\xe3'), 'latin-1')
b"\\n\\xe3?>"
"""
#
# xinclude tests (samples from appendix C of the xinclude specification)
XINCLUDE = {}
XINCLUDE["C1.xml"] = """\
120 Mz is adequate for an average home user.
"""
XINCLUDE["disclaimer.xml"] = """\
The opinions represented herein represent those of the individual
and should not be interpreted as official policy endorsed by this
organization.
"""
XINCLUDE["C2.xml"] = """\
This document has been accessed
times.
"""
XINCLUDE["count.txt"] = "324387"
XINCLUDE["C2b.xml"] = """\
This document has been accessed
times.
"""
XINCLUDE["C3.xml"] = """\
The following is the source of the "data.xml" resource:
"""
XINCLUDE["data.xml"] = """\
Example.
""".format(html.escape(SIMPLE_XMLFILE, True))
def xinclude_loader(href, parse="xml", encoding=None):
try:
data = XINCLUDE[href]
except KeyError:
raise OSError("resource not found")
if parse == "xml":
from xml.etree.ElementTree import XML
return XML(data)
return data
def xinclude():
r"""
Basic inclusion example (XInclude C.1)
>>> from xml.etree import ElementInclude
>>> document = xinclude_loader("C1.xml")
>>> ElementInclude.include(document, xinclude_loader)
>>> print(serialize(document)) # C1
120 Mz is adequate for an average home user.
The opinions represented herein represent those of the individual
and should not be interpreted as official policy endorsed by this
organization.
Textual inclusion example (XInclude C.2)
>>> document = xinclude_loader("C2.xml")
>>> ElementInclude.include(document, xinclude_loader)
>>> print(serialize(document)) # C2
This document has been accessed
324387 times.
Textual inclusion after sibling element (based on modified XInclude C.2)
>>> document = xinclude_loader("C2b.xml")
>>> ElementInclude.include(document, xinclude_loader)
>>> print(serialize(document)) # C2b
This document has been accessed
324387 times.
Textual inclusion of XML example (XInclude C.3)
>>> document = xinclude_loader("C3.xml")
>>> ElementInclude.include(document, xinclude_loader)
>>> print(serialize(document)) # C3
The following is the source of the "data.xml" resource:
<?xml version='1.0'?>
<data>
<item><![CDATA[Brooks & Shields]]></item>
</data>
Fallback example (XInclude C.5)
Note! Fallback support is not yet implemented
>>> document = xinclude_loader("C5.xml")
>>> ElementInclude.include(document, xinclude_loader)
Traceback (most recent call last):
OSError: resource not found
>>> # print(serialize(document)) # C5
"""
def xinclude_default():
"""
>>> from xml.etree import ElementInclude
>>> document = xinclude_loader("default.xml")
>>> ElementInclude.include(document)
>>> print(serialize(document)) # default
Example.
text
texttail
"""
#
# badly formatted xi:include tags
XINCLUDE_BAD = {}
XINCLUDE_BAD["B1.xml"] = """\
120 Mz is adequate for an average home user.
"""
XINCLUDE_BAD["B2.xml"] = """\
"""
def xinclude_failures():
r"""
Test failure to locate included XML file.
>>> from xml.etree import ElementInclude
>>> def none_loader(href, parser, encoding=None):
... return None
>>> document = ET.XML(XINCLUDE["C1.xml"])
>>> ElementInclude.include(document, loader=none_loader)
Traceback (most recent call last):
xml.etree.ElementInclude.FatalIncludeError: cannot load 'disclaimer.xml' as 'xml'
Test failure to locate included text file.
>>> document = ET.XML(XINCLUDE["C2.xml"])
>>> ElementInclude.include(document, loader=none_loader)
Traceback (most recent call last):
xml.etree.ElementInclude.FatalIncludeError: cannot load 'count.txt' as 'text'
Test bad parse type.
>>> document = ET.XML(XINCLUDE_BAD["B1.xml"])
>>> ElementInclude.include(document, loader=none_loader)
Traceback (most recent call last):
xml.etree.ElementInclude.FatalIncludeError: unknown parse type in xi:include tag ('BAD_TYPE')
Test xi:fallback outside xi:include.
>>> document = ET.XML(XINCLUDE_BAD["B2.xml"])
>>> ElementInclude.include(document, loader=none_loader)
Traceback (most recent call last):
xml.etree.ElementInclude.FatalIncludeError: xi:fallback tag must be child of xi:include ('{http://www.w3.org/2001/XInclude}fallback')
"""
# --------------------------------------------------------------------
# reported bugs
def bug_xmltoolkit21():
"""
marshaller gives obscure errors for non-string values
>>> elem = ET.Element(123)
>>> serialize(elem) # tag
Traceback (most recent call last):
TypeError: cannot serialize 123 (type int)
>>> elem = ET.Element("elem")
>>> elem.text = 123
>>> serialize(elem) # text
Traceback (most recent call last):
TypeError: cannot serialize 123 (type int)
>>> elem = ET.Element("elem")
>>> elem.tail = 123
>>> serialize(elem) # tail
Traceback (most recent call last):
TypeError: cannot serialize 123 (type int)
>>> elem = ET.Element("elem")
>>> elem.set(123, "123")
>>> serialize(elem) # attribute key
Traceback (most recent call last):
TypeError: cannot serialize 123 (type int)
>>> elem = ET.Element("elem")
>>> elem.set("123", 123)
>>> serialize(elem) # attribute value
Traceback (most recent call last):
TypeError: cannot serialize 123 (type int)
"""
def bug_xmltoolkit25():
"""
typo in ElementTree.findtext
>>> elem = ET.XML(SAMPLE_XML)
>>> tree = ET.ElementTree(elem)
>>> tree.findtext("tag")
'text'
>>> tree.findtext("section/tag")
'subtext'
"""
def bug_xmltoolkit28():
"""
.//tag causes exceptions
>>> tree = ET.XML("")
>>> summarize_list(tree.findall(".//thead"))
[]
>>> summarize_list(tree.findall(".//tbody"))
['tbody']
"""
def bug_xmltoolkitX1():
"""
dump() doesn't flush the output buffer
>>> tree = ET.XML("")
>>> ET.dump(tree); print("tail")
tail
"""
def bug_xmltoolkit39():
"""
non-ascii element and attribute names doesn't work
>>> tree = ET.XML(b"")
>>> ET.tostring(tree, "utf-8")
b''
>>> tree = ET.XML(b"")
>>> tree.attrib
{'\\xe4ttr': 'v\\xe4lue'}
>>> ET.tostring(tree, "utf-8")
b''
>>> tree = ET.XML(b"text")
>>> ET.tostring(tree, "utf-8")
b'text'
>>> tree = ET.Element("t\u00e4g")
>>> ET.tostring(tree, "utf-8")
b''
>>> tree = ET.Element("tag")
>>> tree.set("\u00e4ttr", "v\u00e4lue")
>>> ET.tostring(tree, "utf-8")
b''
"""
def bug_xmltoolkit54():
"""
problems handling internally defined entities
>>> e = ET.XML("]>&ldots;")
>>> serialize(e, encoding="us-ascii")
b'舰'
>>> serialize(e)
'\u8230'
"""
def bug_xmltoolkit55():
"""
make sure we're reporting the first error, not the last
>>> normalize_exception(ET.XML, b"&ldots;&ndots;&rdots;")
Traceback (most recent call last):
ParseError: undefined entity &ldots;: line 1, column 36
"""
class ExceptionFile:
def read(self, x):
raise OSError
def xmltoolkit60():
"""
Handle crash in stream source.
>>> tree = ET.parse(ExceptionFile())
Traceback (most recent call last):
OSError
"""
XMLTOOLKIT62_DOC = """
A new cultivar of Begonia plant named ‘BCT9801BEG’.
"""
def xmltoolkit62():
"""
Don't crash when using custom entities.
>>> xmltoolkit62()
'A new cultivar of Begonia plant named \u2018BCT9801BEG\u2019.'
"""
ENTITIES = {'rsquo': '\u2019', 'lsquo': '\u2018'}
parser = ET.XMLTreeBuilder()
parser.entity.update(ENTITIES)
parser.feed(XMLTOOLKIT62_DOC)
t = parser.close()
return t.find('.//paragraph').text
def xmltoolkit63():
"""
Check reference leak.
>>> xmltoolkit63()
>>> count = sys.getrefcount(None)
>>> for i in range(1000):
... xmltoolkit63()
>>> sys.getrefcount(None) - count
0
"""
tree = ET.TreeBuilder()
tree.start("tag", {})
tree.data("text")
tree.end("tag")
# --------------------------------------------------------------------
def bug_200708_newline():
r"""
Preserve newlines in attributes.
>>> e = ET.Element('SomeTag', text="def _f():\n return 3\n")
>>> ET.tostring(e)
b''
>>> ET.XML(ET.tostring(e)).get("text")
'def _f():\n return 3\n'
>>> ET.tostring(ET.XML(ET.tostring(e)))
b''
"""
def bug_200708_close():
"""
Test default builder.
>>> parser = ET.XMLParser() # default
>>> parser.feed("some text")
>>> summarize(parser.close())
'element'
Test custom builder.
>>> class EchoTarget:
... def close(self):
... return ET.Element("element") # simulate root
>>> parser = ET.XMLParser(EchoTarget())
>>> parser.feed("some text")
>>> summarize(parser.close())
'element'
"""
def bug_200709_default_namespace():
"""
>>> e = ET.Element("{default}elem")
>>> s = ET.SubElement(e, "{default}elem")
>>> serialize(e, default_namespace="default") # 1
''
>>> e = ET.Element("{default}elem")
>>> s = ET.SubElement(e, "{default}elem")
>>> s = ET.SubElement(e, "{not-default}elem")
>>> serialize(e, default_namespace="default") # 2
''
>>> e = ET.Element("{default}elem")
>>> s = ET.SubElement(e, "{default}elem")
>>> s = ET.SubElement(e, "elem") # unprefixed name
>>> serialize(e, default_namespace="default") # 3
Traceback (most recent call last):
ValueError: cannot use non-qualified names with default_namespace option
"""
def bug_200709_register_namespace():
"""
>>> ET.tostring(ET.Element("{http://namespace.invalid/does/not/exist/}title"))
b''
>>> ET.register_namespace("foo", "http://namespace.invalid/does/not/exist/")
>>> ET.tostring(ET.Element("{http://namespace.invalid/does/not/exist/}title"))
b''
And the Dublin Core namespace is in the default list:
>>> ET.tostring(ET.Element("{http://purl.org/dc/elements/1.1/}title"))
b''
"""
def bug_200709_element_comment():
"""
Not sure if this can be fixed, really (since the serializer needs
ET.Comment, not cET.comment).
>>> a = ET.Element('a')
>>> a.append(ET.Comment('foo'))
>>> a[0].tag == ET.Comment
True
>>> a = ET.Element('a')
>>> a.append(ET.PI('foo'))
>>> a[0].tag == ET.PI
True
"""
def bug_200709_element_insert():
"""
>>> a = ET.Element('a')
>>> b = ET.SubElement(a, 'b')
>>> c = ET.SubElement(a, 'c')
>>> d = ET.Element('d')
>>> a.insert(0, d)
>>> summarize_list(a)
['d', 'b', 'c']
>>> a.insert(-1, d)
>>> summarize_list(a)
['d', 'b', 'd', 'c']
"""
def bug_200709_iter_comment():
"""
>>> a = ET.Element('a')
>>> b = ET.SubElement(a, 'b')
>>> comment_b = ET.Comment("TEST-b")
>>> b.append(comment_b)
>>> summarize_list(a.iter(ET.Comment))
['']
"""
# --------------------------------------------------------------------
# reported on bugs.python.org
def bug_1534630():
"""
>>> bob = ET.TreeBuilder()
>>> e = bob.data("data")
>>> e = bob.start("tag", {})
>>> e = bob.end("tag")
>>> e = bob.close()
>>> serialize(e)
''
"""
def check_issue6233():
"""
>>> e = ET.XML(b"t\\xc3\\xa3g")
>>> ET.tostring(e, 'ascii')
b"\\ntãg"
>>> e = ET.XML(b"t\\xe3g")
>>> ET.tostring(e, 'ascii')
b"\\ntãg"
"""
def check_issue3151():
"""
>>> e = ET.XML('')
>>> e.tag
'{${stuff}}localname'
>>> t = ET.ElementTree(e)
>>> ET.tostring(e)
b''
"""
def check_issue6565():
"""
>>> elem = ET.XML("")
>>> summarize_list(elem)
['tag']
>>> newelem = ET.XML(SAMPLE_XML)
>>> elem[:] = newelem[:]
>>> summarize_list(elem)
['tag', 'tag', 'section']
"""
def check_issue10777():
"""
Registering a namespace twice caused a "dictionary changed size during
iteration" bug.
>>> ET.register_namespace('test10777', 'http://myuri/')
>>> ET.register_namespace('test10777', 'http://myuri/')
"""
# --------------------------------------------------------------------
class ElementTreeTest(unittest.TestCase):
def test_istype(self):
self.assertIsInstance(ET.ParseError, type)
self.assertIsInstance(ET.QName, type)
self.assertIsInstance(ET.ElementTree, type)
self.assertIsInstance(ET.Element, type)
# XXX issue 14128 with C ElementTree
# self.assertIsInstance(ET.TreeBuilder, type)
# self.assertIsInstance(ET.XMLParser, type)
def test_Element_subclass_trivial(self):
class MyElement(ET.Element):
pass
mye = MyElement('foo')
self.assertIsInstance(mye, ET.Element)
self.assertIsInstance(mye, MyElement)
self.assertEqual(mye.tag, 'foo')
def test_Element_subclass_constructor(self):
class MyElement(ET.Element):
def __init__(self, tag, attrib={}, **extra):
super(MyElement, self).__init__(tag + '__', attrib, **extra)
mye = MyElement('foo', {'a': 1, 'b': 2}, c=3, d=4)
self.assertEqual(mye.tag, 'foo__')
self.assertEqual(sorted(mye.items()),
[('a', 1), ('b', 2), ('c', 3), ('d', 4)])
def test_Element_subclass_new_method(self):
class MyElement(ET.Element):
def newmethod(self):
return self.tag
mye = MyElement('joe')
self.assertEqual(mye.newmethod(), 'joe')
class TreeBuilderTest(unittest.TestCase):
sample1 = (''
'text')
def test_dummy_builder(self):
class BaseDummyBuilder:
def close(self):
return 42
class DummyBuilder(BaseDummyBuilder):
data = start = end = lambda *a: None
parser = ET.XMLParser(target=DummyBuilder())
parser.feed(self.sample1)
self.assertEqual(parser.close(), 42)
parser = ET.XMLParser(target=BaseDummyBuilder())
parser.feed(self.sample1)
self.assertEqual(parser.close(), 42)
parser = ET.XMLParser(target=object())
parser.feed(self.sample1)
self.assertIsNone(parser.close())
# XXX in _elementtree, the constructor of TreeBuilder expects no
# arguments
@unittest.expectedFailure
def test_element_factory(self):
tb = ET.TreeBuilder(element_factory=lambda: ET.Element())
@unittest.expectedFailure # XXX issue 14007 with C ElementTree
def test_doctype(self):
class DoctypeParser:
_doctype = None
def doctype(self, name, pubid, system):
self._doctype = (name, pubid, system)
def close(self):
return self._doctype
parser = ET.XMLParser(target=DoctypeParser())
parser.feed(self.sample1)
self.assertEqual(parser.close(),
('html', '-//W3C//DTD XHTML 1.0 Transitional//EN',
'http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd'))
class NoAcceleratorTest(unittest.TestCase):
# Test that the C accelerator was not imported for pyET
def test_correct_import_pyET(self):
self.assertEqual(pyET.Element.__module__, 'xml.etree.ElementTree')
self.assertEqual(pyET.SubElement.__module__, 'xml.etree.ElementTree')
class ElementSlicingTest(unittest.TestCase):
def _elem_tags(self, elemlist):
return [e.tag for e in elemlist]
def _subelem_tags(self, elem):
return self._elem_tags(list(elem))
def _make_elem_with_children(self, numchildren):
"""Create an Element with a tag 'a', with the given amount of children
named 'a0', 'a1' ... and so on.
"""
e = ET.Element('a')
for i in range(numchildren):
ET.SubElement(e, 'a%s' % i)
return e
def test_getslice_single_index(self):
e = self._make_elem_with_children(10)
self.assertEqual(e[1].tag, 'a1')
self.assertEqual(e[-2].tag, 'a8')
self.assertRaises(IndexError, lambda: e[12])
def test_getslice_range(self):
e = self._make_elem_with_children(6)
self.assertEqual(self._elem_tags(e[3:]), ['a3', 'a4', 'a5'])
self.assertEqual(self._elem_tags(e[3:6]), ['a3', 'a4', 'a5'])
self.assertEqual(self._elem_tags(e[3:16]), ['a3', 'a4', 'a5'])
self.assertEqual(self._elem_tags(e[3:5]), ['a3', 'a4'])
self.assertEqual(self._elem_tags(e[3:-1]), ['a3', 'a4'])
self.assertEqual(self._elem_tags(e[:2]), ['a0', 'a1'])
def test_getslice_steps(self):
e = self._make_elem_with_children(10)
self.assertEqual(self._elem_tags(e[8:10:1]), ['a8', 'a9'])
self.assertEqual(self._elem_tags(e[::3]), ['a0', 'a3', 'a6', 'a9'])
self.assertEqual(self._elem_tags(e[::8]), ['a0', 'a8'])
self.assertEqual(self._elem_tags(e[1::8]), ['a1', 'a9'])
def test_getslice_negative_steps(self):
e = self._make_elem_with_children(4)
self.assertEqual(self._elem_tags(e[::-1]), ['a3', 'a2', 'a1', 'a0'])
self.assertEqual(self._elem_tags(e[::-2]), ['a3', 'a1'])
def test_delslice(self):
e = self._make_elem_with_children(4)
del e[0:2]
self.assertEqual(self._subelem_tags(e), ['a2', 'a3'])
e = self._make_elem_with_children(4)
del e[0:]
self.assertEqual(self._subelem_tags(e), [])
e = self._make_elem_with_children(4)
del e[::-1]
self.assertEqual(self._subelem_tags(e), [])
e = self._make_elem_with_children(4)
del e[::-2]
self.assertEqual(self._subelem_tags(e), ['a0', 'a2'])
e = self._make_elem_with_children(4)
del e[1::2]
self.assertEqual(self._subelem_tags(e), ['a0', 'a2'])
e = self._make_elem_with_children(2)
del e[::2]
self.assertEqual(self._subelem_tags(e), ['a1'])
class StringIOTest(unittest.TestCase):
def test_read_from_stringio(self):
tree = ET.ElementTree()
stream = io.StringIO()
stream.write('''''')
stream.seek(0)
tree.parse(stream)
self.assertEqual(tree.getroot().tag, 'site')
class ParseErrorTest(unittest.TestCase):
def test_subclass(self):
self.assertIsInstance(ET.ParseError(), SyntaxError)
def _get_error(self, s):
try:
ET.fromstring(s)
except ET.ParseError as e:
return e
def test_error_position(self):
self.assertEqual(self._get_error('foo').position, (1, 0))
self.assertEqual(self._get_error('&foo;').position, (1, 5))
self.assertEqual(self._get_error('foobar<').position, (1, 6))
def test_error_code(self):
import xml.parsers.expat.errors as ERRORS
self.assertEqual(self._get_error('foo').code,
ERRORS.codes[ERRORS.XML_ERROR_SYNTAX])
# --------------------------------------------------------------------
class CleanContext(object):
"""Provide default namespace mapping and path cache."""
checkwarnings = None
def __init__(self, quiet=False):
if sys.flags.optimize >= 2:
# under -OO, doctests cannot be run and therefore not all warnings
# will be emitted
quiet = True
deprecations = (
# Search behaviour is broken if search path starts with "/".
("This search is broken in 1.3 and earlier, and will be fixed "
"in a future version. If you rely on the current behaviour, "
"change it to '.+'", FutureWarning),
# Element.getchildren() and Element.getiterator() are deprecated.
("This method will be removed in future versions. "
"Use .+ instead.", DeprecationWarning),
("This method will be removed in future versions. "
"Use .+ instead.", PendingDeprecationWarning))
self.checkwarnings = support.check_warnings(*deprecations, quiet=quiet)
def __enter__(self):
from xml.etree import ElementPath
self._nsmap = ET.register_namespace._namespace_map
# Copy the default namespace mapping
self._nsmap_copy = self._nsmap.copy()
# Copy the path cache (should be empty)
self._path_cache = ElementPath._cache
ElementPath._cache = self._path_cache.copy()
self.checkwarnings.__enter__()
def __exit__(self, *args):
from xml.etree import ElementPath
# Restore mapping and path cache
self._nsmap.clear()
self._nsmap.update(self._nsmap_copy)
ElementPath._cache = self._path_cache
self.checkwarnings.__exit__(*args)
def test_main(module=pyET):
from test import test_xml_etree
# The same doctests are used for both the Python and the C implementations
test_xml_etree.ET = module
test_classes = [
ElementSlicingTest,
StringIOTest,
ParseErrorTest,
ElementTreeTest,
TreeBuilderTest]
if module is pyET:
# Run the tests specific to the Python implementation
test_classes += [NoAcceleratorTest]
support.run_unittest(*test_classes)
# XXX the C module should give the same warnings as the Python module
with CleanContext(quiet=(module is not pyET)):
support.run_doctest(test_xml_etree, verbosity=True)
if __name__ == '__main__':
test_main()