[3.8] Upgrade bundled libexpat to 2.6.0 (pythonGH-115399) (pythonGH-115468)

Manual backport due to code differences.
(cherry picked from commit e071b0d)

Co-authored-by: Seth Michael Larson <[email protected]>

Author: sethmlarson

Misc/NEWS.d/next/Security/2024-02-13-15-14-39.gh-issue-115399.xT-scP.rst

@@ -0,0 +1 @@
+Update bundled libexpat to 2.6.0

Modules/expat/expat.h

@@ -11,11 +11,13 @@
    Copyright (c) 2000-2005 Fred L. Drake, Jr. <[email protected]>
    Copyright (c) 2001-2002 Greg Stein <[email protected]>
    Copyright (c) 2002-2016 Karl Waclawek <[email protected]>
-   Copyright (c) 2016-2022 Sebastian Pipping <[email protected]>
+   Copyright (c) 2016-2024 Sebastian Pipping <[email protected]>
    Copyright (c) 2016      Cristian Rodríguez <[email protected]>
    Copyright (c) 2016      Thomas Beutlich <[email protected]>
    Copyright (c) 2017      Rhodri James <[email protected]>
    Copyright (c) 2022      Thijs Schreijer <[email protected]>
+   Copyright (c) 2023      Hanno Böck <[email protected]>
+   Copyright (c) 2023      Sony Corporation / Snild Dolkow <[email protected]>
    Licensed under the MIT license:
 
    Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -269,7 +271,7 @@ XML_ParserCreate_MM(const XML_Char *encoding,
                     const XML_Memory_Handling_Suite *memsuite,
                     const XML_Char *namespaceSeparator);
 
-/* Prepare a parser object to be re-used.  This is particularly
+/* Prepare a parser object to be reused.  This is particularly
    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
@@ -951,7 +953,7 @@ XMLPARSEAPI(XML_Index) XML_GetCurrentByteIndex(XML_Parser parser);
 XMLPARSEAPI(int)
 XML_GetCurrentByteCount(XML_Parser parser);
 
-/* If XML_CONTEXT_BYTES is defined, returns the input buffer, sets
+/* If XML_CONTEXT_BYTES is >=1, returns the input buffer, sets
    the integer pointed to by offset to the offset within this buffer
    of the current parse position, and sets the integer pointed to by size
    to the size of this buffer (the number of input bytes). Otherwise
@@ -1025,7 +1027,9 @@ enum XML_FeatureEnum {
   XML_FEATURE_ATTR_INFO,
   /* Added in Expat 2.4.0. */
   XML_FEATURE_BILLION_LAUGHS_ATTACK_PROTECTION_MAXIMUM_AMPLIFICATION_DEFAULT,
-  XML_FEATURE_BILLION_LAUGHS_ATTACK_PROTECTION_ACTIVATION_THRESHOLD_DEFAULT
+  XML_FEATURE_BILLION_LAUGHS_ATTACK_PROTECTION_ACTIVATION_THRESHOLD_DEFAULT,
+  /* Added in Expat 2.6.0. */
+  XML_FEATURE_GE
   /* Additional features must be added to the end of this enum. */
 };
 
@@ -1038,23 +1042,29 @@ typedef struct {
 XMLPARSEAPI(const XML_Feature *)
 XML_GetFeatureList(void);
 
-#ifdef XML_DTD
-/* Added in Expat 2.4.0. */
+#if XML_GE == 1
+/* Added in Expat 2.4.0 for XML_DTD defined and
+ * added in Expat 2.6.0 for XML_GE == 1. */
 XMLPARSEAPI(XML_Bool)
 XML_SetBillionLaughsAttackProtectionMaximumAmplification(
     XML_Parser parser, float maximumAmplificationFactor);
 
-/* Added in Expat 2.4.0. */
+/* Added in Expat 2.4.0 for XML_DTD defined and
+ * added in Expat 2.6.0 for XML_GE == 1. */
 XMLPARSEAPI(XML_Bool)
 XML_SetBillionLaughsAttackProtectionActivationThreshold(
     XML_Parser parser, unsigned long long activationThresholdBytes);
 #endif
 
+/* Added in Expat 2.6.0. */
+XMLPARSEAPI(XML_Bool)
+XML_SetReparseDeferralEnabled(XML_Parser parser, XML_Bool enabled);
+
 /* Expat follows the semantic versioning convention.
-   See http://semver.org.
+   See https://semver.org
 */
 #define XML_MAJOR_VERSION 2
-#define XML_MINOR_VERSION 5
+#define XML_MINOR_VERSION 6
 #define XML_MICRO_VERSION 0
 
 #ifdef __cplusplus

Modules/expat/expat_config.h

@@ -16,6 +16,7 @@
 
 #define XML_NS 1
 #define XML_DTD 1
+#define XML_GE 1
 #define XML_CONTEXT_BYTES 1024
 
 #endif /* EXPAT_CONFIG_H */

Modules/expat/internal.h

@@ -28,9 +28,10 @@
    Copyright (c) 2002-2003 Fred L. Drake, Jr. <[email protected]>
    Copyright (c) 2002-2006 Karl Waclawek <[email protected]>
    Copyright (c) 2003      Greg Stein <[email protected]>
-   Copyright (c) 2016-2022 Sebastian Pipping <[email protected]>
+   Copyright (c) 2016-2023 Sebastian Pipping <[email protected]>
    Copyright (c) 2018      Yury Gribov <[email protected]>
    Copyright (c) 2019      David Loffredo <[email protected]>
+   Copyright (c) 2023      Sony Corporation / Snild Dolkow <[email protected]>
    Licensed under the MIT license:
 
    Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -154,12 +155,15 @@ extern "C" {
 void _INTERNAL_trim_to_complete_utf8_characters(const char *from,
                                                 const char **fromLimRef);
 
-#if defined(XML_DTD)
+#if XML_GE == 1
 unsigned long long testingAccountingGetCountBytesDirect(XML_Parser parser);
 unsigned long long testingAccountingGetCountBytesIndirect(XML_Parser parser);
 const char *unsignedCharToPrintable(unsigned char c);
 #endif
 
+extern XML_Bool g_reparseDeferralEnabledDefault; // written ONLY in runtests.c
+extern unsigned int g_parseAttempts;             // used for testing only
+
 #ifdef __cplusplus
 }
 #endif

Modules/expat/siphash.h

@@ -106,7 +106,7 @@
  * if this code is included and compiled as C++; related GCC warning is:
  * warning: use of C++11 long long integer constant [-Wlong-long]
  */
-#define _SIP_ULL(high, low) ((((uint64_t)high) << 32) | (low))
+#define SIP_ULL(high, low) ((((uint64_t)high) << 32) | (low))
 
 #define SIP_ROTL(x, b) (uint64_t)(((x) << (b)) | ((x) >> (64 - (b))))
 
@@ -190,10 +190,10 @@ sip_round(struct siphash *H, const int rounds) {
 
 static struct siphash *
 sip24_init(struct siphash *H, const struct sipkey *key) {
-  H->v0 = _SIP_ULL(0x736f6d65U, 0x70736575U) ^ key->k[0];
-  H->v1 = _SIP_ULL(0x646f7261U, 0x6e646f6dU) ^ key->k[1];
-  H->v2 = _SIP_ULL(0x6c796765U, 0x6e657261U) ^ key->k[0];
-  H->v3 = _SIP_ULL(0x74656462U, 0x79746573U) ^ key->k[1];
+  H->v0 = SIP_ULL(0x736f6d65U, 0x70736575U) ^ key->k[0];
+  H->v1 = SIP_ULL(0x646f7261U, 0x6e646f6dU) ^ key->k[1];
+  H->v2 = SIP_ULL(0x6c796765U, 0x6e657261U) ^ key->k[0];
+  H->v3 = SIP_ULL(0x74656462U, 0x79746573U) ^ key->k[1];
 
   H->p = H->buf;
   H->c = 0;

Modules/expat/winconfig.h

@@ -9,7 +9,8 @@
    Copyright (c) 2000      Clark Cooper <[email protected]>
    Copyright (c) 2002      Greg Stein <[email protected]>
    Copyright (c) 2005      Karl Waclawek <[email protected]>
-   Copyright (c) 2017-2021 Sebastian Pipping <[email protected]>
+   Copyright (c) 2017-2023 Sebastian Pipping <[email protected]>
+   Copyright (c) 2023      Orgad Shaneh <[email protected]>
    Licensed under the MIT license:
 
    Permission is  hereby granted,  free of charge,  to any  person obtaining
@@ -35,7 +36,9 @@
 #ifndef WINCONFIG_H
 #define WINCONFIG_H
 
-#define WIN32_LEAN_AND_MEAN
+#ifndef WIN32_LEAN_AND_MEAN
+#  define WIN32_LEAN_AND_MEAN
+#endif
 #include <windows.h>
 #undef WIN32_LEAN_AND_MEAN