toUtf16(const ENCODING *enc,
+     case BT_LEAD4:
+       {
+         unsigned long n;
+-        if (to + 1 == toLim)
++        if (toLim - to < 2) {
++          res = XML_CONVERT_OUTPUT_EXHAUSTED;
+           goto after;
++        }
++        if (fromLim - from < 4) {
++          res = XML_CONVERT_INPUT_INCOMPLETE;
++          goto after;
++        }
+         n = ((from[0] & 0x7) << 18) | ((from[1] & 0x3f) << 12)
+             | ((from[2] & 0x3f) << 6) | (from[3] & 0x3f);
+         n -= 0x10000;
+@@ -377,6 +399,7 @@ utf8_toUtf16(const ENCODING *enc,
+ after:
+   *fromP = from;
+   *toP = to;
++  return res;
+ }
+ 
+ #ifdef XML_NS
+@@ -425,7 +448,7 @@ static const struct normal_encoding internal_utf8_encoding = {
+   STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
+ };
+ 
+-static void PTRCALL
++static enum XML_Convert_Result PTRCALL
+ latin1_toUtf8(const ENCODING *enc,
+               const char **fromP, const char *fromLim,
+               char **toP, const char *toLim)
+@@ -433,30 +456,35 @@ latin1_toUtf8(const ENCODING *enc,
+   for (;;) {
+     unsigned char c;
+     if (*fromP == fromLim)
+-      break;
++      return XML_CONVERT_COMPLETED;
+     c = (unsigned char)**fromP;
+     if (c & 0x80) {
+       if (toLim - *toP < 2)
+-        break;
++        return XML_CONVERT_OUTPUT_EXHAUSTED;
+       *(*toP)++ = (char)((c >> 6) | UTF8_cval2);
+       *(*toP)++ = (char)((c & 0x3f) | 0x80);
+       (*fromP)++;
+     }
+     else {
+       if (*toP == toLim)
+-        break;
++        return XML_CONVERT_OUTPUT_EXHAUSTED;
+       *(*toP)++ = *(*fromP)++;
+     }
+   }
+ }
+ 
+-static void PTRCALL
++static enum XML_Convert_Result PTRCALL
+ latin1_toUtf16(const ENCODING *enc,
+                const char **fromP, const char *fromLim,
+                unsigned short **toP, const unsigned short *toLim)
+ {
+-  while (*fromP != fromLim && *toP != toLim)
++  while (*fromP < fromLim && *toP < toLim)
+     *(*toP)++ = (unsigned char)*(*fromP)++;
++
++  if ((*toP == toLim) && (*fromP < fromLim))
++    return XML_CONVERT_OUTPUT_EXHAUSTED;
++  else
++    return XML_CONVERT_COMPLETED;
+ }
+ 
+ #ifdef XML_NS
+@@ -483,13 +511,18 @@ static const struct normal_encoding latin1_encoding = {
+   STANDARD_VTABLE(sb_)
+ };
+ 
+-static void PTRCALL
++static enum XML_Convert_Result PTRCALL
+ ascii_toUtf8(const ENCODING *enc,
+              const char **fromP, const char *fromLim,
+              char **toP, const char *toLim)
+ {
+-  while (*fromP != fromLim && *toP != toLim)
++  while (*fromP < fromLim && *toP < toLim)
+     *(*toP)++ = *(*fromP)++;
++
++  if ((*toP == toLim) && (*fromP < fromLim))
++    return XML_CONVERT_OUTPUT_EXHAUSTED;
++  else
++    return XML_CONVERT_COMPLETED;
+ }
+ 
+ #ifdef XML_NS
+@@ -536,13 +569,14 @@ unicode_byte_type(char hi, char lo)
+ }
+ 
+ #define DEFINE_UTF16_TO_UTF8(E) \
+-static void  PTRCALL \
++static enum XML_Convert_Result  PTRCALL \
+ E ## toUtf8(const ENCODING *enc, \
+             const char **fromP, const char *fromLim, \
+             char **toP, const char *toLim) \
+ { \
+-  const char *from; \
+-  for (from = *fromP; from != fromLim; from += 2) { \
++  const char *from = *fromP; \
++  fromLim = from + (((fromLim - from) >> 1) << 1);  /* shrink to even */ \
++  for (; from < fromLim; from += 2) { \
+     int plane; \
+     unsigned char lo2; \
+     unsigned char lo = GET_LO(from); \
+@@ -552,7 +586,7 @@ E ## toUtf8(const ENCODING *enc, \
+       if (lo < 0x80) { \
+         if (*toP == toLim) { \
+           *fromP = from; \
+-          return; \
++          return XML_CONVERT_OUTPUT_EXHAUSTED; \
+         } \
+         *(*toP)++ = lo; \
+         break; \
+@@ -562,7 +596,7 @@ E ## toUtf8(const ENCODING *enc, \
+     case 0x4: case 0x5: case 0x6: case 0x7: \
+       if (toLim -  *toP < 2) { \
+         *fromP = from; \
+-        return; \
++        return XML_CONVERT_OUTPUT_EXHAUSTED; \
+       } \
+       *(*toP)++ = ((lo >> 6) | (hi << 2) |  UTF8_cval2); \
+       *(*toP)++ = ((lo & 0x3f) | 0x80); \
+@@ -570,7 +604,7 @@ E ## toUtf8(const ENCODING *enc, \
+     default: \
+       if (toLim -  *toP < 3)  { \
+         *fromP = from; \
+-        return; \
++        return XML_CONVERT_OUTPUT_EXHAUSTED; \
+       } \
+       /* 16 bits divided 4, 6, 6 amongst 3 bytes */ \
+       *(*toP)++ = ((hi >> 4) | UTF8_cval3); \
+@@ -580,7 +614,11 @@ E ## toUtf8(const ENCODING *enc, \
+     case 0xD8: case 0xD9: case 0xDA: case 0xDB: \
+       if (toLim -  *toP < 4) { \
+         *fromP = from; \
+-        return; \
++        return XML_CONVERT_OUTPUT_EXHAUSTED; \
++      } \
++      if (fromLim - from < 4) { \
++        *fromP = from; \
++        return XML_CONVERT_INPUT_INCOMPLETE; \
+       } \
+       plane = (((hi & 0x3) << 2) | ((lo >> 6) & 0x3)) + 1; \
+       *(*toP)++ = ((plane >> 2) | UTF8_cval4); \
+@@ -596,20 +634,32 @@ E ## toUtf8(const ENCODING *enc, \
+     } \
+   } \
+   *fromP = from; \
++  if (from < fromLim) \
++    return XML_CONVERT_INPUT_INCOMPLETE; \
++  else \
++    return XML_CONVERT_COMPLETED; \
+ }
+ 
+ #define DEFINE_UTF16_TO_UTF16(E) \
+-static void  PTRCALL \
++static enum XML_Convert_Result  PTRCALL \
+ E ## toUtf16(const ENCODING *enc, \
+              const char **fromP, const char *fromLim, \
+              unsigned short **toP, const unsigned short *toLim) \
+ { \
++  enum XML_Convert_Result res = XML_CONVERT_COMPLETED; \
++  fromLim = *fromP + (((fromLim - *fromP) >> 1) << 1);  /* shrink to even */ \
+   /* Avoid copying first half only of surrogate */ \
+   if (fromLim - *fromP > ((toLim - *toP) << 1) \
+-      && (GET_HI(fromLim - 2) & 0xF8) == 0xD8) \
++      && (GET_HI(fromLim - 2) & 0xF8) == 0xD8) { \
+     fromLim -= 2; \
+-  for (; *fromP != fromLim && *toP != toLim; *fromP += 2) \
++    res = XML_CONVERT_INPUT_INCOMPLETE; \
++  } \
++  for (; *fromP < fromLim && *toP < toLim; *fromP += 2) \
+     *(*toP)++ = (GET_HI(*fromP) << 8) | GET_LO(*fromP); \
++  if ((*toP == toLim) && (*fromP < fromLim)) \
++    return XML_CONVERT_OUTPUT_EXHAUSTED; \
++  else \
++    return res; \
+ }
+ 
+ #define SET2(ptr, ch) \
+@@ -1288,7 +1338,7 @@ unknown_isInvalid(const ENCODING *enc, const char *p)
+   return (c & ~0xFFFF) || checkCharRefNumber(c) < 0;
+ }
+ 
+-static void PTRCALL
++static enum XML_Convert_Result PTRCALL
+ unknown_toUtf8(const ENCODING *enc,
+                const char **fromP, const char *fromLim,
+                char **toP, const char *toLim)
+@@ -1299,21 +1349,21 @@ unknown_toUtf8(const ENCODING *enc,
+     const char *utf8;
+     int n;
+     if (*fromP == fromLim)
+-      break;
++      return XML_CONVERT_COMPLETED;
+     utf8 = uenc->utf8[(unsigned char)**fromP];
+     n = *utf8++;
+     if (n == 0) {
+       int c = uenc->convert(uenc->userData, *fromP);
+       n = XmlUtf8Encode(c, buf);
+       if (n > toLim - *toP)
+-        break;
++        return XML_CONVERT_OUTPUT_EXHAUSTED;
+       utf8 = buf;
+       *fromP += (AS_NORMAL_ENCODING(enc)->type[(unsigned char)**fromP]
+                  - (BT_LEAD2 - 2));
+     }
+     else {
+       if (n > toLim - *toP)
+-        break;
++        return XML_CONVERT_OUTPUT_EXHAUSTED;
+       (*fromP)++;
+     }
+     do {
+@@ -1322,13 +1372,13 @@ unknown_toUtf8(const ENCODING *enc,
+   }
+ }
+ 
+-static void PTRCALL
++static enum XML_Convert_Result PTRCALL
+ unknown_toUtf16(const ENCODING *enc,
+                 const char **fromP, const char *fromLim,
+                 unsigned short **toP, const unsigned short *toLim)
+ {
+   const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc);
+-  while (*fromP != fromLim && *toP != toLim) {
++  while (*fromP < fromLim && *toP < toLim) {
+     unsigned short c = uenc->utf16[(unsigned char)**fromP];
+     if (c == 0) {
+       c = (unsigned short)
+@@ -1340,6 +1390,11 @@ unknown_toUtf16(const ENCODING *enc,
+       (*fromP)++;
+     *(*toP)++ = c;
+   }
++
++  if ((*toP == toLim) && (*fromP < fromLim))
++    return XML_CONVERT_OUTPUT_EXHAUSTED;
++  else
++    return XML_CONVERT_COMPLETED;
+ }
+ 
+ ENCODING *
+@@ -1503,7 +1558,7 @@ initScan(const ENCODING * const *encodingTable,
+ {
+   const ENCODING **encPtr;
+ 
+-  if (ptr == end)
++  if (ptr >= end)
+     return XML_TOK_NONE;
+   encPtr = enc->encPtr;
+   if (ptr + 1 == end) {
+diff --git a/lib/xmltok.h b/lib/xmltok.h
+index ca867aa..752007e 100644
+--- a/lib/xmltok.h
++++ b/lib/xmltok.h
+@@ -130,6 +130,12 @@ typedef int (PTRCALL *SCANNER)(const ENCODING *,
+                                const char *,
+                                const char **);
+ 
++enum XML_Convert_Result {
++  XML_CONVERT_COMPLETED = 0,
++  XML_CONVERT_INPUT_INCOMPLETE = 1,
++  XML_CONVERT_OUTPUT_EXHAUSTED = 2  /* and therefore potentially input remaining as well */
++};
++
+ struct encoding {
+   SCANNER scanners[XML_N_STATES];
+   SCANNER literalScanners[XML_N_LITERAL_TYPES];
+@@ -158,12 +164,12 @@ struct encoding {
+                             const char *ptr,
+                             const char *end,
+                             const char **badPtr);
+-  void (PTRCALL *utf8Convert)(const ENCODING *enc,
++  enum XML_Convert_Result (PTRCALL *utf8Convert)(const ENCODING *enc,
+                               const char **fromP,
+                               const char *fromLim,
+                               char **toP,
+                               const char *toLim);
+-  void (PTRCALL *utf16Convert)(const ENCODING *enc,
++  enum XML_Convert_Result (PTRCALL *utf16Convert)(const ENCODING *enc,
+                                const char **fromP,
+                                const char *fromLim,
+                                unsigned short **toP,
+diff --git a/lib/xmltok_impl.c b/lib/xmltok_impl.c
+index 9c2895b..6c5a3ba 100644
+--- a/lib/xmltok_impl.c
++++ b/lib/xmltok_impl.c
+@@ -93,13 +93,13 @@ static int PTRCALL
+ PREFIX(scanComment)(const ENCODING *enc, const char *ptr,
+                     const char *end, const char **nextTokPtr)
+ {
+-  if (ptr != end) {
++  if (ptr < end) {
+     if (!CHAR_MATCHES(enc, ptr, ASCII_MINUS)) {
+       *nextTokPtr = ptr;
+       return XML_TOK_INVALID;
+     }
+     ptr += MINBPC(enc);
+-    while (ptr != end) {
++    while (ptr < end) {
+       switch (BYTE_TYPE(enc, ptr)) {
+       INVALID_CASES(ptr, nextTokPtr)
+       case BT_MINUS:
+@@ -147,7 +147,7 @@ PREFIX(scanDecl)(const ENCODING *enc, const char *ptr,
+     *nextTokPtr = ptr;
+     return XML_TOK_INVALID;
+   }
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     case BT_PERCNT:
+       if (ptr + MINBPC(enc) == end)
+@@ -233,7 +233,7 @@ PREFIX(scanPi)(const ENCODING *enc, const char *ptr,
+     *nextTokPtr = ptr;
+     return XML_TOK_INVALID;
+   }
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+     case BT_S: case BT_CR: case BT_LF:
+@@ -242,7 +242,7 @@ PREFIX(scanPi)(const ENCODING *enc, const char *ptr,
+         return XML_TOK_INVALID;
+       }
+       ptr += MINBPC(enc);
+-      while (ptr != end) {
++      while (ptr < end) {
+         switch (BYTE_TYPE(enc, ptr)) {
+         INVALID_CASES(ptr, nextTokPtr)
+         case BT_QUEST:
+@@ -305,7 +305,7 @@ static int PTRCALL
+ PREFIX(cdataSectionTok)(const ENCODING *enc, const char *ptr,
+                         const char *end, const char **nextTokPtr)
+ {
+-  if (ptr == end)
++  if (ptr >= end)
+     return XML_TOK_NONE;
+   if (MINBPC(enc) > 1) {
+     size_t n = end - ptr;
+@@ -348,7 +348,7 @@ PREFIX(cdataSectionTok)(const ENCODING *enc, const char *ptr,
+     ptr += MINBPC(enc);
+     break;
+   }
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+ #define LEAD_CASE(n) \
+     case BT_LEAD ## n: \
+@@ -391,11 +391,11 @@ PREFIX(scanEndTag)(const ENCODING *enc, const char *ptr,
+     *nextTokPtr = ptr;
+     return XML_TOK_INVALID;
+   }
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+     case BT_S: case BT_CR: case BT_LF:
+-      for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) {
++      for (ptr += MINBPC(enc); ptr < end; ptr += MINBPC(enc)) {
+         switch (BYTE_TYPE(enc, ptr)) {
+         case BT_S: case BT_CR: case BT_LF:
+           break;
+@@ -432,7 +432,7 @@ static int PTRCALL
+ PREFIX(scanHexCharRef)(const ENCODING *enc, const char *ptr,
+                        const char *end, const char **nextTokPtr)
+ {
+-  if (ptr != end) {
++  if (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     case BT_DIGIT:
+     case BT_HEX:
+@@ -441,7 +441,7 @@ PREFIX(scanHexCharRef)(const ENCODING *enc, const char *ptr,
+       *nextTokPtr = ptr;
+       return XML_TOK_INVALID;
+     }
+-    for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) {
++    for (ptr += MINBPC(enc); ptr < end; ptr += MINBPC(enc)) {
+       switch (BYTE_TYPE(enc, ptr)) {
+       case BT_DIGIT:
+       case BT_HEX:
+@@ -464,7 +464,7 @@ static int PTRCALL
+ PREFIX(scanCharRef)(const ENCODING *enc, const char *ptr,
+                     const char *end, const char **nextTokPtr)
+ {
+-  if (ptr != end) {
++  if (ptr < end) {
+     if (CHAR_MATCHES(enc, ptr, ASCII_x))
+       return PREFIX(scanHexCharRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+     switch (BYTE_TYPE(enc, ptr)) {
+@@ -474,7 +474,7 @@ PREFIX(scanCharRef)(const ENCODING *enc, const char *ptr,
+       *nextTokPtr = ptr;
+       return XML_TOK_INVALID;
+     }
+-    for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) {
++    for (ptr += MINBPC(enc); ptr < end; ptr += MINBPC(enc)) {
+       switch (BYTE_TYPE(enc, ptr)) {
+       case BT_DIGIT:
+         break;
+@@ -506,7 +506,7 @@ PREFIX(scanRef)(const ENCODING *enc, const char *ptr, const char *end,
+     *nextTokPtr = ptr;
+     return XML_TOK_INVALID;
+   }
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+     case BT_SEMI:
+@@ -529,7 +529,7 @@ PREFIX(scanAtts)(const ENCODING *enc, const char *ptr, const char *end,
+ #ifdef XML_NS
+   int hadColon = 0;
+ #endif
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+ #ifdef XML_NS
+@@ -716,7 +716,7 @@ PREFIX(scanLt)(const ENCODING *enc, const char *ptr, const char *end,
+   hadColon = 0;
+ #endif
+   /* we have a start-tag */
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+ #ifdef XML_NS
+@@ -740,7 +740,7 @@ PREFIX(scanLt)(const ENCODING *enc, const char *ptr, const char *end,
+     case BT_S: case BT_CR: case BT_LF:
+       {
+         ptr += MINBPC(enc);
+-        while (ptr != end) {
++        while (ptr < end) {
+           switch (BYTE_TYPE(enc, ptr)) {
+           CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+           case BT_GT:
+@@ -785,7 +785,7 @@ static int PTRCALL
+ PREFIX(contentTok)(const ENCODING *enc, const char *ptr, const char *end,
+                    const char **nextTokPtr)
+ {
+-  if (ptr == end)
++  if (ptr >= end)
+     return XML_TOK_NONE;
+   if (MINBPC(enc) > 1) {
+     size_t n = end - ptr;
+@@ -832,7 +832,7 @@ PREFIX(contentTok)(const ENCODING *enc, const char *ptr, const char *end,
+     ptr += MINBPC(enc);
+     break;
+   }
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+ #define LEAD_CASE(n) \
+     case BT_LEAD ## n: \
+@@ -895,7 +895,7 @@ PREFIX(scanPercent)(const ENCODING *enc, const char *ptr, const char *end,
+     *nextTokPtr = ptr;
+     return XML_TOK_INVALID;
+   }
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+     case BT_SEMI:
+@@ -921,7 +921,7 @@ PREFIX(scanPoundName)(const ENCODING *enc, const char *ptr, const char *end,
+     *nextTokPtr = ptr;
+     return XML_TOK_INVALID;
+   }
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+     case BT_CR: case BT_LF: case BT_S:
+@@ -941,7 +941,7 @@ PREFIX(scanLit)(int open, const ENCODING *enc,
+                 const char *ptr, const char *end,
+                 const char **nextTokPtr)
+ {
+-  while (ptr != end) {
++  while (ptr < end) {
+     int t = BYTE_TYPE(enc, ptr);
+     switch (t) {
+     INVALID_CASES(ptr, nextTokPtr)
+@@ -973,7 +973,7 @@ PREFIX(prologTok)(const ENCODING *enc, const char *ptr, const char *end,
+                   const char **nextTokPtr)
+ {
+   int tok;
+-  if (ptr == end)
++  if (ptr >= end)
+     return XML_TOK_NONE;
+   if (MINBPC(enc) > 1) {
+     size_t n = end - ptr;
+@@ -1141,7 +1141,7 @@ PREFIX(prologTok)(const ENCODING *enc, const char *ptr, const char *end,
+     *nextTokPtr = ptr;
+     return XML_TOK_INVALID;
+   }
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+     case BT_GT: case BT_RPAR: case BT_COMMA:
+@@ -1204,10 +1204,10 @@ PREFIX(attributeValueTok)(const ENCODING *enc, const char *ptr,
+                           const char *end, const char **nextTokPtr)
+ {
+   const char *start;
+-  if (ptr == end)
++  if (ptr >= end)
+     return XML_TOK_NONE;
+   start = ptr;
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+ #define LEAD_CASE(n) \
+     case BT_LEAD ## n: ptr += n; break;
+@@ -1262,10 +1262,10 @@ PREFIX(entityValueTok)(const ENCODING *enc, const char *ptr,
+                        const char *end, const char **nextTokPtr)
+ {
+   const char *start;
+-  if (ptr == end)
++  if (ptr >= end)
+     return XML_TOK_NONE;
+   start = ptr;
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+ #define LEAD_CASE(n) \
+     case BT_LEAD ## n: ptr += n; break;
+@@ -1326,7 +1326,7 @@ PREFIX(ignoreSectionTok)(const ENCODING *enc, const char *ptr,
+       end = ptr + n;
+     }
+   }
+-  while (ptr != end) {
++  while (ptr < end) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     INVALID_CASES(ptr, nextTokPtr)
+     case BT_LT:
+@@ -1373,7 +1373,7 @@ PREFIX(isPublicId)(const ENCODING *enc, const char *ptr, const char *end,
+ {
+   ptr += MINBPC(enc);
+   end -= MINBPC(enc);
+-  for (; ptr != end; ptr += MINBPC(enc)) {
++  for (; ptr < end; ptr += MINBPC(enc)) {
+     switch (BYTE_TYPE(enc, ptr)) {
+     case BT_DIGIT:
+     case BT_HEX:
+@@ -1760,7 +1760,7 @@ PREFIX(updatePosition)(const ENCODING *enc,
+     case BT_CR:
+       pos->lineNumber++;
+       ptr += MINBPC(enc);
+-      if (ptr != end && BYTE_TYPE(enc, ptr) == BT_LF)
++      if (ptr < end && BYTE_TYPE(enc, ptr) == BT_LF)
+         ptr += MINBPC(enc);
+       pos->columnNumber = (XML_Size)-1;
+       break;
+-- 
+2.8.2
+