Merge pull request #890 from libexpat/taiyou-xml-parsebuffer-len

[CVE-2024-45490] lib: Reject negative len for `XML_ParseBuffer` (fixes #887)

expat/doc/reference.html

@@ -1135,7 +1135,9 @@ containing part (or perhaps all) of the document. The number of bytes of s
 that are part of the document is indicated by <code>len</code>. This means
 that <code>s</code> doesn't have to be null-terminated. It also means that
 if <code>len</code> is larger than the number of bytes in the block of
-memory that <code>s</code> points at, then a memory fault is likely. The
+memory that <code>s</code> points at, then a memory fault is likely.
+Negative values for <code>len</code> are rejected since Expat 2.2.1.
+The
 <code>isFinal</code> parameter informs the parser that this is the last
 piece of the document. Frequently, the last piece is empty (i.e.
 <code>len</code> is zero.)
@@ -1183,11 +1185,17 @@ XML_ParseBuffer(XML_Parser p,
                 int isFinal);
 </pre>
 <div class="fcndef">
+<p>
 This is just like <code><a href= "#XML_Parse" >XML_Parse</a></code>,
 except in this case Expat provides the buffer.  By obtaining the
 buffer from Expat with the <code><a href= "#XML_GetBuffer"
 >XML_GetBuffer</a></code> function, the application can avoid double
 copying of the input.
+</p>
+
+<p>
+Negative values for <code>len</code> are rejected since Expat 2.6.3.
+</p>
 </div>
 
 <h4 id="XML_GetBuffer">XML_GetBuffer</h4>

expat/lib/xmlparse.c

@@ -2038,6 +2038,12 @@ XML_ParseBuffer(XML_Parser parser, int len, int isFinal) {
 
   if (parser == NULL)
     return XML_STATUS_ERROR;
+
+  if (len < 0) {
+    parser->m_errorCode = XML_ERROR_INVALID_ARGUMENT;
+    return XML_STATUS_ERROR;
+  }
+
   switch (parser->m_parsingStatus.parsing) {
   case XML_SUSPENDED:
     parser->m_errorCode = XML_ERROR_SUSPENDED;

expat/tests/basic_tests.c

@@ -2804,6 +2804,61 @@ START_TEST(test_empty_parse) {
 }
 END_TEST
 
+/* Test XML_Parse for len < 0 */
+START_TEST(test_negative_len_parse) {
+  const char *const doc = "<root/>";
+  for (int isFinal = 0; isFinal < 2; isFinal++) {
+    set_subtest("isFinal=%d", isFinal);
+
+    XML_Parser parser = XML_ParserCreate(NULL);
+
+    if (XML_GetErrorCode(parser) != XML_ERROR_NONE)
+      fail("There was not supposed to be any initial parse error.");
+
+    const enum XML_Status status = XML_Parse(parser, doc, -1, isFinal);
+
+    if (status != XML_STATUS_ERROR)
+      fail("Negative len was expected to fail the parse but did not.");
+
+    if (XML_GetErrorCode(parser) != XML_ERROR_INVALID_ARGUMENT)
+      fail("Parse error does not match XML_ERROR_INVALID_ARGUMENT.");
+
+    XML_ParserFree(parser);
+  }
+}
+END_TEST
+
+/* Test XML_ParseBuffer for len < 0 */
+START_TEST(test_negative_len_parse_buffer) {
+  const char *const doc = "<root/>";
+  for (int isFinal = 0; isFinal < 2; isFinal++) {
+    set_subtest("isFinal=%d", isFinal);
+
+    XML_Parser parser = XML_ParserCreate(NULL);
+
+    if (XML_GetErrorCode(parser) != XML_ERROR_NONE)
+      fail("There was not supposed to be any initial parse error.");
+
+    void *const buffer = XML_GetBuffer(parser, (int)strlen(doc));
+
+    if (buffer == NULL)
+      fail("XML_GetBuffer failed.");
+
+    memcpy(buffer, doc, strlen(doc));
+
+    const enum XML_Status status = XML_ParseBuffer(parser, -1, isFinal);
+
+    if (status != XML_STATUS_ERROR)
+      fail("Negative len was expected to fail the parse but did not.");
+
+    if (XML_GetErrorCode(parser) != XML_ERROR_INVALID_ARGUMENT)
+      fail("Parse error does not match XML_ERROR_INVALID_ARGUMENT.");
+
+    XML_ParserFree(parser);
+  }
+}
+END_TEST
+
 /* Test odd corners of the XML_GetBuffer interface */
 static enum XML_Status
 get_feature(enum XML_FeatureEnum feature_id, long *presult) {
@@ -5955,6 +6010,8 @@ make_basic_test_case(Suite *s) {
   tcase_add_test__ifdef_xml_dtd(tc_basic, test_user_parameters);
   tcase_add_test__ifdef_xml_dtd(tc_basic, test_ext_entity_ref_parameter);
   tcase_add_test(tc_basic, test_empty_parse);
+  tcase_add_test(tc_basic, test_negative_len_parse);
+  tcase_add_test(tc_basic, test_negative_len_parse_buffer);
   tcase_add_test(tc_basic, test_get_buffer_1);
   tcase_add_test(tc_basic, test_get_buffer_2);
 #if XML_CONTEXT_BYTES > 0