\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":
data = ET.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
"""
#
# 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 BasicElementTest(ElementTestCase, unittest.TestCase):
def test_augmentation_type_errors(self):
e = ET.Element('joe')
self.assertRaises(TypeError, e.append, 'b')
self.assertRaises(TypeError, e.extend, [ET.Element('bar'), 'foo'])
self.assertRaises(TypeError, e.insert, 0, 'foo')
def test_cyclic_gc(self):
class Dummy:
pass
# Test the shortest cycle: d->element->d
d = Dummy()
d.dummyref = ET.Element('joe', attr=d)
wref = weakref.ref(d)
del d
gc_collect()
self.assertIsNone(wref())
# A longer cycle: d->e->e2->d
e = ET.Element('joe')
d = Dummy()
d.dummyref = e
wref = weakref.ref(d)
e2 = ET.SubElement(e, 'foo', attr=d)
del d, e, e2
gc_collect()
self.assertIsNone(wref())
# A cycle between Element objects as children of one another
# e1->e2->e3->e1
e1 = ET.Element('e1')
e2 = ET.Element('e2')
e3 = ET.Element('e3')
e1.append(e2)
e2.append(e2)
e3.append(e1)
wref = weakref.ref(e1)
del e1, e2, e3
gc_collect()
self.assertIsNone(wref())
def test_weakref(self):
flag = False
def wref_cb(w):
nonlocal flag
flag = True
e = ET.Element('e')
wref = weakref.ref(e, wref_cb)
self.assertEqual(wref().tag, 'e')
del e
self.assertEqual(flag, True)
self.assertEqual(wref(), None)
def test_get_keyword_args(self):
e1 = ET.Element('foo' , x=1, y=2, z=3)
self.assertEqual(e1.get('x', default=7), 1)
self.assertEqual(e1.get('w', default=7), 7)
def test_pickle(self):
# issue #16076: the C implementation wasn't pickleable.
for dumper, loader in product(self.modules, repeat=2):
e = dumper.Element('foo', bar=42)
e.text = "text goes here"
e.tail = "opposite of head"
dumper.SubElement(e, 'child').append(dumper.Element('grandchild'))
e.append(dumper.Element('child'))
e.findall('.//grandchild')[0].set('attr', 'other value')
e2 = self.pickleRoundTrip(e, 'xml.etree.ElementTree',
dumper, loader)
self.assertEqual(e2.tag, 'foo')
self.assertEqual(e2.attrib['bar'], 42)
self.assertEqual(len(e2), 2)
self.assertEqualElements(e, e2)
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)
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')
# test that attribute assignment works (issue 14849)
mye.text = "joe"
self.assertEqual(mye.text, "joe")
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')
def test_html_empty_elems_serialization(self):
# issue 15970
# from http://www.w3.org/TR/html401/index/elements.html
for element in ['AREA', 'BASE', 'BASEFONT', 'BR', 'COL', 'FRAME', 'HR',
'IMG', 'INPUT', 'ISINDEX', 'LINK', 'META', 'PARAM']:
for elem in [element, element.lower()]:
expected = '<%s>' % elem
serialized = serialize(ET.XML('<%s />' % elem), method='html')
self.assertEqual(serialized, expected)
serialized = serialize(ET.XML('<%s>' % (elem,elem)),
method='html')
self.assertEqual(serialized, expected)
class ElementFindTest(unittest.TestCase):
def test_find_simple(self):
e = ET.XML(SAMPLE_XML)
self.assertEqual(e.find('tag').tag, 'tag')
self.assertEqual(e.find('section/tag').tag, 'tag')
self.assertEqual(e.find('./tag').tag, 'tag')
e[2] = ET.XML(SAMPLE_SECTION)
self.assertEqual(e.find('section/nexttag').tag, 'nexttag')
self.assertEqual(e.findtext('./tag'), 'text')
self.assertEqual(e.findtext('section/tag'), 'subtext')
# section/nexttag is found but has no text
self.assertEqual(e.findtext('section/nexttag'), '')
self.assertEqual(e.findtext('section/nexttag', 'default'), '')
# tog doesn't exist and 'default' kicks in
self.assertIsNone(e.findtext('tog'))
self.assertEqual(e.findtext('tog', 'default'), 'default')
# Issue #16922
self.assertEqual(ET.XML('').findtext('empty'), '')
def test_find_xpath(self):
LINEAR_XML = '''
'''
e = ET.XML(LINEAR_XML)
# Test for numeric indexing and last()
self.assertEqual(e.find('./tag[1]').attrib['class'], 'a')
self.assertEqual(e.find('./tag[2]').attrib['class'], 'b')
self.assertEqual(e.find('./tag[last()]').attrib['class'], 'd')
self.assertEqual(e.find('./tag[last()-1]').attrib['class'], 'c')
self.assertEqual(e.find('./tag[last()-2]').attrib['class'], 'b')
def test_findall(self):
e = ET.XML(SAMPLE_XML)
e[2] = ET.XML(SAMPLE_SECTION)
self.assertEqual(summarize_list(e.findall('.')), ['body'])
self.assertEqual(summarize_list(e.findall('tag')), ['tag', 'tag'])
self.assertEqual(summarize_list(e.findall('tog')), [])
self.assertEqual(summarize_list(e.findall('tog/foo')), [])
self.assertEqual(summarize_list(e.findall('*')),
['tag', 'tag', 'section'])
self.assertEqual(summarize_list(e.findall('.//tag')),
['tag'] * 4)
self.assertEqual(summarize_list(e.findall('section/tag')), ['tag'])
self.assertEqual(summarize_list(e.findall('section//tag')), ['tag'] * 2)
self.assertEqual(summarize_list(e.findall('section/*')),
['tag', 'nexttag', 'nextsection'])
self.assertEqual(summarize_list(e.findall('section//*')),
['tag', 'nexttag', 'nextsection', 'tag'])
self.assertEqual(summarize_list(e.findall('section/.//*')),
['tag', 'nexttag', 'nextsection', 'tag'])
self.assertEqual(summarize_list(e.findall('*/*')),
['tag', 'nexttag', 'nextsection'])
self.assertEqual(summarize_list(e.findall('*//*')),
['tag', 'nexttag', 'nextsection', 'tag'])
self.assertEqual(summarize_list(e.findall('*/tag')), ['tag'])
self.assertEqual(summarize_list(e.findall('*/./tag')), ['tag'])
self.assertEqual(summarize_list(e.findall('./tag')), ['tag'] * 2)
self.assertEqual(summarize_list(e.findall('././tag')), ['tag'] * 2)
self.assertEqual(summarize_list(e.findall('.//tag[@class]')),
['tag'] * 3)
self.assertEqual(summarize_list(e.findall('.//tag[@class="a"]')),
['tag'])
self.assertEqual(summarize_list(e.findall('.//tag[@class="b"]')),
['tag'] * 2)
self.assertEqual(summarize_list(e.findall('.//tag[@id]')),
['tag'])
self.assertEqual(summarize_list(e.findall('.//section[tag]')),
['section'])
self.assertEqual(summarize_list(e.findall('.//section[element]')), [])
self.assertEqual(summarize_list(e.findall('../tag')), [])
self.assertEqual(summarize_list(e.findall('section/../tag')),
['tag'] * 2)
self.assertEqual(e.findall('section//'), e.findall('section//*'))
def test_test_find_with_ns(self):
e = ET.XML(SAMPLE_XML_NS)
self.assertEqual(summarize_list(e.findall('tag')), [])
self.assertEqual(
summarize_list(e.findall("{http://effbot.org/ns}tag")),
['{http://effbot.org/ns}tag'] * 2)
self.assertEqual(
summarize_list(e.findall(".//{http://effbot.org/ns}tag")),
['{http://effbot.org/ns}tag'] * 3)
def test_bad_find(self):
e = ET.XML(SAMPLE_XML)
with self.assertRaisesRegex(SyntaxError, 'cannot use absolute path'):
e.findall('/tag')
def test_find_through_ElementTree(self):
e = ET.XML(SAMPLE_XML)
self.assertEqual(ET.ElementTree(e).find('tag').tag, 'tag')
self.assertEqual(ET.ElementTree(e).findtext('tag'), 'text')
self.assertEqual(summarize_list(ET.ElementTree(e).findall('tag')),
['tag'] * 2)
# this produces a warning
self.assertEqual(summarize_list(ET.ElementTree(e).findall('//tag')),
['tag'] * 3)
class ElementIterTest(unittest.TestCase):
def _ilist(self, elem, tag=None):
return summarize_list(elem.iter(tag))
def test_basic(self):
doc = ET.XML("this is a paragraph...")
self.assertEqual(self._ilist(doc), ['html', 'body', 'i'])
self.assertEqual(self._ilist(doc.find('body')), ['body', 'i'])
self.assertEqual(next(doc.iter()).tag, 'html')
self.assertEqual(''.join(doc.itertext()), 'this is a paragraph...')
self.assertEqual(''.join(doc.find('body').itertext()),
'this is a paragraph.')
self.assertEqual(next(doc.itertext()), 'this is a ')
# iterparse should return an iterator
sourcefile = serialize(doc, to_string=False)
self.assertEqual(next(ET.iterparse(sourcefile))[0], 'end')
tree = ET.ElementTree(None)
self.assertRaises(AttributeError, tree.iter)
# Issue #16913
doc = ET.XML("a&b&c&")
self.assertEqual(''.join(doc.itertext()), 'a&b&c&')
def test_corners(self):
# single root, no subelements
a = ET.Element('a')
self.assertEqual(self._ilist(a), ['a'])
# one child
b = ET.SubElement(a, 'b')
self.assertEqual(self._ilist(a), ['a', 'b'])
# one child and one grandchild
c = ET.SubElement(b, 'c')
self.assertEqual(self._ilist(a), ['a', 'b', 'c'])
# two children, only first with grandchild
d = ET.SubElement(a, 'd')
self.assertEqual(self._ilist(a), ['a', 'b', 'c', 'd'])
# replace first child by second
a[0] = a[1]
del a[1]
self.assertEqual(self._ilist(a), ['a', 'd'])
def test_iter_by_tag(self):
doc = ET.XML('''
bedroom1
bedroom2
nothing here
bedroom8
''')
self.assertEqual(self._ilist(doc, 'room'), ['room'] * 3)
self.assertEqual(self._ilist(doc, 'house'), ['house'] * 2)
# test that iter also accepts 'tag' as a keyword arg
self.assertEqual(
summarize_list(doc.iter(tag='room')),
['room'] * 3)
# make sure both tag=None and tag='*' return all tags
all_tags = ['document', 'house', 'room', 'room',
'shed', 'house', 'room']
self.assertEqual(self._ilist(doc), all_tags)
self.assertEqual(self._ilist(doc, '*'), all_tags)
class TreeBuilderTest(unittest.TestCase):
sample1 = (''
'textsubtext
tail')
sample2 = '''sometext'''
def _check_sample1_element(self, e):
self.assertEqual(e.tag, 'html')
self.assertEqual(e.text, 'text')
self.assertEqual(e.tail, None)
self.assertEqual(e.attrib, {})
children = list(e)
self.assertEqual(len(children), 1)
child = children[0]
self.assertEqual(child.tag, 'div')
self.assertEqual(child.text, 'subtext')
self.assertEqual(child.tail, 'tail')
self.assertEqual(child.attrib, {})
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())
def test_subclass(self):
class MyTreeBuilder(ET.TreeBuilder):
def foobar(self, x):
return x * 2
tb = MyTreeBuilder()
self.assertEqual(tb.foobar(10), 20)
parser = ET.XMLParser(target=tb)
parser.feed(self.sample1)
e = parser.close()
self._check_sample1_element(e)
def test_element_factory(self):
lst = []
def myfactory(tag, attrib):
nonlocal lst
lst.append(tag)
return ET.Element(tag, attrib)
tb = ET.TreeBuilder(element_factory=myfactory)
parser = ET.XMLParser(target=tb)
parser.feed(self.sample2)
parser.close()
self.assertEqual(lst, ['toplevel'])
def _check_element_factory_class(self, cls):
tb = ET.TreeBuilder(element_factory=cls)
parser = ET.XMLParser(target=tb)
parser.feed(self.sample1)
e = parser.close()
self.assertIsInstance(e, cls)
self._check_sample1_element(e)
def test_element_factory_subclass(self):
class MyElement(ET.Element):
pass
self._check_element_factory_class(MyElement)
def test_element_factory_pure_python_subclass(self):
# Mimick SimpleTAL's behaviour (issue #16089): both versions of
# TreeBuilder should be able to cope with a subclass of the
# pure Python Element class.
base = ET._Element
# Not from a C extension
self.assertEqual(base.__module__, 'xml.etree.ElementTree')
# Force some multiple inheritance with a C class to make things
# more interesting.
class MyElement(base, ValueError):
pass
self._check_element_factory_class(MyElement)
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 XincludeTest(unittest.TestCase):
def _my_loader(self, href, parse):
# Used to avoid a test-dependency problem where the default loader
# of ElementInclude uses the pyET parser for cET tests.
if parse == 'xml':
with open(href, 'rb') as f:
return ET.parse(f).getroot()
else:
return None
def test_xinclude_default(self):
from xml.etree import ElementInclude
doc = xinclude_loader('default.xml')
ElementInclude.include(doc, self._my_loader)
s = serialize(doc)
self.assertEqual(s.strip(), '''
Example.
text
texttail
''')
class XMLParserTest(unittest.TestCase):
sample1 = '22'
sample2 = (''
'text')
def _check_sample_element(self, e):
self.assertEqual(e.tag, 'file')
self.assertEqual(e[0].tag, 'line')
self.assertEqual(e[0].text, '22')
def test_constructor_args(self):
# Positional args. The first (html) is not supported, but should be
# nevertheless correctly accepted.
parser = ET.XMLParser(None, ET.TreeBuilder(), 'utf-8')
parser.feed(self.sample1)
self._check_sample_element(parser.close())
# Now as keyword args.
parser2 = ET.XMLParser(encoding='utf-8', html=[{}], target=ET.TreeBuilder())
parser2.feed(self.sample1)
self._check_sample_element(parser2.close())
def test_subclass(self):
class MyParser(ET.XMLParser):
pass
parser = MyParser()
parser.feed(self.sample1)
self._check_sample_element(parser.close())
def test_subclass_doctype(self):
_doctype = None
class MyParserWithDoctype(ET.XMLParser):
def doctype(self, name, pubid, system):
nonlocal _doctype
_doctype = (name, pubid, system)
parser = MyParserWithDoctype()
parser.feed(self.sample2)
parser.close()
self.assertEqual(_doctype,
('html', '-//W3C//DTD XHTML 1.0 Transitional//EN',
'http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd'))
class NamespaceParseTest(unittest.TestCase):
def test_find_with_namespace(self):
nsmap = {'h': 'hello', 'f': 'foo'}
doc = ET.fromstring(SAMPLE_XML_NS_ELEMS)
self.assertEqual(len(doc.findall('{hello}table', nsmap)), 1)
self.assertEqual(len(doc.findall('.//{hello}td', nsmap)), 2)
self.assertEqual(len(doc.findall('.//{foo}name', nsmap)), 1)
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 IOTest(unittest.TestCase):
def tearDown(self):
unlink(TESTFN)
def test_encoding(self):
# Test encoding issues.
elem = ET.Element("tag")
elem.text = "abc"
self.assertEqual(serialize(elem), 'abc')
self.assertEqual(serialize(elem, encoding="utf-8"),
b'abc')
self.assertEqual(serialize(elem, encoding="us-ascii"),
b'abc')
for enc in ("iso-8859-1", "utf-16", "utf-32"):
self.assertEqual(serialize(elem, encoding=enc),
("\n"
"abc" % enc).encode(enc))
elem = ET.Element("tag")
elem.text = "<&\"\'>"
self.assertEqual(serialize(elem), '<&"\'>')
self.assertEqual(serialize(elem, encoding="utf-8"),
b'<&"\'>')
self.assertEqual(serialize(elem, encoding="us-ascii"),
b'<&"\'>')
for enc in ("iso-8859-1", "utf-16", "utf-32"):
self.assertEqual(serialize(elem, encoding=enc),
("\n"
"<&\"'>" % enc).encode(enc))
elem = ET.Element("tag")
elem.attrib["key"] = "<&\"\'>"
self.assertEqual(serialize(elem), '')
self.assertEqual(serialize(elem, encoding="utf-8"),
b'')
self.assertEqual(serialize(elem, encoding="us-ascii"),
b'')
for enc in ("iso-8859-1", "utf-16", "utf-32"):
self.assertEqual(serialize(elem, encoding=enc),
("\n"
"" % enc).encode(enc))
elem = ET.Element("tag")
elem.text = '\xe5\xf6\xf6<>'
self.assertEqual(serialize(elem), '\xe5\xf6\xf6<>')
self.assertEqual(serialize(elem, encoding="utf-8"),
b'\xc3\xa5\xc3\xb6\xc3\xb6<>')
self.assertEqual(serialize(elem, encoding="us-ascii"),
b'åöö<>')
for enc in ("iso-8859-1", "utf-16", "utf-32"):
self.assertEqual(serialize(elem, encoding=enc),
("\n"
"åöö<>" % enc).encode(enc))
elem = ET.Element("tag")
elem.attrib["key"] = '\xe5\xf6\xf6<>'
self.assertEqual(serialize(elem), '')
self.assertEqual(serialize(elem, encoding="utf-8"),
b'')
self.assertEqual(serialize(elem, encoding="us-ascii"),
b'')
for enc in ("iso-8859-1", "utf-16", "utf-16le", "utf-16be", "utf-32"):
self.assertEqual(serialize(elem, encoding=enc),
("\n"
"" % enc).encode(enc))
def test_write_to_filename(self):
tree = ET.ElementTree(ET.XML(''''''))
tree.write(TESTFN)
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(), b'''''')
def test_write_to_text_file(self):
tree = ET.ElementTree(ET.XML(''''''))
with open(TESTFN, 'w', encoding='utf-8') as f:
tree.write(f, encoding='unicode')
self.assertFalse(f.closed)
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(), b'''''')
def test_write_to_binary_file(self):
tree = ET.ElementTree(ET.XML(''''''))
with open(TESTFN, 'wb') as f:
tree.write(f)
self.assertFalse(f.closed)
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(), b'''''')
def test_write_to_binary_file_with_bom(self):
tree = ET.ElementTree(ET.XML(''''''))
# test BOM writing to buffered file
with open(TESTFN, 'wb') as f:
tree.write(f, encoding='utf-16')
self.assertFalse(f.closed)
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(),
'''\n'''
''''''.encode("utf-16"))
# test BOM writing to non-buffered file
with open(TESTFN, 'wb', buffering=0) as f:
tree.write(f, encoding='utf-16')
self.assertFalse(f.closed)
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(),
'''\n'''
''''''.encode("utf-16"))
def test_read_from_stringio(self):
tree = ET.ElementTree()
stream = io.StringIO('''''')
tree.parse(stream)
self.assertEqual(tree.getroot().tag, 'site')
def test_write_to_stringio(self):
tree = ET.ElementTree(ET.XML(''''''))
stream = io.StringIO()
tree.write(stream, encoding='unicode')
self.assertEqual(stream.getvalue(), '''''')
def test_read_from_bytesio(self):
tree = ET.ElementTree()
raw = io.BytesIO(b'''''')
tree.parse(raw)
self.assertEqual(tree.getroot().tag, 'site')
def test_write_to_bytesio(self):
tree = ET.ElementTree(ET.XML(''''''))
raw = io.BytesIO()
tree.write(raw)
self.assertEqual(raw.getvalue(), b'''''')
class dummy:
pass
def test_read_from_user_text_reader(self):
stream = io.StringIO('''''')
reader = self.dummy()
reader.read = stream.read
tree = ET.ElementTree()
tree.parse(reader)
self.assertEqual(tree.getroot().tag, 'site')
def test_write_to_user_text_writer(self):
tree = ET.ElementTree(ET.XML(''''''))
stream = io.StringIO()
writer = self.dummy()
writer.write = stream.write
tree.write(writer, encoding='unicode')
self.assertEqual(stream.getvalue(), '''''')
def test_read_from_user_binary_reader(self):
raw = io.BytesIO(b'''''')
reader = self.dummy()
reader.read = raw.read
tree = ET.ElementTree()
tree.parse(reader)
self.assertEqual(tree.getroot().tag, 'site')
tree = ET.ElementTree()
def test_write_to_user_binary_writer(self):
tree = ET.ElementTree(ET.XML(''''''))
raw = io.BytesIO()
writer = self.dummy()
writer.write = raw.write
tree.write(writer)
self.assertEqual(raw.getvalue(), b'''''')
def test_write_to_user_binary_writer_with_bom(self):
tree = ET.ElementTree(ET.XML(''''''))
raw = io.BytesIO()
writer = self.dummy()
writer.write = raw.write
writer.seekable = lambda: True
writer.tell = raw.tell
tree.write(writer, encoding="utf-16")
self.assertEqual(raw.getvalue(),
'''\n'''
''''''.encode("utf-16"))
def test_tostringlist_invariant(self):
root = ET.fromstring('foo')
self.assertEqual(
ET.tostring(root, 'unicode'),
''.join(ET.tostringlist(root, 'unicode')))
self.assertEqual(
ET.tostring(root, 'utf-16'),
b''.join(ET.tostringlist(root, 'utf-16')))
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 KeywordArgsTest(unittest.TestCase):
# Test various issues with keyword arguments passed to ET.Element
# constructor and methods
def test_issue14818(self):
x = ET.XML("foo")
self.assertEqual(x.find('a', None),
x.find(path='a', namespaces=None))
self.assertEqual(x.findtext('a', None, None),
x.findtext(path='a', default=None, namespaces=None))
self.assertEqual(x.findall('a', None),
x.findall(path='a', namespaces=None))
self.assertEqual(list(x.iterfind('a', None)),
list(x.iterfind(path='a', namespaces=None)))
self.assertEqual(ET.Element('a').attrib, {})
elements = [
ET.Element('a', dict(href="#", id="foo")),
ET.Element('a', attrib=dict(href="#", id="foo")),
ET.Element('a', dict(href="#"), id="foo"),
ET.Element('a', href="#", id="foo"),
ET.Element('a', dict(href="#", id="foo"), href="#", id="foo"),
]
for e in elements:
self.assertEqual(e.tag, 'a')
self.assertEqual(e.attrib, dict(href="#", id="foo"))
e2 = ET.SubElement(elements[0], 'foobar', attrib={'key1': 'value1'})
self.assertEqual(e2.attrib['key1'], 'value1')
with self.assertRaisesRegex(TypeError, 'must be dict, not str'):
ET.Element('a', "I'm not a dict")
with self.assertRaisesRegex(TypeError, 'must be dict, not str'):
ET.Element('a', attrib="I'm not a dict")
# --------------------------------------------------------------------
class NoAcceleratorTest(unittest.TestCase):
def setUp(self):
if not pyET:
raise unittest.SkipTest('only for the Python version')
# 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 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=None):
# When invoked without a module, runs the Python ET tests by loading pyET.
# Otherwise, uses the given module as the ET.
global pyET
pyET = import_fresh_module('xml.etree.ElementTree',
blocked=['_elementtree'])
if module is None:
module = pyET
global ET
ET = module
test_classes = [
ElementSlicingTest,
BasicElementTest,
IOTest,
ParseErrorTest,
XincludeTest,
ElementTreeTest,
ElementFindTest,
ElementIterTest,
TreeBuilderTest,
]
# These tests will only run for the pure-Python version that doesn't import
# _elementtree. We can't use skipUnless here, because pyET is filled in only
# after the module is loaded.
if pyET is not ET:
test_classes.extend([
NoAcceleratorTest,
])
try:
# XXX the C module should give the same warnings as the Python module
with CleanContext(quiet=(pyET is not ET)):
support.run_unittest(*test_classes)
support.run_doctest(sys.modules[__name__], verbosity=True)
finally:
# don't interfere with subsequent tests
ET = pyET = None
if __name__ == '__main__':
test_main()