# XXX TypeErrors on calling handlers, or on bad return values from a # handler, are obscure and unhelpful. import abc import functools import os import re import sys import sysconfig import textwrap import unittest import traceback from io import BytesIO from test import support from test.support import import_helper, os_helper from test.support import sortdict from unittest import mock from xml.parsers import expat from xml.parsers.expat import errors class SetAttributeTest(unittest.TestCase): def setUp(self): self.parser = expat.ParserCreate(namespace_separator='!') def test_buffer_text(self): self.assertIs(self.parser.buffer_text, False) for x in 0, 1, 2, 0: self.parser.buffer_text = x self.assertIs(self.parser.buffer_text, bool(x)) def test_namespace_prefixes(self): self.assertIs(self.parser.namespace_prefixes, False) for x in 0, 1, 2, 0: self.parser.namespace_prefixes = x self.assertIs(self.parser.namespace_prefixes, bool(x)) def test_ordered_attributes(self): self.assertIs(self.parser.ordered_attributes, False) for x in 0, 1, 2, 0: self.parser.ordered_attributes = x self.assertIs(self.parser.ordered_attributes, bool(x)) def test_specified_attributes(self): self.assertIs(self.parser.specified_attributes, False) for x in 0, 1, 2, 0: self.parser.specified_attributes = x self.assertIs(self.parser.specified_attributes, bool(x)) def test_invalid_attributes(self): with self.assertRaises(AttributeError): self.parser.returns_unicode = 1 with self.assertRaises(AttributeError): self.parser.returns_unicode # Issue #25019 self.assertRaises(TypeError, setattr, self.parser, range(0xF), 0) self.assertRaises(TypeError, self.parser.__setattr__, range(0xF), 0) self.assertRaises(TypeError, getattr, self.parser, range(0xF)) data = b'''\ %unparsed_entity; ]> Contents of subelements &external_entity; &skipped_entity; \xb5 ''' # Produce UTF-8 output class ParseTest(unittest.TestCase): class Outputter: def __init__(self): self.out = [] def StartElementHandler(self, name, attrs): self.out.append('Start element: ' + repr(name) + ' ' + sortdict(attrs)) def EndElementHandler(self, name): self.out.append('End element: ' + repr(name)) def CharacterDataHandler(self, data): data = data.strip() if data: self.out.append('Character data: ' + repr(data)) def ProcessingInstructionHandler(self, target, data): self.out.append('PI: ' + repr(target) + ' ' + repr(data)) def StartNamespaceDeclHandler(self, prefix, uri): self.out.append('NS decl: ' + repr(prefix) + ' ' + repr(uri)) def EndNamespaceDeclHandler(self, prefix): self.out.append('End of NS decl: ' + repr(prefix)) def StartCdataSectionHandler(self): self.out.append('Start of CDATA section') def EndCdataSectionHandler(self): self.out.append('End of CDATA section') def CommentHandler(self, text): self.out.append('Comment: ' + repr(text)) def NotationDeclHandler(self, *args): name, base, sysid, pubid = args self.out.append('Notation declared: %s' %(args,)) def UnparsedEntityDeclHandler(self, *args): entityName, base, systemId, publicId, notationName = args self.out.append('Unparsed entity decl: %s' %(args,)) def NotStandaloneHandler(self): self.out.append('Not standalone') return 1 def ExternalEntityRefHandler(self, *args): context, base, sysId, pubId = args self.out.append('External entity ref: %s' %(args[1:],)) return 1 def StartDoctypeDeclHandler(self, *args): self.out.append(('Start doctype', args)) return 1 def EndDoctypeDeclHandler(self): self.out.append("End doctype") return 1 def EntityDeclHandler(self, *args): self.out.append(('Entity declaration', args)) return 1 def XmlDeclHandler(self, *args): self.out.append(('XML declaration', args)) return 1 def ElementDeclHandler(self, *args): self.out.append(('Element declaration', args)) return 1 def AttlistDeclHandler(self, *args): self.out.append(('Attribute list declaration', args)) return 1 def SkippedEntityHandler(self, *args): self.out.append(("Skipped entity", args)) return 1 def DefaultHandler(self, userData): pass def DefaultHandlerExpand(self, userData): pass handler_names = [ 'StartElementHandler', 'EndElementHandler', 'CharacterDataHandler', 'ProcessingInstructionHandler', 'UnparsedEntityDeclHandler', 'NotationDeclHandler', 'StartNamespaceDeclHandler', 'EndNamespaceDeclHandler', 'CommentHandler', 'StartCdataSectionHandler', 'EndCdataSectionHandler', 'DefaultHandler', 'DefaultHandlerExpand', 'NotStandaloneHandler', 'ExternalEntityRefHandler', 'StartDoctypeDeclHandler', 'EndDoctypeDeclHandler', 'EntityDeclHandler', 'XmlDeclHandler', 'ElementDeclHandler', 'AttlistDeclHandler', 'SkippedEntityHandler', ] def _hookup_callbacks(self, parser, handler): """ Set each of the callbacks defined on handler and named in self.handler_names on the given parser. """ for name in self.handler_names: setattr(parser, name, getattr(handler, name)) def _verify_parse_output(self, operations): expected_operations = [ ('XML declaration', ('1.0', 'iso-8859-1', 0)), 'PI: \'xml-stylesheet\' \'href="stylesheet.css"\'', "Comment: ' comment data '", "Not standalone", ("Start doctype", ('quotations', 'quotations.dtd', None, 1)), ('Element declaration', ('root', (2, 0, None, ()))), ('Attribute list declaration', ('root', 'attr1', 'CDATA', None, 1)), ('Attribute list declaration', ('root', 'attr2', 'CDATA', None, 0)), "Notation declared: ('notation', None, 'notation.jpeg', None)", ('Entity declaration', ('acirc', 0, '\xe2', None, None, None, None)), ('Entity declaration', ('external_entity', 0, None, None, 'entity.file', None, None)), "Unparsed entity decl: ('unparsed_entity', None, 'entity.file', None, 'notation')", "Not standalone", "End doctype", "Start element: 'root' {'attr1': 'value1', 'attr2': 'value2\u1f40'}", "NS decl: 'myns' 'http://www.python.org/namespace'", "Start element: 'http://www.python.org/namespace!subelement' {}", "Character data: 'Contents of subelements'", "End element: 'http://www.python.org/namespace!subelement'", "End of NS decl: 'myns'", "Start element: 'sub2' {}", 'Start of CDATA section', "Character data: 'contents of CDATA section'", 'End of CDATA section', "End element: 'sub2'", "External entity ref: (None, 'entity.file', None)", ('Skipped entity', ('skipped_entity', 0)), "Character data: '\xb5'", "End element: 'root'", ] for operation, expected_operation in zip(operations, expected_operations): self.assertEqual(operation, expected_operation) def test_parse_bytes(self): out = self.Outputter() parser = expat.ParserCreate(namespace_separator='!') self._hookup_callbacks(parser, out) parser.Parse(data, True) operations = out.out self._verify_parse_output(operations) # Issue #6697. self.assertRaises(AttributeError, getattr, parser, '\uD800') def test_parse_str(self): out = self.Outputter() parser = expat.ParserCreate(namespace_separator='!') self._hookup_callbacks(parser, out) parser.Parse(data.decode('iso-8859-1'), True) operations = out.out self._verify_parse_output(operations) def test_parse_file(self): # Try parsing a file out = self.Outputter() parser = expat.ParserCreate(namespace_separator='!') self._hookup_callbacks(parser, out) file = BytesIO(data) parser.ParseFile(file) operations = out.out self._verify_parse_output(operations) def test_parse_again(self): parser = expat.ParserCreate() file = BytesIO(data) parser.ParseFile(file) # Issue 6676: ensure a meaningful exception is raised when attempting # to parse more than one XML document per xmlparser instance, # a limitation of the Expat library. with self.assertRaises(expat.error) as cm: parser.ParseFile(file) self.assertEqual(expat.ErrorString(cm.exception.code), expat.errors.XML_ERROR_FINISHED) class NamespaceSeparatorTest(unittest.TestCase): def test_legal(self): # Tests that make sure we get errors when the namespace_separator value # is illegal, and that we don't for good values: expat.ParserCreate() expat.ParserCreate(namespace_separator=None) expat.ParserCreate(namespace_separator=' ') def test_illegal(self): with self.assertRaisesRegex(TypeError, r"ParserCreate\(\) argument (2|'namespace_separator') " r"must be str or None, not int"): expat.ParserCreate(namespace_separator=42) try: expat.ParserCreate(namespace_separator='too long') self.fail() except ValueError as e: self.assertEqual(str(e), 'namespace_separator must be at most one character, omitted, or None') def test_zero_length(self): # ParserCreate() needs to accept a namespace_separator of zero length # to satisfy the requirements of RDF applications that are required # to simply glue together the namespace URI and the localname. Though # considered a wart of the RDF specifications, it needs to be supported. # # See XML-SIG mailing list thread starting with # http://mail.python.org/pipermail/xml-sig/2001-April/005202.html # expat.ParserCreate(namespace_separator='') # too short class InterningTest(unittest.TestCase): def test(self): # Test the interning machinery. p = expat.ParserCreate() L = [] def collector(name, *args): L.append(name) p.StartElementHandler = collector p.EndElementHandler = collector p.Parse(b" ", True) tag = L[0] self.assertEqual(len(L), 6) for entry in L: # L should have the same string repeated over and over. self.assertTrue(tag is entry) def test_issue9402(self): # create an ExternalEntityParserCreate with buffer text class ExternalOutputter: def __init__(self, parser): self.parser = parser self.parser_result = None def ExternalEntityRefHandler(self, context, base, sysId, pubId): external_parser = self.parser.ExternalEntityParserCreate("") self.parser_result = external_parser.Parse(b"", True) return 1 parser = expat.ParserCreate(namespace_separator='!') parser.buffer_text = 1 out = ExternalOutputter(parser) parser.ExternalEntityRefHandler = out.ExternalEntityRefHandler parser.Parse(data, True) self.assertEqual(out.parser_result, 1) class BufferTextTest(unittest.TestCase): def setUp(self): self.stuff = [] self.parser = expat.ParserCreate() self.parser.buffer_text = 1 self.parser.CharacterDataHandler = self.CharacterDataHandler def check(self, expected, label): self.assertEqual(self.stuff, expected, "%s\nstuff = %r\nexpected = %r" % (label, self.stuff, map(str, expected))) def CharacterDataHandler(self, text): self.stuff.append(text) def StartElementHandler(self, name, attrs): self.stuff.append("<%s>" % name) bt = attrs.get("buffer-text") if bt == "yes": self.parser.buffer_text = 1 elif bt == "no": self.parser.buffer_text = 0 def EndElementHandler(self, name): self.stuff.append("" % name) def CommentHandler(self, data): self.stuff.append("" % data) def setHandlers(self, handlers=[]): for name in handlers: setattr(self.parser, name, getattr(self, name)) def test_default_to_disabled(self): parser = expat.ParserCreate() self.assertFalse(parser.buffer_text) def test_buffering_enabled(self): # Make sure buffering is turned on self.assertTrue(self.parser.buffer_text) self.parser.Parse(b"123", True) self.assertEqual(self.stuff, ['123'], "buffered text not properly collapsed") def test1(self): # XXX This test exposes more detail of Expat's text chunking than we # XXX like, but it tests what we need to concisely. self.setHandlers(["StartElementHandler"]) self.parser.Parse(b"12\n34\n5", True) self.assertEqual(self.stuff, ["", "1", "", "2", "\n", "3", "", "4\n5"], "buffering control not reacting as expected") def test2(self): self.parser.Parse(b"1<2> \n 3", True) self.assertEqual(self.stuff, ["1<2> \n 3"], "buffered text not properly collapsed") def test3(self): self.setHandlers(["StartElementHandler"]) self.parser.Parse(b"123", True) self.assertEqual(self.stuff, ["", "1", "", "2", "", "3"], "buffered text not properly split") def test4(self): self.setHandlers(["StartElementHandler", "EndElementHandler"]) self.parser.CharacterDataHandler = None self.parser.Parse(b"123", True) self.assertEqual(self.stuff, ["", "", "", "", "", ""]) def test5(self): self.setHandlers(["StartElementHandler", "EndElementHandler"]) self.parser.Parse(b"123", True) self.assertEqual(self.stuff, ["", "1", "", "", "2", "", "", "3", ""]) def test6(self): self.setHandlers(["CommentHandler", "EndElementHandler", "StartElementHandler"]) self.parser.Parse(b"12345 ", True) self.assertEqual(self.stuff, ["", "1", "", "", "2", "", "", "345", ""], "buffered text not properly split") def test7(self): self.setHandlers(["CommentHandler", "EndElementHandler", "StartElementHandler"]) self.parser.Parse(b"12345 ", True) self.assertEqual(self.stuff, ["", "1", "", "", "2", "", "", "3", "", "4", "", "5", ""], "buffered text not properly split") def test_change_character_data_handler_in_callback(self): # Test that xmlparse_handler_setter() properly handles # the special case "parser.CharacterDataHandler = None". def handler(*args): parser.CharacterDataHandler = None handler_wrapper = mock.Mock(wraps=handler) parser = expat.ParserCreate() parser.CharacterDataHandler = handler_wrapper parser.Parse(b"12345 ", True) handler_wrapper.assert_called_once() self.assertIsNone(parser.CharacterDataHandler) # Test handling of exception from callback: class HandlerExceptionTest(unittest.TestCase): def StartElementHandler(self, name, attrs): raise RuntimeError(f'StartElementHandler: <{name}>') def check_traceback_entry(self, entry, filename, funcname): self.assertEqual(os.path.basename(entry.filename), filename) self.assertEqual(entry.name, funcname) @support.cpython_only def test_exception(self): # gh-66652: test _PyTraceback_Add() used by pyexpat.c to inject frames # Change the current directory to the Python source code directory # if it is available. src_dir = sysconfig.get_config_var('abs_builddir') if src_dir: have_source = os.path.isdir(src_dir) else: have_source = False if have_source: with os_helper.change_cwd(src_dir): self._test_exception(have_source) else: self._test_exception(have_source) def _test_exception(self, have_source): # Use path relative to the current directory which should be the Python # source code directory (if it is available). PYEXPAT_C = os.path.join('Modules', 'pyexpat.c') parser = expat.ParserCreate() parser.StartElementHandler = self.StartElementHandler try: parser.Parse(b"", True) self.fail("the parser did not raise RuntimeError") except RuntimeError as exc: self.assertEqual(exc.args[0], 'StartElementHandler: ', exc) entries = traceback.extract_tb(exc.__traceback__) self.assertEqual(len(entries), 3, entries) self.check_traceback_entry(entries[0], "test_pyexpat.py", "_test_exception") self.check_traceback_entry(entries[1], os.path.basename(PYEXPAT_C), "StartElement") self.check_traceback_entry(entries[2], "test_pyexpat.py", "StartElementHandler") # Check that the traceback contains the relevant line in # Modules/pyexpat.c. Skip the test if Modules/pyexpat.c is not # available. if have_source and os.path.exists(PYEXPAT_C): self.assertIn('call_with_frame("StartElement"', entries[1].line) # Test Current* members: class PositionTest(unittest.TestCase): def StartElementHandler(self, name, attrs): self.check_pos('s') def EndElementHandler(self, name): self.check_pos('e') def check_pos(self, event): pos = (event, self.parser.CurrentByteIndex, self.parser.CurrentLineNumber, self.parser.CurrentColumnNumber) self.assertTrue(self.upto < len(self.expected_list), 'too many parser events') expected = self.expected_list[self.upto] self.assertEqual(pos, expected, 'Expected position %s, got position %s' %(pos, expected)) self.upto += 1 def test(self): self.parser = expat.ParserCreate() self.parser.StartElementHandler = self.StartElementHandler self.parser.EndElementHandler = self.EndElementHandler self.upto = 0 self.expected_list = [('s', 0, 1, 0), ('s', 5, 2, 1), ('s', 11, 3, 2), ('e', 15, 3, 6), ('e', 17, 4, 1), ('e', 22, 5, 0)] xml = b'\n \n \n \n' self.parser.Parse(xml, True) class sf1296433Test(unittest.TestCase): def test_parse_only_xml_data(self): # https://bugs.python.org/issue1296433 # xml = "%s" % ('a' * 1025) # this one doesn't crash #xml = "%s" % ('a' * 10000) class SpecificException(Exception): pass def handler(text): raise SpecificException parser = expat.ParserCreate() parser.CharacterDataHandler = handler self.assertRaises(SpecificException, parser.Parse, xml.encode('iso8859')) class ChardataBufferTest(unittest.TestCase): """ test setting of chardata buffer size """ def test_1025_bytes(self): self.assertEqual(self.small_buffer_test(1025), 2) def test_1000_bytes(self): self.assertEqual(self.small_buffer_test(1000), 1) def test_wrong_size(self): parser = expat.ParserCreate() parser.buffer_text = 1 with self.assertRaises(ValueError): parser.buffer_size = -1 with self.assertRaises(ValueError): parser.buffer_size = 0 with self.assertRaises((ValueError, OverflowError)): parser.buffer_size = sys.maxsize + 1 with self.assertRaises(TypeError): parser.buffer_size = 512.0 def test_unchanged_size(self): xml1 = b"" + b'a' * 512 xml2 = b'a'*512 + b'' parser = expat.ParserCreate() parser.CharacterDataHandler = self.counting_handler parser.buffer_size = 512 parser.buffer_text = 1 # Feed 512 bytes of character data: the handler should be called # once. self.n = 0 parser.Parse(xml1) self.assertEqual(self.n, 1) # Reassign to buffer_size, but assign the same size. parser.buffer_size = parser.buffer_size self.assertEqual(self.n, 1) # Try parsing rest of the document parser.Parse(xml2) self.assertEqual(self.n, 2) def test_disabling_buffer(self): xml1 = b"" + b'a' * 512 xml2 = b'b' * 1024 xml3 = b'c' * 1024 + b'' parser = expat.ParserCreate() parser.CharacterDataHandler = self.counting_handler parser.buffer_text = 1 parser.buffer_size = 1024 self.assertEqual(parser.buffer_size, 1024) # Parse one chunk of XML self.n = 0 parser.Parse(xml1, False) self.assertEqual(parser.buffer_size, 1024) self.assertEqual(self.n, 1) # Turn off buffering and parse the next chunk. parser.buffer_text = 0 self.assertFalse(parser.buffer_text) self.assertEqual(parser.buffer_size, 1024) for i in range(10): parser.Parse(xml2, False) self.assertEqual(self.n, 11) parser.buffer_text = 1 self.assertTrue(parser.buffer_text) self.assertEqual(parser.buffer_size, 1024) parser.Parse(xml3, True) self.assertEqual(self.n, 12) def counting_handler(self, text): self.n += 1 def small_buffer_test(self, buffer_len): xml = b"" + b'a' * buffer_len + b'' parser = expat.ParserCreate() parser.CharacterDataHandler = self.counting_handler parser.buffer_size = 1024 parser.buffer_text = 1 self.n = 0 parser.Parse(xml) return self.n def test_change_size_1(self): xml1 = b"" + b'a' * 1024 xml2 = b'aaa' + b'a' * 1025 + b'' parser = expat.ParserCreate() parser.CharacterDataHandler = self.counting_handler parser.buffer_text = 1 parser.buffer_size = 1024 self.assertEqual(parser.buffer_size, 1024) self.n = 0 parser.Parse(xml1, False) parser.buffer_size *= 2 self.assertEqual(parser.buffer_size, 2048) parser.Parse(xml2, True) self.assertEqual(self.n, 2) def test_change_size_2(self): xml1 = b"a" + b'a' * 1023 xml2 = b'aaa' + b'a' * 1025 + b'' parser = expat.ParserCreate() parser.CharacterDataHandler = self.counting_handler parser.buffer_text = 1 parser.buffer_size = 2048 self.assertEqual(parser.buffer_size, 2048) self.n=0 parser.Parse(xml1, False) parser.buffer_size = parser.buffer_size // 2 self.assertEqual(parser.buffer_size, 1024) parser.Parse(xml2, True) self.assertEqual(self.n, 4) class MalformedInputTest(unittest.TestCase): def test1(self): xml = b"\0\r\n" parser = expat.ParserCreate() try: parser.Parse(xml, True) self.fail() except expat.ExpatError as e: self.assertEqual(str(e), 'unclosed token: line 2, column 0') def test2(self): # \xc2\x85 is UTF-8 encoded U+0085 (NEXT LINE) xml = b"\r\n" parser = expat.ParserCreate() err_pattern = r'XML declaration not well-formed: line 1, column \d+' with self.assertRaisesRegex(expat.ExpatError, err_pattern): parser.Parse(xml, True) class ErrorMessageTest(unittest.TestCase): def test_codes(self): # verify mapping of errors.codes and errors.messages self.assertEqual(errors.XML_ERROR_SYNTAX, errors.messages[errors.codes[errors.XML_ERROR_SYNTAX]]) def test_expaterror(self): xml = b'<' parser = expat.ParserCreate() try: parser.Parse(xml, True) self.fail() except expat.ExpatError as e: self.assertEqual(e.code, errors.codes[errors.XML_ERROR_UNCLOSED_TOKEN]) class ForeignDTDTests(unittest.TestCase): """ Tests for the UseForeignDTD method of expat parser objects. """ def test_use_foreign_dtd(self): """ If UseForeignDTD is passed True and a document without an external entity reference is parsed, ExternalEntityRefHandler is first called with None for the public and system ids. """ handler_call_args = [] def resolve_entity(context, base, system_id, public_id): handler_call_args.append((public_id, system_id)) return 1 parser = expat.ParserCreate() parser.UseForeignDTD(True) parser.SetParamEntityParsing(expat.XML_PARAM_ENTITY_PARSING_ALWAYS) parser.ExternalEntityRefHandler = resolve_entity parser.Parse(b"") self.assertEqual(handler_call_args, [(None, None)]) # test UseForeignDTD() is equal to UseForeignDTD(True) handler_call_args[:] = [] parser = expat.ParserCreate() parser.UseForeignDTD() parser.SetParamEntityParsing(expat.XML_PARAM_ENTITY_PARSING_ALWAYS) parser.ExternalEntityRefHandler = resolve_entity parser.Parse(b"") self.assertEqual(handler_call_args, [(None, None)]) def test_ignore_use_foreign_dtd(self): """ If UseForeignDTD is passed True and a document with an external entity reference is parsed, ExternalEntityRefHandler is called with the public and system ids from the document. """ handler_call_args = [] def resolve_entity(context, base, system_id, public_id): handler_call_args.append((public_id, system_id)) return 1 parser = expat.ParserCreate() parser.UseForeignDTD(True) parser.SetParamEntityParsing(expat.XML_PARAM_ENTITY_PARSING_ALWAYS) parser.ExternalEntityRefHandler = resolve_entity parser.Parse( b"") self.assertEqual(handler_call_args, [("bar", "baz")]) class ParentParserLifetimeTest(unittest.TestCase): """ Subparsers make use of their parent XML_Parser inside of Expat. As a result, parent parsers need to outlive subparsers. See https://github.com/python/cpython/issues/139400. """ def test_parent_parser_outlives_its_subparsers__single(self): parser = expat.ParserCreate() subparser = parser.ExternalEntityParserCreate(None) # Now try to cause garbage collection of the parent parser # while it's still being referenced by a related subparser. del parser def test_parent_parser_outlives_its_subparsers__multiple(self): parser = expat.ParserCreate() subparser_one = parser.ExternalEntityParserCreate(None) subparser_two = parser.ExternalEntityParserCreate(None) # Now try to cause garbage collection of the parent parser # while it's still being referenced by a related subparser. del parser def test_parent_parser_outlives_its_subparsers__chain(self): parser = expat.ParserCreate() subparser = parser.ExternalEntityParserCreate(None) subsubparser = subparser.ExternalEntityParserCreate(None) # Now try to cause garbage collection of the parent parsers # while they are still being referenced by a related subparser. del parser del subparser class ReparseDeferralTest(unittest.TestCase): def test_getter_setter_round_trip(self): parser = expat.ParserCreate() enabled = (expat.version_info >= (2, 6, 0)) self.assertIs(parser.GetReparseDeferralEnabled(), enabled) parser.SetReparseDeferralEnabled(False) self.assertIs(parser.GetReparseDeferralEnabled(), False) parser.SetReparseDeferralEnabled(True) self.assertIs(parser.GetReparseDeferralEnabled(), enabled) def test_reparse_deferral_enabled(self): if expat.version_info < (2, 6, 0): self.skipTest(f'Expat {expat.version_info} does not ' 'support reparse deferral') started = [] def start_element(name, _): started.append(name) parser = expat.ParserCreate() parser.StartElementHandler = start_element self.assertTrue(parser.GetReparseDeferralEnabled()) for chunk in (b''): parser.Parse(chunk, False) # The key test: Have handlers already fired? Expecting: no. self.assertEqual(started, []) parser.Parse(b'', True) self.assertEqual(started, ['doc']) def test_reparse_deferral_disabled(self): started = [] def start_element(name, _): started.append(name) parser = expat.ParserCreate() parser.StartElementHandler = start_element if expat.version_info >= (2, 6, 0): parser.SetReparseDeferralEnabled(False) self.assertFalse(parser.GetReparseDeferralEnabled()) for chunk in (b''): parser.Parse(chunk, False) # The key test: Have handlers already fired? Expecting: yes. self.assertEqual(started, ['doc']) class AttackProtectionTestBase(abc.ABC): """ Base class for testing protections against XML payloads with disproportionate amplification. The protections being tested should detect and prevent attacks that leverage disproportionate amplification from small inputs. """ @staticmethod def exponential_expansion_payload(*, nrows, ncols, text='.'): """Create a billion laughs attack payload. Be careful: the number of total items is pow(n, k), thereby requiring at least pow(ncols, nrows) * sizeof(text) memory! """ template = textwrap.dedent(f"""\ {{body}} ]> &row{nrows}; """).rstrip() body = '\n'.join( f'' for i in range(nrows) ) body = textwrap.indent(body, ' ' * 4) return template.format(body=body) def test_payload_generation(self): # self-test for exponential_expansion_payload() payload = self.exponential_expansion_payload(nrows=2, ncols=3) self.assertEqual(payload, textwrap.dedent("""\ ]> &row2; """).rstrip()) def assert_root_parser_failure(self, func, /, *args, **kwargs): """Check that func(*args, **kwargs) is invalid for a sub-parser.""" msg = "parser must be a root parser" self.assertRaisesRegex(expat.ExpatError, msg, func, *args, **kwargs) @abc.abstractmethod def assert_rejected(self, func, /, *args, **kwargs): """Assert that func(*args, **kwargs) triggers the attack protection. Note: this method must ensure that the attack protection being tested is the one that is actually triggered at runtime, e.g., by matching the exact error message. """ @abc.abstractmethod def set_activation_threshold(self, parser, threshold): """Set the activation threshold for the tested protection.""" @abc.abstractmethod def set_maximum_amplification(self, parser, max_factor): """Set the maximum amplification factor for the tested protection.""" @abc.abstractmethod def test_set_activation_threshold__threshold_reached(self): """Test when the activation threshold is exceeded.""" @abc.abstractmethod def test_set_activation_threshold__threshold_not_reached(self): """Test when the activation threshold is not exceeded.""" def test_set_activation_threshold__invalid_threshold_type(self): parser = expat.ParserCreate() setter = functools.partial(self.set_activation_threshold, parser) self.assertRaises(TypeError, setter, 1.0) self.assertRaises(TypeError, setter, -1.5) self.assertRaises(ValueError, setter, -5) def test_set_activation_threshold__invalid_threshold_range(self): _testcapi = import_helper.import_module("_testcapi") parser = expat.ParserCreate() setter = functools.partial(self.set_activation_threshold, parser) self.assertRaises(OverflowError, setter, _testcapi.ULLONG_MAX + 1) def test_set_activation_threshold__fail_for_subparser(self): parser = expat.ParserCreate() subparser = parser.ExternalEntityParserCreate(None) setter = functools.partial(self.set_activation_threshold, subparser) self.assert_root_parser_failure(setter, 12345) @abc.abstractmethod def test_set_maximum_amplification__amplification_exceeded(self): """Test when the amplification factor is exceeded.""" @abc.abstractmethod def test_set_maximum_amplification__amplification_not_exceeded(self): """Test when the amplification factor is not exceeded.""" def test_set_maximum_amplification__infinity(self): inf = float('inf') # an 'inf' threshold is allowed by Expat parser = expat.ParserCreate() self.assertIsNone(self.set_maximum_amplification(parser, inf)) def test_set_maximum_amplification__invalid_max_factor_type(self): parser = expat.ParserCreate() setter = functools.partial(self.set_maximum_amplification, parser) self.assertRaises(TypeError, setter, None) self.assertRaises(TypeError, setter, 'abc') def test_set_maximum_amplification__invalid_max_factor_range(self): parser = expat.ParserCreate() setter = functools.partial(self.set_maximum_amplification, parser) msg = re.escape("'max_factor' must be at least 1.0") self.assertRaisesRegex(expat.ExpatError, msg, setter, float('nan')) self.assertRaisesRegex(expat.ExpatError, msg, setter, 0.99) def test_set_maximum_amplification__fail_for_subparser(self): parser = expat.ParserCreate() subparser = parser.ExternalEntityParserCreate(None) setter = functools.partial(self.set_maximum_amplification, subparser) self.assert_root_parser_failure(setter, 123.45) @unittest.skipIf(expat.version_info < (2, 4, 0), "requires Expat >= 2.4.0") class ExpansionProtectionTest(AttackProtectionTestBase, unittest.TestCase): def assert_rejected(self, func, /, *args, **kwargs): """Check that func(*args, **kwargs) hits the allocation limit.""" msg = ( r"limit on input amplification factor \(from DTD and entities\) " r"breached: line \d+, column \d+" ) self.assertRaisesRegex(expat.ExpatError, msg, func, *args, **kwargs) def set_activation_threshold(self, parser, threshold): return parser.SetBillionLaughsAttackProtectionActivationThreshold(threshold) def set_maximum_amplification(self, parser, max_factor): return parser.SetBillionLaughsAttackProtectionMaximumAmplification(max_factor) def test_set_activation_threshold__threshold_reached(self): parser = expat.ParserCreate() # Choose a threshold expected to be always reached. self.set_activation_threshold(parser, 3) # Check that the threshold is reached by choosing a small factor # and a payload whose peak amplification factor exceeds it. self.assertIsNone(self.set_maximum_amplification(parser, 1.0)) payload = self.exponential_expansion_payload(ncols=10, nrows=4) self.assert_rejected(parser.Parse, payload, True) def test_set_activation_threshold__threshold_not_reached(self): parser = expat.ParserCreate() # Choose a threshold expected to be never reached. self.set_activation_threshold(parser, pow(10, 5)) # Check that the threshold is reached by choosing a small factor # and a payload whose peak amplification factor exceeds it. self.assertIsNone(self.set_maximum_amplification(parser, 1.0)) payload = self.exponential_expansion_payload(ncols=10, nrows=4) self.assertIsNotNone(parser.Parse(payload, True)) def test_set_maximum_amplification__amplification_exceeded(self): parser = expat.ParserCreate() # Unconditionally enable maximum activation factor. self.set_activation_threshold(parser, 0) # Choose a max amplification factor expected to always be exceeded. self.assertIsNone(self.set_maximum_amplification(parser, 1.0)) # Craft a payload for which the peak amplification factor is > 1.0. payload = self.exponential_expansion_payload(ncols=1, nrows=2) self.assert_rejected(parser.Parse, payload, True) def test_set_maximum_amplification__amplification_not_exceeded(self): parser = expat.ParserCreate() # Unconditionally enable maximum activation factor. self.set_activation_threshold(parser, 0) # Choose a max amplification factor expected to never be exceeded. self.assertIsNone(self.set_maximum_amplification(parser, 1e4)) # Craft a payload for which the peak amplification factor is < 1e4. payload = self.exponential_expansion_payload(ncols=1, nrows=2) self.assertIsNotNone(parser.Parse(payload, True)) @unittest.skipIf(expat.version_info < (2, 7, 2), "requires Expat >= 2.7.2") class MemoryProtectionTest(AttackProtectionTestBase, unittest.TestCase): # NOTE: with the default Expat configuration, the billion laughs protection # may hit before the allocation limiter if exponential_expansion_payload() # is not carefully parametrized. As such, the payloads should be chosen so # that either the allocation limiter is hit before other protections are # triggered or no protection at all is triggered. def assert_rejected(self, func, /, *args, **kwargs): """Check that func(*args, **kwargs) hits the allocation limit.""" msg = r"out of memory: line \d+, column \d+" self.assertRaisesRegex(expat.ExpatError, msg, func, *args, **kwargs) def set_activation_threshold(self, parser, threshold): return parser.SetAllocTrackerActivationThreshold(threshold) def set_maximum_amplification(self, parser, max_factor): return parser.SetAllocTrackerMaximumAmplification(max_factor) def test_set_activation_threshold__threshold_reached(self): parser = expat.ParserCreate() # Choose a threshold expected to be always reached. self.set_activation_threshold(parser, 3) # Check that the threshold is reached by choosing a small factor # and a payload whose peak amplification factor exceeds it. self.assertIsNone(self.set_maximum_amplification(parser, 1.0)) payload = self.exponential_expansion_payload(ncols=10, nrows=4) self.assert_rejected(parser.Parse, payload, True) def test_set_activation_threshold__threshold_not_reached(self): parser = expat.ParserCreate() # Choose a threshold expected to be never reached. self.set_activation_threshold(parser, pow(10, 5)) # Check that the threshold is reached by choosing a small factor # and a payload whose peak amplification factor exceeds it. self.assertIsNone(self.set_maximum_amplification(parser, 1.0)) payload = self.exponential_expansion_payload(ncols=10, nrows=4) self.assertIsNotNone(parser.Parse(payload, True)) def test_set_maximum_amplification__amplification_exceeded(self): parser = expat.ParserCreate() # Unconditionally enable maximum activation factor. self.set_activation_threshold(parser, 0) # Choose a max amplification factor expected to always be exceeded. self.assertIsNone(self.set_maximum_amplification(parser, 1.0)) # Craft a payload for which the peak amplification factor is > 1.0. payload = self.exponential_expansion_payload(ncols=1, nrows=2) self.assert_rejected(parser.Parse, payload, True) def test_set_maximum_amplification__amplification_not_exceeded(self): parser = expat.ParserCreate() # Unconditionally enable maximum activation factor. self.set_activation_threshold(parser, 0) # Choose a max amplification factor expected to never be exceeded. self.assertIsNone(self.set_maximum_amplification(parser, 1e4)) # Craft a payload for which the peak amplification factor is < 1e4. payload = self.exponential_expansion_payload(ncols=1, nrows=2) self.assertIsNotNone(parser.Parse(payload, True)) if __name__ == "__main__": unittest.main()