[Python-checkins] bpo-34625: Update vendorized expat version to 2.2.6. (GH-9150)

Miss Islington (bot) webhook-mailer at python.org
Tue Sep 11 00:55:30 EDT 2018


https://github.com/python/cpython/commit/2bc4eea8f6675b157adcaea4a59d982d58a5e7ae
commit: 2bc4eea8f6675b157adcaea4a59d982d58a5e7ae
branch: 3.7
author: Miss Islington (bot) <31488909+miss-islington at users.noreply.github.com>
committer: GitHub <noreply at github.com>
date: 2018-09-10T21:55:27-07:00
summary:

bpo-34625: Update vendorized expat version to 2.2.6. (GH-9150)

(cherry picked from commit 5033aa77aacaa5505636f150e8d54baac5bdca9c)

Co-authored-by: Benjamin Peterson <benjamin at python.org>

files:
A Misc/NEWS.d/next/Library/2018-09-10-17-46-51.bpo-34625.D2YfDz.rst
M Modules/expat/expat.h
M Modules/expat/siphash.h
M Modules/expat/xmlparse.c
M Modules/expat/xmltok_impl.c

diff --git a/Misc/NEWS.d/next/Library/2018-09-10-17-46-51.bpo-34625.D2YfDz.rst b/Misc/NEWS.d/next/Library/2018-09-10-17-46-51.bpo-34625.D2YfDz.rst
new file mode 100644
index 000000000000..0747ec54470f
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2018-09-10-17-46-51.bpo-34625.D2YfDz.rst
@@ -0,0 +1 @@
+Update vendorized expat library version to 2.2.6.
diff --git a/Modules/expat/expat.h b/Modules/expat/expat.h
index 1f608c02d6fd..174c3fafda3f 100644
--- a/Modules/expat/expat.h
+++ b/Modules/expat/expat.h
@@ -264,7 +264,7 @@ XML_ParserCreate_MM(const XML_Char *encoding,
                     const XML_Char *namespaceSeparator);
 
 /* Prepare a parser object to be re-used.  This is particularly
-   valuable when memory allocation overhead is disproportionatly high,
+   valuable when memory allocation overhead is disproportionately high,
    such as when a large number of small documnents need to be parsed.
    All handlers are cleared from the parser, except for the
    unknownEncodingHandler. The parser's external state is re-initialized
@@ -1076,7 +1076,7 @@ XML_GetFeatureList(void);
 */
 #define XML_MAJOR_VERSION 2
 #define XML_MINOR_VERSION 2
-#define XML_MICRO_VERSION 5
+#define XML_MICRO_VERSION 6
 
 #ifdef __cplusplus
 }
diff --git a/Modules/expat/siphash.h b/Modules/expat/siphash.h
index 581872df7b49..4d6786d7839a 100644
--- a/Modules/expat/siphash.h
+++ b/Modules/expat/siphash.h
@@ -11,6 +11,12 @@
  * --------------------------------------------------------------------------
  * HISTORY:
  *
+ * 2018-07-08  (Anton Maklakov)
+ *   - Add "fall through" markers for GCC's -Wimplicit-fallthrough
+ *
+ * 2017-11-03  (Sebastian Pipping)
+ *   - Hide sip_tobin and sip_binof unless SIPHASH_TOBIN macro is defined
+ *
  * 2017-07-25  (Vadim Zeitlin)
  *   - Fix use of SIPHASH_MAIN macro
  *
@@ -151,6 +157,8 @@ static struct sipkey *sip_tokey(struct sipkey *key, const void *src) {
 } /* sip_tokey() */
 
 
+#ifdef SIPHASH_TOBIN
+
 #define sip_binof(v) sip_tobin((unsigned char[8]){ 0 }, (v))
 
 static void *sip_tobin(void *dst, uint64_t u64) {
@@ -158,6 +166,8 @@ static void *sip_tobin(void *dst, uint64_t u64) {
 	return dst;
 } /* sip_tobin() */
 
+#endif  /* SIPHASH_TOBIN */
+
 
 static void sip_round(struct siphash *H, const int rounds) {
 	int i;
@@ -231,12 +241,19 @@ static uint64_t sip24_final(struct siphash *H) {
 
 	switch (left) {
 	case 7: b |= (uint64_t)H->buf[6] << 48;
+		/* fall through */
 	case 6: b |= (uint64_t)H->buf[5] << 40;
+		/* fall through */
 	case 5: b |= (uint64_t)H->buf[4] << 32;
+		/* fall through */
 	case 4: b |= (uint64_t)H->buf[3] << 24;
+		/* fall through */
 	case 3: b |= (uint64_t)H->buf[2] << 16;
+		/* fall through */
 	case 2: b |= (uint64_t)H->buf[1] << 8;
+		/* fall through */
 	case 1: b |= (uint64_t)H->buf[0] << 0;
+		/* fall through */
 	case 0: break;
 	}
 
diff --git a/Modules/expat/xmlparse.c b/Modules/expat/xmlparse.c
index 90a237f30eb8..c4f3ffc215c9 100644
--- a/Modules/expat/xmlparse.c
+++ b/Modules/expat/xmlparse.c
@@ -1,4 +1,4 @@
-/* 4b74aa710b4ed5ce464b0ce544852cb47bf905c85a49c7bae2749f5885cb966d (2.2.5+)
+/* 19ac4776051591216f1874e34ee99b6a43a3784c8bd7d70efeb9258dd22b906a (2.2.6+)
                             __  __            _
                          ___\ \/ /_ __   __ _| |_
                         / _ \\  /| '_ \ / _` | __|
@@ -161,6 +161,9 @@ typedef char ICHAR;
 /* Round up n to be a multiple of sz, where sz is a power of 2. */
 #define ROUND_UP(n, sz) (((n) + ((sz) - 1)) & ~((sz) - 1))
 
+/* Do safe (NULL-aware) pointer arithmetic */
+#define EXPAT_SAFE_PTR_DIFF(p, q) (((p) && (q)) ? ((p) - (q)) : 0)
+
 /* Handle the case where memmove() doesn't exist. */
 #ifndef HAVE_MEMMOVE
 #ifdef HAVE_BCOPY
@@ -1820,6 +1823,7 @@ XML_Parse(XML_Parser parser, const char *s, int len, int isFinal)
       parser->m_errorCode = XML_ERROR_NO_MEMORY;
       return XML_STATUS_ERROR;
     }
+    /* fall through */
   default:
     parser->m_parsingStatus.parsing = XML_PARSING;
   }
@@ -1969,6 +1973,7 @@ XML_ParseBuffer(XML_Parser parser, int len, int isFinal)
       parser->m_errorCode = XML_ERROR_NO_MEMORY;
       return XML_STATUS_ERROR;
     }
+    /* fall through */
   default:
     parser->m_parsingStatus.parsing = XML_PARSING;
   }
@@ -2026,39 +2031,46 @@ XML_GetBuffer(XML_Parser parser, int len)
   default: ;
   }
 
-  if (len > parser->m_bufferLim - parser->m_bufferEnd) {
+  if (len > EXPAT_SAFE_PTR_DIFF(parser->m_bufferLim, parser->m_bufferEnd)) {
 #ifdef XML_CONTEXT_BYTES
     int keep;
 #endif  /* defined XML_CONTEXT_BYTES */
     /* Do not invoke signed arithmetic overflow: */
-    int neededSize = (int) ((unsigned)len + (unsigned)(parser->m_bufferEnd - parser->m_bufferPtr));
+    int neededSize = (int) ((unsigned)len +
+                            (unsigned)EXPAT_SAFE_PTR_DIFF(parser->m_bufferEnd,
+                                                          parser->m_bufferPtr));
     if (neededSize < 0) {
       parser->m_errorCode = XML_ERROR_NO_MEMORY;
       return NULL;
     }
 #ifdef XML_CONTEXT_BYTES
-    keep = (int)(parser->m_bufferPtr - parser->m_buffer);
+    keep = (int)EXPAT_SAFE_PTR_DIFF(parser->m_bufferPtr, parser->m_buffer);
     if (keep > XML_CONTEXT_BYTES)
       keep = XML_CONTEXT_BYTES;
     neededSize += keep;
 #endif  /* defined XML_CONTEXT_BYTES */
-    if (neededSize  <= parser->m_bufferLim - parser->m_buffer) {
+    if (neededSize <= EXPAT_SAFE_PTR_DIFF(parser->m_bufferLim, parser->m_buffer)) {
 #ifdef XML_CONTEXT_BYTES
-      if (keep < parser->m_bufferPtr - parser->m_buffer) {
-        int offset = (int)(parser->m_bufferPtr - parser->m_buffer) - keep;
+      if (keep < EXPAT_SAFE_PTR_DIFF(parser->m_bufferPtr, parser->m_buffer)) {
+          int offset = (int)EXPAT_SAFE_PTR_DIFF(parser->m_bufferPtr, parser->m_buffer) - keep;
+        /* The buffer pointers cannot be NULL here; we have at least some bytes in the buffer */
         memmove(parser->m_buffer, &parser->m_buffer[offset], parser->m_bufferEnd - parser->m_bufferPtr + keep);
         parser->m_bufferEnd -= offset;
         parser->m_bufferPtr -= offset;
       }
 #else
-      memmove(parser->m_buffer, parser->m_bufferPtr, parser->m_bufferEnd - parser->m_bufferPtr);
-      parser->m_bufferEnd = parser->m_buffer + (parser->m_bufferEnd - parser->m_bufferPtr);
-      parser->m_bufferPtr = parser->m_buffer;
+      if (parser->m_buffer && parser->m_bufferPtr) {
+        memmove(parser->m_buffer, parser->m_bufferPtr,
+                EXPAT_SAFE_PTR_DIFF(parser->m_bufferEnd, parser->m_bufferPtr));
+        parser->m_bufferEnd = parser->m_buffer +
+            EXPAT_SAFE_PTR_DIFF(parser->m_bufferEnd, parser->m_bufferPtr);
+        parser->m_bufferPtr = parser->m_buffer;
+      }
 #endif  /* not defined XML_CONTEXT_BYTES */
     }
     else {
       char *newBuf;
-      int bufferSize = (int)(parser->m_bufferLim - parser->m_bufferPtr);
+      int bufferSize = (int)EXPAT_SAFE_PTR_DIFF(parser->m_bufferLim, parser->m_bufferPtr);
       if (bufferSize == 0)
         bufferSize = INIT_BUFFER_SIZE;
       do {
@@ -2077,25 +2089,34 @@ XML_GetBuffer(XML_Parser parser, int len)
       parser->m_bufferLim = newBuf + bufferSize;
 #ifdef XML_CONTEXT_BYTES
       if (parser->m_bufferPtr) {
-        int keep = (int)(parser->m_bufferPtr - parser->m_buffer);
+        int keep = (int)EXPAT_SAFE_PTR_DIFF(parser->m_bufferPtr, parser->m_buffer);
         if (keep > XML_CONTEXT_BYTES)
           keep = XML_CONTEXT_BYTES;
-        memcpy(newBuf, &parser->m_bufferPtr[-keep], parser->m_bufferEnd - parser->m_bufferPtr + keep);
+        memcpy(newBuf, &parser->m_bufferPtr[-keep],
+               EXPAT_SAFE_PTR_DIFF(parser->m_bufferEnd, parser->m_bufferPtr) + keep);
         FREE(parser, parser->m_buffer);
         parser->m_buffer = newBuf;
-        parser->m_bufferEnd = parser->m_buffer + (parser->m_bufferEnd - parser->m_bufferPtr) + keep;
+        parser->m_bufferEnd = parser->m_buffer +
+            EXPAT_SAFE_PTR_DIFF(parser->m_bufferEnd, parser->m_bufferPtr) + keep;
         parser->m_bufferPtr = parser->m_buffer + keep;
       }
       else {
-        parser->m_bufferEnd = newBuf + (parser->m_bufferEnd - parser->m_bufferPtr);
+        /* This must be a brand new buffer with no data in it yet */
+        parser->m_bufferEnd = newBuf;
         parser->m_bufferPtr = parser->m_buffer = newBuf;
       }
 #else
       if (parser->m_bufferPtr) {
-        memcpy(newBuf, parser->m_bufferPtr, parser->m_bufferEnd - parser->m_bufferPtr);
+        memcpy(newBuf, parser->m_bufferPtr,
+               EXPAT_SAFE_PTR_DIFF(parser->m_bufferEnd, parser->m_bufferPtr));
         FREE(parser, parser->m_buffer);
+        parser->m_bufferEnd = newBuf +
+            EXPAT_SAFE_PTR_DIFF(parser->m_bufferEnd, parser->m_bufferPtr);
+      }
+      else {
+        /* This must be a brand new buffer with no data in it yet */
+        parser->m_bufferEnd = newBuf;
       }
-      parser->m_bufferEnd = newBuf + (parser->m_bufferEnd - parser->m_bufferPtr);
       parser->m_bufferPtr = parser->m_buffer = newBuf;
 #endif  /* not defined XML_CONTEXT_BYTES */
     }
@@ -2908,9 +2929,11 @@ doContent(XML_Parser parser,
         poolClear(&parser->m_tempPool);
         freeBindings(parser, bindings);
       }
-      if ((parser->m_tagLevel == 0) &&
-          !((parser->m_parsingStatus.parsing == XML_FINISHED) || (parser->m_parsingStatus.parsing == XML_SUSPENDED))) {
-        return epilogProcessor(parser, next, end, nextPtr);
+      if ((parser->m_tagLevel == 0) && (parser->m_parsingStatus.parsing != XML_FINISHED)) {
+        if (parser->m_parsingStatus.parsing == XML_SUSPENDED)
+          parser->m_processor = epilogProcessor;
+        else
+          return epilogProcessor(parser, next, end, nextPtr);
       }
       break;
     case XML_TOK_END_TAG:
@@ -4746,8 +4769,8 @@ doProlog(XML_Parser parser,
           return XML_ERROR_NO_MEMORY;
         parser->m_declEntity->publicId = NULL;
       }
-      /* fall through */
 #endif /* XML_DTD */
+      /* fall through */
     case XML_ROLE_ENTITY_SYSTEM_ID:
       if (dtd->keepProcessing && parser->m_declEntity) {
         parser->m_declEntity->systemId = poolStoreString(&dtd->pool, enc,
@@ -6643,7 +6666,6 @@ hash(XML_Parser parser, KEY s)
 {
   struct siphash state;
   struct sipkey key;
-  (void)sip_tobin;
   (void)sip24_valid;
   copy_salt_to_sipkey(parser, &key);
   sip24_init(&state, &key);
diff --git a/Modules/expat/xmltok_impl.c b/Modules/expat/xmltok_impl.c
index 0403dd3d09a0..4d9ae7dc3896 100644
--- a/Modules/expat/xmltok_impl.c
+++ b/Modules/expat/xmltok_impl.c
@@ -74,6 +74,7 @@
       *nextTokPtr = ptr; \
       return XML_TOK_INVALID; \
     } \
+    /* fall through */ \
   case BT_NMSTRT: \
   case BT_HEX: \
   case BT_DIGIT: \
@@ -102,6 +103,7 @@
       *nextTokPtr = ptr; \
       return XML_TOK_INVALID; \
     } \
+    /* fall through */ \
   case BT_NMSTRT: \
   case BT_HEX: \
     ptr += MINBPC(enc); \
@@ -602,7 +604,7 @@ PREFIX(scanAtts)(const ENCODING *enc, const char *ptr, const char *end,
           return XML_TOK_INVALID;
         }
       }
-    /* fall through */
+      /* fall through */
     case BT_EQUALS:
       {
         int open;
@@ -1442,6 +1444,7 @@ PREFIX(isPublicId)(const ENCODING *enc, const char *ptr, const char *end,
     case BT_NMSTRT:
       if (!(BYTE_TO_ASCII(enc, ptr) & ~0x7f))
         break;
+      /* fall through */
     default:
       switch (BYTE_TO_ASCII(enc, ptr)) {
       case 0x24: /* $ */
@@ -1659,8 +1662,8 @@ PREFIX(nameMatchesAscii)(const ENCODING *UNUSED_P(enc), const char *ptr1,
 {
   for (; *ptr2; ptr1 += MINBPC(enc), ptr2++) {
     if (end1 - ptr1 < MINBPC(enc)) {
-      /* This line cannot be executed.  THe incoming data has already
-       * been tokenized once, so imcomplete characters like this have
+      /* This line cannot be executed.  The incoming data has already
+       * been tokenized once, so incomplete characters like this have
        * already been eliminated from the input.  Retaining the
        * paranoia check is still valuable, however.
        */



More information about the Python-checkins mailing list