https://github.com/libexpat/libexpat/labels/help%20wanted
If you can help, please get in touch. Thanks!
+Release 2.5.0 Tue October 25 2022
+ Security fixes:
+ #616 #649 #650 CVE-2022-43680 -- Fix heap use-after-free after overeager
+ destruction of a shared DTD in function
+ XML_ExternalEntityParserCreate in out-of-memory situations.
+ Expected impact is denial of service or potentially
+ arbitrary code execution.
+
+ Bug fixes:
+ #612 #645 Fix curruption from undefined entities
+ #613 #654 Fix case when parsing was suspended while processing nested
+ entities
+ #616 #652 #653 Stop leaking opening tag bindings after a closing tag
+ mismatch error where a parser is reset through
+ XML_ParserReset and then reused to parse
+ #656 CMake: Fix generation of pkg-config file
+ #658 MinGW|CMake: Fix static library name
+
+ Other changes:
+ #663 Protect header expat_config.h from multiple inclusion
+ #666 examples: Make use of XML_GetBuffer and be more
+ consistent across examples
+ #648 Address compiler warnings
+ #667 #668 Version info bumped from 9:9:8 to 9:10:8;
+ see https://verbump.de/ for what these numbers do
+
+ Special thanks to:
+ Jann Horn
+ Mark Brand
+ Osyotr
+ Rhodri James
+ and
+ Google Project Zero
+
Release 2.4.9 Tue September 20 2022
Security fixes:
#629 #640 CVE-2022-40674 -- Heap use-after-free vulnerability in
[](https://github.com/libexpat/libexpat/releases)
-# Expat, Release 2.4.9
+# Expat, Release 2.5.0
This is Expat, a C library for parsing XML, started by
[James Clark](https://en.wikipedia.org/wiki/James_Clark_%28programmer%29) in 1997.
<div>
<h1>
The Expat XML Parser
- <small>Release 2.4.9</small>
+ <small>Release 2.5.0</small>
</h1>
</div>
<div class="content">
Copyright (c) 2001-2003 Fred L. Drake, Jr. <fdrake@users.sourceforge.net>
Copyright (c) 2004-2006 Karl Waclawek <karl@waclawek.net>
Copyright (c) 2005-2007 Steven Solie <steven@solie.ca>
- Copyright (c) 2016-2019 Sebastian Pipping <sebastian@pipping.org>
+ Copyright (c) 2016-2022 Sebastian Pipping <sebastian@pipping.org>
Copyright (c) 2017 Rhodri James <rhodri@wildebeest.org.uk>
Copyright (c) 2019 Zhongyuan Zhou <zhouzhongyuan@huawei.com>
Licensed under the MIT license:
#endif
#ifdef XML_UNICODE_WCHAR_T
-# include <wchar.h>
# define XML_FMT_STR "ls"
#else
# define XML_FMT_STR "s"
static void XMLCALL
startElement(void *userData, const XML_Char *name, const XML_Char **atts) {
int i;
- int *depthPtr = (int *)userData;
+ int *const depthPtr = (int *)userData;
(void)atts;
for (i = 0; i < *depthPtr; i++)
static void XMLCALL
endElement(void *userData, const XML_Char *name) {
- int *depthPtr = (int *)userData;
+ int *const depthPtr = (int *)userData;
(void)name;
*depthPtr -= 1;
}
int
-main(int argc, char *argv[]) {
- char buf[BUFSIZ];
+main(void) {
XML_Parser parser = XML_ParserCreate(NULL);
int done;
int depth = 0;
- (void)argc;
- (void)argv;
+
+ if (! parser) {
+ fprintf(stderr, "Couldn't allocate memory for parser\n");
+ return 1;
+ }
XML_SetUserData(parser, &depth);
XML_SetElementHandler(parser, startElement, endElement);
+
do {
- size_t len = fread(buf, 1, sizeof(buf), stdin);
- done = len < sizeof(buf);
- if (XML_Parse(parser, buf, (int)len, done) == XML_STATUS_ERROR) {
- fprintf(stderr, "%" XML_FMT_STR " at line %" XML_FMT_INT_MOD "u\n",
- XML_ErrorString(XML_GetErrorCode(parser)),
- XML_GetCurrentLineNumber(parser));
+ void *const buf = XML_GetBuffer(parser, BUFSIZ);
+ if (! buf) {
+ fprintf(stderr, "Couldn't allocate memory for buffer\n");
+ XML_ParserFree(parser);
+ return 1;
+ }
+
+ const size_t len = fread(buf, 1, BUFSIZ, stdin);
+
+ if (ferror(stdin)) {
+ fprintf(stderr, "Read error\n");
+ XML_ParserFree(parser);
+ return 1;
+ }
+
+ done = feof(stdin);
+
+ if (XML_ParseBuffer(parser, (int)len, done) == XML_STATUS_ERROR) {
+ fprintf(stderr,
+ "Parse error at line %" XML_FMT_INT_MOD "u:\n%" XML_FMT_STR "\n",
+ XML_GetCurrentLineNumber(parser),
+ XML_ErrorString(XML_GetErrorCode(parser)));
XML_ParserFree(parser);
return 1;
}
} while (! done);
+
XML_ParserFree(parser);
return 0;
}
Copyright (c) 2001-2003 Fred L. Drake, Jr. <fdrake@users.sourceforge.net>
Copyright (c) 2005-2007 Steven Solie <steven@solie.ca>
Copyright (c) 2005-2006 Karl Waclawek <karl@waclawek.net>
- Copyright (c) 2016-2019 Sebastian Pipping <sebastian@pipping.org>
+ Copyright (c) 2016-2022 Sebastian Pipping <sebastian@pipping.org>
Copyright (c) 2017 Rhodri James <rhodri@wildebeest.org.uk>
Licensed under the MIT license:
# define XML_FMT_STR "s"
#endif
-#define BUFFSIZE 8192
-
-char Buff[BUFFSIZE];
-
-int Depth;
-
static void XMLCALL
-start(void *data, const XML_Char *el, const XML_Char **attr) {
+startElement(void *userData, const XML_Char *name, const XML_Char **atts) {
int i;
- (void)data;
+ int *const depthPtr = (int *)userData;
- for (i = 0; i < Depth; i++)
+ for (i = 0; i < *depthPtr; i++)
printf(" ");
- printf("%" XML_FMT_STR, el);
+ printf("%" XML_FMT_STR, name);
- for (i = 0; attr[i]; i += 2) {
- printf(" %" XML_FMT_STR "='%" XML_FMT_STR "'", attr[i], attr[i + 1]);
+ for (i = 0; atts[i]; i += 2) {
+ printf(" %" XML_FMT_STR "='%" XML_FMT_STR "'", atts[i], atts[i + 1]);
}
printf("\n");
- Depth++;
+ *depthPtr += 1;
}
static void XMLCALL
-end(void *data, const XML_Char *el) {
- (void)data;
- (void)el;
+endElement(void *userData, const XML_Char *name) {
+ int *const depthPtr = (int *)userData;
+ (void)name;
- Depth--;
+ *depthPtr -= 1;
}
int
-main(int argc, char *argv[]) {
- XML_Parser p = XML_ParserCreate(NULL);
- (void)argc;
- (void)argv;
+main(void) {
+ XML_Parser parser = XML_ParserCreate(NULL);
+ int done;
+ int depth = 0;
- if (! p) {
+ if (! parser) {
fprintf(stderr, "Couldn't allocate memory for parser\n");
- exit(-1);
+ return 1;
}
- XML_SetElementHandler(p, start, end);
+ XML_SetUserData(parser, &depth);
+ XML_SetElementHandler(parser, startElement, endElement);
- for (;;) {
- int done;
- int len;
+ do {
+ void *const buf = XML_GetBuffer(parser, BUFSIZ);
+ if (! buf) {
+ fprintf(stderr, "Couldn't allocate memory for buffer\n");
+ XML_ParserFree(parser);
+ return 1;
+ }
+
+ const size_t len = fread(buf, 1, BUFSIZ, stdin);
- len = (int)fread(Buff, 1, BUFFSIZE, stdin);
if (ferror(stdin)) {
fprintf(stderr, "Read error\n");
- exit(-1);
+ XML_ParserFree(parser);
+ return 1;
}
+
done = feof(stdin);
- if (XML_Parse(p, Buff, len, done) == XML_STATUS_ERROR) {
+ if (XML_ParseBuffer(parser, (int)len, done) == XML_STATUS_ERROR) {
fprintf(stderr,
"Parse error at line %" XML_FMT_INT_MOD "u:\n%" XML_FMT_STR "\n",
- XML_GetCurrentLineNumber(p),
- XML_ErrorString(XML_GetErrorCode(p)));
- exit(-1);
+ XML_GetCurrentLineNumber(parser),
+ XML_ErrorString(XML_GetErrorCode(parser)));
+ XML_ParserFree(parser);
+ return 1;
}
+ } while (! done);
- if (done)
- break;
- }
- XML_ParserFree(p);
+ XML_ParserFree(parser);
return 0;
}
See http://semver.org.
*/
#define XML_MAJOR_VERSION 2
-#define XML_MINOR_VERSION 4
-#define XML_MICRO_VERSION 9
+#define XML_MINOR_VERSION 5
+#define XML_MICRO_VERSION 0
#ifdef __cplusplus
}
-/* 90815a2b2c80c03b2b889fe1d427bb2b9e3282aa065e42784e001db4f23de324 (2.4.9+)
+/* 5ab094ffadd6edfc94c3eee53af44a86951f9f1f0933ada3114bbce2bfb02c99 (2.5.0+)
__ __ _
___\ \/ /_ __ __ _| |_
/ _ \\ /| '_ \ / _` | __|
Copyright (c) 2021 Dong-hee Na <donghee.na@python.org>
Copyright (c) 2022 Samanta Navarro <ferivoz@riseup.net>
Copyright (c) 2022 Jeffrey Walton <noloader@gmail.com>
+ Copyright (c) 2022 Jann Horn <jannh@google.com>
Licensed under the MIT license:
Permission is hereby granted, free of charge, to any person obtaining
parserInit(parser, encodingName);
if (encodingName && ! parser->m_protocolEncodingName) {
+ if (dtd) {
+ // We need to stop the upcoming call to XML_ParserFree from happily
+ // destroying parser->m_dtd because the DTD is shared with the parent
+ // parser and the only guard that keeps XML_ParserFree from destroying
+ // parser->m_dtd is parser->m_isParamEntity but it will be set to
+ // XML_TRUE only later in XML_ExternalEntityParserCreate (or not at all).
+ parser->m_dtd = NULL;
+ }
XML_ParserFree(parser);
return NULL;
}
int len;
const char *rawName;
TAG *tag = parser->m_tagStack;
- parser->m_tagStack = tag->parent;
- tag->parent = parser->m_freeTagList;
- parser->m_freeTagList = tag;
rawName = s + enc->minBytesPerChar * 2;
len = XmlNameLength(enc, rawName);
if (len != tag->rawNameLength
*eventPP = rawName;
return XML_ERROR_TAG_MISMATCH;
}
+ parser->m_tagStack = tag->parent;
+ tag->parent = parser->m_freeTagList;
+ parser->m_freeTagList = tag;
--parser->m_tagLevel;
if (parser->m_endElementHandler) {
const XML_Char *localPart;
parser->m_handlerArg, parser->m_declElementType->name,
parser->m_declAttributeId->name, parser->m_declAttributeType, 0,
role == XML_ROLE_REQUIRED_ATTRIBUTE_VALUE);
- poolClear(&parser->m_tempPool);
handleDefault = XML_FALSE;
}
}
+ poolClear(&parser->m_tempPool);
break;
case XML_ROLE_DEFAULT_ATTRIBUTE_VALUE:
case XML_ROLE_FIXED_ATTRIBUTE_VALUE:
*
* If 'standalone' is false, the DTD must have no
* parameter entities or we wouldn't have passed the outer
- * 'if' statement. That measn the only entity in the hash
+ * 'if' statement. That means the only entity in the hash
* table is the external subset name "#" which cannot be
* given as a parameter entity name in XML syntax, so the
* lookup must have returned NULL and we don't even reach
if (result != XML_ERROR_NONE)
return result;
- else if (textEnd != next
- && parser->m_parsingStatus.parsing == XML_SUSPENDED) {
+
+ if (textEnd != next && parser->m_parsingStatus.parsing == XML_SUSPENDED) {
entity->processed = (int)(next - (const char *)entity->textPtr);
return result;
- } else {
+ }
+
#ifdef XML_DTD
- entityTrackingOnClose(parser, entity, __LINE__);
+ entityTrackingOnClose(parser, entity, __LINE__);
#endif
- entity->open = XML_FALSE;
- parser->m_openInternalEntities = openEntity->next;
- /* put openEntity back in list of free instances */
- openEntity->next = parser->m_freeInternalEntities;
- parser->m_freeInternalEntities = openEntity;
+ entity->open = XML_FALSE;
+ parser->m_openInternalEntities = openEntity->next;
+ /* put openEntity back in list of free instances */
+ openEntity->next = parser->m_freeInternalEntities;
+ parser->m_freeInternalEntities = openEntity;
+
+ // If there are more open entities we want to stop right here and have the
+ // upcoming call to XML_ResumeParser continue with entity content, or it would
+ // be ignored altogether.
+ if (parser->m_openInternalEntities != NULL
+ && parser->m_parsingStatus.parsing == XML_SUSPENDED) {
+ return XML_ERROR_NONE;
}
#ifdef XML_DTD
BT_LF, /* line feed = "\n" */
BT_GT, /* greater than = ">" */
BT_QUOT, /* quotation character = "\"" */
- BT_APOS, /* aposthrophe = "'" */
+ BT_APOS, /* apostrophe = "'" */
BT_EQUALS, /* equal sign = "=" */
BT_QUEST, /* question mark = "?" */
BT_EXCL, /* exclamation mark = "!" */
Copyright (c) 2005-2007 Steven Solie <steven@solie.ca>
Copyright (c) 2005-2012 Karl Waclawek <karl@waclawek.net>
Copyright (c) 2016-2022 Sebastian Pipping <sebastian@pipping.org>
- Copyright (c) 2017-2018 Rhodri James <rhodri@wildebeest.org.uk>
+ Copyright (c) 2017-2022 Rhodri James <rhodri@wildebeest.org.uk>
Copyright (c) 2017 Joe Orton <jorton@redhat.com>
Copyright (c) 2017 José Gutiérrez de la Concha <jose@zeroc.com>
Copyright (c) 2018 Marco Maggi <marco.maggi-ipsu@poste.it>
}
END_TEST
-void
+static void XMLCALL
suspending_comment_handler(void *userData, const XML_Char *data) {
UNUSED_P(data);
XML_Parser parser = (XML_Parser)userData;
}
END_TEST
+/* Regression test for GH issue #612: unfinished m_declAttributeType
+ * allocation in ->m_tempPool can corrupt following allocation.
+ */
+static int XMLCALL
+external_entity_unfinished_attlist(XML_Parser parser, const XML_Char *context,
+ const XML_Char *base,
+ const XML_Char *systemId,
+ const XML_Char *publicId) {
+ const char *text = "<!ELEMENT barf ANY>\n"
+ "<!ATTLIST barf my_attr (blah|%blah;a|foo) #REQUIRED>\n"
+ "<!--COMMENT-->\n";
+ XML_Parser ext_parser;
+
+ UNUSED_P(base);
+ UNUSED_P(publicId);
+ if (systemId == NULL)
+ return XML_STATUS_OK;
+
+ ext_parser = XML_ExternalEntityParserCreate(parser, context, NULL);
+ if (ext_parser == NULL)
+ fail("Could not create external entity parser");
+
+ if (_XML_Parse_SINGLE_BYTES(ext_parser, text, (int)strlen(text), XML_TRUE)
+ == XML_STATUS_ERROR)
+ xml_failure(ext_parser);
+
+ XML_ParserFree(ext_parser);
+ return XML_STATUS_OK;
+}
+
+START_TEST(test_pool_integrity_with_unfinished_attr) {
+ const char *text = "<?xml version='1.0' encoding='UTF-8'?>\n"
+ "<!DOCTYPE foo [\n"
+ "<!ELEMENT foo ANY>\n"
+ "<!ENTITY % entp SYSTEM \"external.dtd\">\n"
+ "%entp;\n"
+ "]>\n"
+ "<a></a>\n";
+ const XML_Char *expected = XCS("COMMENT");
+ CharData storage;
+
+ CharData_Init(&storage);
+ XML_SetParamEntityParsing(g_parser, XML_PARAM_ENTITY_PARSING_ALWAYS);
+ XML_SetExternalEntityRefHandler(g_parser, external_entity_unfinished_attlist);
+ XML_SetAttlistDeclHandler(g_parser, dummy_attlist_decl_handler);
+ XML_SetCommentHandler(g_parser, accumulate_comment);
+ XML_SetUserData(g_parser, &storage);
+ if (_XML_Parse_SINGLE_BYTES(g_parser, text, (int)strlen(text), XML_TRUE)
+ == XML_STATUS_ERROR)
+ xml_failure(g_parser);
+ CharData_CheckXMLChars(&storage, expected);
+}
+END_TEST
+
+typedef struct {
+ XML_Parser parser;
+ CharData *storage;
+} ParserPlusStorage;
+
+static void XMLCALL
+accumulate_and_suspend_comment_handler(void *userData, const XML_Char *data) {
+ ParserPlusStorage *const parserPlusStorage = (ParserPlusStorage *)userData;
+ accumulate_comment(parserPlusStorage->storage, data);
+ XML_StopParser(parserPlusStorage->parser, XML_TRUE);
+}
+
+START_TEST(test_nested_entity_suspend) {
+ const char *const text = "<!DOCTYPE a [\n"
+ " <!ENTITY e1 '<!--e1-->'>\n"
+ " <!ENTITY e2 '<!--e2 head-->&e1;<!--e2 tail-->'>\n"
+ " <!ENTITY e3 '<!--e3 head-->&e2;<!--e3 tail-->'>\n"
+ "]>\n"
+ "<a><!--start-->&e3;<!--end--></a>";
+ const XML_Char *const expected = XCS("start") XCS("e3 head") XCS("e2 head")
+ XCS("e1") XCS("e2 tail") XCS("e3 tail") XCS("end");
+ CharData storage;
+ XML_Parser parser = XML_ParserCreate(NULL);
+ ParserPlusStorage parserPlusStorage = {parser, &storage};
+
+ CharData_Init(&storage);
+ XML_SetParamEntityParsing(parser, XML_PARAM_ENTITY_PARSING_ALWAYS);
+ XML_SetCommentHandler(parser, accumulate_and_suspend_comment_handler);
+ XML_SetUserData(parser, &parserPlusStorage);
+
+ enum XML_Status status = XML_Parse(parser, text, (int)strlen(text), XML_TRUE);
+ while (status == XML_STATUS_SUSPENDED) {
+ status = XML_ResumeParser(parser);
+ }
+ if (status != XML_STATUS_OK)
+ xml_failure(parser);
+
+ CharData_CheckXMLChars(&storage, expected);
+ XML_ParserFree(parser);
+}
+END_TEST
+
/*
* Namespaces tests.
*/
fail("Version mismatch");
#if ! defined(XML_UNICODE) || defined(XML_UNICODE_WCHAR_T)
- if (xcstrcmp(version_text, XCS("expat_2.4.9"))) /* needs bump on releases */
+ if (xcstrcmp(version_text, XCS("expat_2.5.0"))) /* needs bump on releases */
fail("XML_*_VERSION in expat.h out of sync?\n");
#else
/* If we have XML_UNICODE defined but not XML_UNICODE_WCHAR_T
}
END_TEST
+START_TEST(test_misc_tag_mismatch_reset_leak) {
+#ifdef XML_NS
+ const char *const text = "<open xmlns='https://namespace1.test'></close>";
+ XML_Parser parser = XML_ParserCreateNS(NULL, XCS('\n'));
+
+ if (XML_Parse(parser, text, (int)strlen(text), XML_TRUE) != XML_STATUS_ERROR)
+ fail("Call to parse was expected to fail");
+ if (XML_GetErrorCode(parser) != XML_ERROR_TAG_MISMATCH)
+ fail("Call to parse was expected to fail from a closing tag mismatch");
+
+ XML_ParserReset(parser, NULL);
+
+ if (XML_Parse(parser, text, (int)strlen(text), XML_TRUE) != XML_STATUS_ERROR)
+ fail("Call to parse was expected to fail");
+ if (XML_GetErrorCode(parser) != XML_ERROR_TAG_MISMATCH)
+ fail("Call to parse was expected to fail from a closing tag mismatch");
+
+ XML_ParserFree(parser);
+#endif
+}
+END_TEST
+
static void
alloc_setup(void) {
XML_Memory_Handling_Suite memsuite = {duff_allocator, duff_reallocator, free};
}
END_TEST
+static int XMLCALL
+external_entity_parser_create_alloc_fail_handler(XML_Parser parser,
+ const XML_Char *context,
+ const XML_Char *base,
+ const XML_Char *systemId,
+ const XML_Char *publicId) {
+ UNUSED_P(base);
+ UNUSED_P(systemId);
+ UNUSED_P(publicId);
+
+ if (context != NULL)
+ fail("Unexpected non-NULL context");
+
+ // The following number intends to fail the upcoming allocation in line
+ // "parser->m_protocolEncodingName = copyString(encodingName,
+ // &(parser->m_mem));" in function parserInit.
+ allocation_count = 3;
+
+ const XML_Char *const encodingName = XCS("UTF-8"); // needs something non-NULL
+ const XML_Parser ext_parser
+ = XML_ExternalEntityParserCreate(parser, context, encodingName);
+ if (ext_parser != NULL)
+ fail(
+ "Call to XML_ExternalEntityParserCreate was expected to fail out-of-memory");
+
+ allocation_count = ALLOC_ALWAYS_SUCCEED;
+ return XML_STATUS_ERROR;
+}
+
+START_TEST(test_alloc_reset_after_external_entity_parser_create_fail) {
+ const char *const text = "<!DOCTYPE doc SYSTEM 'foo'><doc/>";
+
+ XML_SetExternalEntityRefHandler(
+ g_parser, external_entity_parser_create_alloc_fail_handler);
+ XML_SetParamEntityParsing(g_parser, XML_PARAM_ENTITY_PARSING_ALWAYS);
+
+ if (XML_Parse(g_parser, text, (int)strlen(text), XML_TRUE)
+ != XML_STATUS_ERROR)
+ fail("Call to parse was expected to fail");
+
+ if (XML_GetErrorCode(g_parser) != XML_ERROR_EXTERNAL_ENTITY_HANDLING)
+ fail("Call to parse was expected to fail from the external entity handler");
+
+ XML_ParserReset(g_parser, NULL);
+}
+END_TEST
+
static void
nsalloc_setup(void) {
XML_Memory_Handling_Suite memsuite = {duff_allocator, duff_reallocator, free};
tcase_add_test(tc_basic, test_bad_notation);
tcase_add_test(tc_basic, test_default_doctype_handler);
tcase_add_test(tc_basic, test_empty_element_abort);
+ tcase_add_test__ifdef_xml_dtd(tc_basic,
+ test_pool_integrity_with_unfinished_attr);
+ tcase_add_test(tc_basic, test_nested_entity_suspend);
suite_add_tcase(s, tc_namespace);
tcase_add_checked_fixture(tc_namespace, namespace_setup, namespace_teardown);
tcase_add_test(tc_misc, test_misc_stop_during_end_handler_issue_240_2);
tcase_add_test__ifdef_xml_dtd(
tc_misc, test_misc_deny_internal_entity_closing_doctype_issue_317);
+ tcase_add_test(tc_misc, test_misc_tag_mismatch_reset_leak);
suite_add_tcase(s, tc_alloc);
tcase_add_checked_fixture(tc_alloc, alloc_setup, alloc_teardown);
tcase_add_test(tc_alloc, test_alloc_long_public_id);
tcase_add_test(tc_alloc, test_alloc_long_entity_value);
tcase_add_test(tc_alloc, test_alloc_long_notation);
+ tcase_add_test__ifdef_xml_dtd(
+ tc_alloc, test_alloc_reset_after_external_entity_parser_create_fail);
suite_add_tcase(s, tc_nsalloc);
tcase_add_checked_fixture(tc_nsalloc, nsalloc_setup, nsalloc_teardown);
projects / openbsd / commitdiff