\xe3'), 'latin-1'),
b"\n"
b"\xe3?>")
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)
def test_dump_attribute_order(self):
# See BPO 34160
e = ET.Element('cirriculum', status='public', company='example')
with support.captured_stdout() as stdout:
ET.dump(e)
self.assertEqual(stdout.getvalue(),
'\n')
def test_tree_write_attribute_order(self):
# See BPO 34160
root = ET.Element('cirriculum', status='public', company='example')
self.assertEqual(serialize(root),
'')
self.assertEqual(serialize(root, method='html'),
'')
def test_attlist_default(self):
# Test default attribute values; See BPO 42151.
root = ET.fromstring(ATTLIST_XML)
self.assertEqual(root[0].attrib,
{'{http://www.w3.org/XML/1998/namespace}lang': 'eng'})
class XMLPullParserTest(unittest.TestCase):
def _feed(self, parser, data, chunk_size=None):
if chunk_size is None:
parser.feed(data)
else:
for i in range(0, len(data), chunk_size):
parser.feed(data[i:i+chunk_size])
def assert_events(self, parser, expected, max_events=None):
self.assertEqual(
[(event, (elem.tag, elem.text))
for event, elem in islice(parser.read_events(), max_events)],
expected)
def assert_event_tuples(self, parser, expected, max_events=None):
self.assertEqual(
list(islice(parser.read_events(), max_events)),
expected)
def assert_event_tags(self, parser, expected, max_events=None):
events = islice(parser.read_events(), max_events)
self.assertEqual([(action, elem.tag) for action, elem in events],
expected)
def test_simple_xml(self):
for chunk_size in (None, 1, 5):
with self.subTest(chunk_size=chunk_size):
parser = ET.XMLPullParser()
self.assert_event_tags(parser, [])
self._feed(parser, "\n", chunk_size)
self.assert_event_tags(parser, [])
self._feed(parser,
"\n text\n", chunk_size)
self.assert_event_tags(parser, [('end', 'element')])
self._feed(parser, "texttail\n", chunk_size)
self._feed(parser, "\n", chunk_size)
self.assert_event_tags(parser, [
('end', 'element'),
('end', 'empty-element'),
])
self._feed(parser, "\n", chunk_size)
self.assert_event_tags(parser, [('end', 'root')])
self.assertIsNone(parser.close())
def test_feed_while_iterating(self):
parser = ET.XMLPullParser()
it = parser.read_events()
self._feed(parser, "\n text\n")
action, elem = next(it)
self.assertEqual((action, elem.tag), ('end', 'element'))
self._feed(parser, "\n")
action, elem = next(it)
self.assertEqual((action, elem.tag), ('end', 'root'))
with self.assertRaises(StopIteration):
next(it)
def test_simple_xml_with_ns(self):
parser = ET.XMLPullParser()
self.assert_event_tags(parser, [])
self._feed(parser, "\n")
self.assert_event_tags(parser, [])
self._feed(parser, "\n")
self.assert_event_tags(parser, [])
self._feed(parser, "text\n")
self.assert_event_tags(parser, [('end', '{namespace}element')])
self._feed(parser, "texttail\n")
self._feed(parser, "\n")
self.assert_event_tags(parser, [
('end', '{namespace}element'),
('end', '{namespace}empty-element'),
])
self._feed(parser, "\n")
self.assert_event_tags(parser, [('end', '{namespace}root')])
self.assertIsNone(parser.close())
def test_ns_events(self):
parser = ET.XMLPullParser(events=('start-ns', 'end-ns'))
self._feed(parser, "\n")
self._feed(parser, "\n")
self.assertEqual(
list(parser.read_events()),
[('start-ns', ('', 'namespace'))])
self._feed(parser, "text\n")
self._feed(parser, "texttail\n")
self._feed(parser, "\n")
self._feed(parser, "\n")
self.assertEqual(list(parser.read_events()), [('end-ns', None)])
self.assertIsNone(parser.close())
def test_ns_events_start(self):
parser = ET.XMLPullParser(events=('start-ns', 'start', 'end'))
self._feed(parser, "\n")
self.assert_event_tuples(parser, [
('start-ns', ('', 'abc')),
('start-ns', ('p', 'xyz')),
], max_events=2)
self.assert_event_tags(parser, [
('start', '{abc}tag'),
], max_events=1)
self._feed(parser, "\n")
self.assert_event_tags(parser, [
('start', '{abc}child'),
('end', '{abc}child'),
])
self._feed(parser, "\n")
parser.close()
self.assert_event_tags(parser, [
('end', '{abc}tag'),
])
def test_ns_events_start_end(self):
parser = ET.XMLPullParser(events=('start-ns', 'start', 'end', 'end-ns'))
self._feed(parser, "\n")
self.assert_event_tuples(parser, [
('start-ns', ('', 'abc')),
('start-ns', ('p', 'xyz')),
], max_events=2)
self.assert_event_tags(parser, [
('start', '{abc}tag'),
], max_events=1)
self._feed(parser, "\n")
self.assert_event_tags(parser, [
('start', '{abc}child'),
('end', '{abc}child'),
])
self._feed(parser, "\n")
parser.close()
self.assert_event_tags(parser, [
('end', '{abc}tag'),
], max_events=1)
self.assert_event_tuples(parser, [
('end-ns', None),
('end-ns', None),
])
def test_events(self):
parser = ET.XMLPullParser(events=())
self._feed(parser, "\n")
self.assert_event_tags(parser, [])
parser = ET.XMLPullParser(events=('start', 'end'))
self._feed(parser, "\n")
self.assert_events(parser, [])
parser = ET.XMLPullParser(events=('start', 'end'))
self._feed(parser, "\n")
self.assert_event_tags(parser, [('start', 'root')])
self._feed(parser, "text\n")
self.assert_event_tags(parser, [('end', 'element')])
self._feed(parser,
"texttail\n")
self.assert_event_tags(parser, [
('start', '{foo}element'),
('start', '{foo}empty-element'),
('end', '{foo}empty-element'),
('end', '{foo}element'),
])
self._feed(parser, "")
self.assertIsNone(parser.close())
self.assert_event_tags(parser, [('end', 'root')])
parser = ET.XMLPullParser(events=('start',))
self._feed(parser, "\n")
self.assert_event_tags(parser, [])
self._feed(parser, "\n")
self.assert_event_tags(parser, [('start', 'root')])
self._feed(parser, "text\n")
self.assert_event_tags(parser, [])
self._feed(parser,
"texttail\n")
self.assert_event_tags(parser, [
('start', '{foo}element'),
('start', '{foo}empty-element'),
])
self._feed(parser, "")
self.assertIsNone(parser.close())
def test_events_comment(self):
parser = ET.XMLPullParser(events=('start', 'comment', 'end'))
self._feed(parser, "\n")
self.assert_events(parser, [('comment', (ET.Comment, ' text here '))])
self._feed(parser, "\n")
self.assert_events(parser, [('comment', (ET.Comment, ' more text here '))])
self._feed(parser, "text")
self.assert_event_tags(parser, [('start', 'root-tag')])
self._feed(parser, "\n")
self.assert_events(parser, [('comment', (ET.Comment, ' inner comment'))])
self._feed(parser, "\n")
self.assert_event_tags(parser, [('end', 'root-tag')])
self._feed(parser, "\n")
self.assert_events(parser, [('comment', (ET.Comment, ' outer comment '))])
parser = ET.XMLPullParser(events=('comment',))
self._feed(parser, "\n")
self.assert_events(parser, [('comment', (ET.Comment, ' text here '))])
def test_events_pi(self):
parser = ET.XMLPullParser(events=('start', 'pi', 'end'))
self._feed(parser, "\n")
self.assert_events(parser, [('pi', (ET.PI, 'pitarget'))])
parser = ET.XMLPullParser(events=('pi',))
self._feed(parser, "\n")
self.assert_events(parser, [('pi', (ET.PI, 'pitarget some text '))])
def test_events_sequence(self):
# Test that events can be some sequence that's not just a tuple or list
eventset = {'end', 'start'}
parser = ET.XMLPullParser(events=eventset)
self._feed(parser, "bar")
self.assert_event_tags(parser, [('start', 'foo'), ('end', 'foo')])
class DummyIter:
def __init__(self):
self.events = iter(['start', 'end', 'start-ns'])
def __iter__(self):
return self
def __next__(self):
return next(self.events)
parser = ET.XMLPullParser(events=DummyIter())
self._feed(parser, "bar")
self.assert_event_tags(parser, [('start', 'foo'), ('end', 'foo')])
def test_unknown_event(self):
with self.assertRaises(ValueError):
ET.XMLPullParser(events=('start', 'end', 'bogus'))
#
# 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))
XINCLUDE["include_c1_repeated.xml"] = """\
The following is the source code of Recursive1.xml:
"""
#
# badly formatted xi:include tags
XINCLUDE_BAD = {}
XINCLUDE_BAD["B1.xml"] = """\
120 Mz is adequate for an average home user.
"""
XINCLUDE_BAD["B2.xml"] = """\
"""
XINCLUDE["Recursive1.xml"] = """\
The following is the source code of Recursive2.xml:
"""
XINCLUDE["Recursive2.xml"] = """\
The following is the source code of Recursive3.xml:
"""
XINCLUDE["Recursive3.xml"] = """\
The following is the source code of Recursive1.xml:
"""
class XIncludeTest(unittest.TestCase):
def xinclude_loader(self, 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 none_loader(self, href, parser, encoding=None):
return None
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 = self.xinclude_loader('default.xml')
ElementInclude.include(doc, self._my_loader)
self.assertEqual(serialize(doc),
'\n'
' Example.
\n'
' \n'
' text\n'
' texttail\n'
' \n'
'\n'
'')
def test_xinclude(self):
from xml.etree import ElementInclude
# Basic inclusion example (XInclude C.1)
document = self.xinclude_loader("C1.xml")
ElementInclude.include(document, self.xinclude_loader)
self.assertEqual(serialize(document),
'\n'
' 120 Mz is adequate for an average home user.
\n'
' \n'
' The opinions represented herein represent those of the individual\n'
' and should not be interpreted as official policy endorsed by this\n'
' organization.
\n'
'\n'
'') # C1
# Textual inclusion example (XInclude C.2)
document = self.xinclude_loader("C2.xml")
ElementInclude.include(document, self.xinclude_loader)
self.assertEqual(serialize(document),
'\n'
' This document has been accessed\n'
' 324387 times.
\n'
'') # C2
# Textual inclusion after sibling element (based on modified XInclude C.2)
document = self.xinclude_loader("C2b.xml")
ElementInclude.include(document, self.xinclude_loader)
self.assertEqual(serialize(document),
'\n'
' This document has been accessed\n'
' 324387 times.
\n'
'') # C2b
# Textual inclusion of XML example (XInclude C.3)
document = self.xinclude_loader("C3.xml")
ElementInclude.include(document, self.xinclude_loader)
self.assertEqual(serialize(document),
'\n'
' The following is the source of the "data.xml" resource:
\n'
" <?xml version='1.0'?>\n"
'<data>\n'
' <item><![CDATA[Brooks & Shields]]></item>\n'
'</data>\n'
'\n'
'') # C3
# Fallback example (XInclude C.5)
# Note! Fallback support is not yet implemented
document = self.xinclude_loader("C5.xml")
with self.assertRaises(OSError) as cm:
ElementInclude.include(document, self.xinclude_loader)
self.assertEqual(str(cm.exception), 'resource not found')
self.assertEqual(serialize(document),
'') # C5
def test_xinclude_repeated(self):
from xml.etree import ElementInclude
document = self.xinclude_loader("include_c1_repeated.xml")
ElementInclude.include(document, self.xinclude_loader)
self.assertEqual(1+4*2, len(document.findall(".//p")))
def test_xinclude_failures(self):
from xml.etree import ElementInclude
# Test failure to locate included XML file.
document = ET.XML(XINCLUDE["C1.xml"])
with self.assertRaises(ElementInclude.FatalIncludeError) as cm:
ElementInclude.include(document, loader=self.none_loader)
self.assertEqual(str(cm.exception),
"cannot load 'disclaimer.xml' as 'xml'")
# Test failure to locate included text file.
document = ET.XML(XINCLUDE["C2.xml"])
with self.assertRaises(ElementInclude.FatalIncludeError) as cm:
ElementInclude.include(document, loader=self.none_loader)
self.assertEqual(str(cm.exception),
"cannot load 'count.txt' as 'text'")
# Test bad parse type.
document = ET.XML(XINCLUDE_BAD["B1.xml"])
with self.assertRaises(ElementInclude.FatalIncludeError) as cm:
ElementInclude.include(document, loader=self.none_loader)
self.assertEqual(str(cm.exception),
"unknown parse type in xi:include tag ('BAD_TYPE')")
# Test xi:fallback outside xi:include.
document = ET.XML(XINCLUDE_BAD["B2.xml"])
with self.assertRaises(ElementInclude.FatalIncludeError) as cm:
ElementInclude.include(document, loader=self.none_loader)
self.assertEqual(str(cm.exception),
"xi:fallback tag must be child of xi:include "
"('{http://www.w3.org/2001/XInclude}fallback')")
# Test infinitely recursive includes.
document = self.xinclude_loader("Recursive1.xml")
with self.assertRaises(ElementInclude.FatalIncludeError) as cm:
ElementInclude.include(document, self.xinclude_loader)
self.assertEqual(str(cm.exception),
"recursive include of Recursive2.xml")
# Test 'max_depth' limitation.
document = self.xinclude_loader("Recursive1.xml")
with self.assertRaises(ElementInclude.FatalIncludeError) as cm:
ElementInclude.include(document, self.xinclude_loader, max_depth=None)
self.assertEqual(str(cm.exception),
"recursive include of Recursive2.xml")
document = self.xinclude_loader("Recursive1.xml")
with self.assertRaises(ElementInclude.LimitedRecursiveIncludeError) as cm:
ElementInclude.include(document, self.xinclude_loader, max_depth=0)
self.assertEqual(str(cm.exception),
"maximum xinclude depth reached when including file Recursive2.xml")
document = self.xinclude_loader("Recursive1.xml")
with self.assertRaises(ElementInclude.LimitedRecursiveIncludeError) as cm:
ElementInclude.include(document, self.xinclude_loader, max_depth=1)
self.assertEqual(str(cm.exception),
"maximum xinclude depth reached when including file Recursive3.xml")
document = self.xinclude_loader("Recursive1.xml")
with self.assertRaises(ElementInclude.LimitedRecursiveIncludeError) as cm:
ElementInclude.include(document, self.xinclude_loader, max_depth=2)
self.assertEqual(str(cm.exception),
"maximum xinclude depth reached when including file Recursive1.xml")
document = self.xinclude_loader("Recursive1.xml")
with self.assertRaises(ElementInclude.FatalIncludeError) as cm:
ElementInclude.include(document, self.xinclude_loader, max_depth=3)
self.assertEqual(str(cm.exception),
"recursive include of Recursive2.xml")
# --------------------------------------------------------------------
# reported bugs
class BugsTest(unittest.TestCase):
def test_bug_xmltoolkit21(self):
# marshaller gives obscure errors for non-string values
def check(elem):
with self.assertRaises(TypeError) as cm:
serialize(elem)
self.assertEqual(str(cm.exception),
'cannot serialize 123 (type int)')
elem = ET.Element(123)
check(elem) # tag
elem = ET.Element("elem")
elem.text = 123
check(elem) # text
elem = ET.Element("elem")
elem.tail = 123
check(elem) # tail
elem = ET.Element("elem")
elem.set(123, "123")
check(elem) # attribute key
elem = ET.Element("elem")
elem.set("123", 123)
check(elem) # attribute value
def test_bug_xmltoolkit25(self):
# typo in ElementTree.findtext
elem = ET.XML(SAMPLE_XML)
tree = ET.ElementTree(elem)
self.assertEqual(tree.findtext("tag"), 'text')
self.assertEqual(tree.findtext("section/tag"), 'subtext')
def test_bug_xmltoolkit28(self):
# .//tag causes exceptions
tree = ET.XML("")
self.assertEqual(summarize_list(tree.findall(".//thead")), [])
self.assertEqual(summarize_list(tree.findall(".//tbody")), ['tbody'])
def test_bug_xmltoolkitX1(self):
# dump() doesn't flush the output buffer
tree = ET.XML("")
with support.captured_stdout() as stdout:
ET.dump(tree)
self.assertEqual(stdout.getvalue(), '\n')
def test_bug_xmltoolkit39(self):
# non-ascii element and attribute names doesn't work
tree = ET.XML(b"")
self.assertEqual(ET.tostring(tree, "utf-8"), b'')
tree = ET.XML(b""
b"")
self.assertEqual(tree.attrib, {'\xe4ttr': 'v\xe4lue'})
self.assertEqual(ET.tostring(tree, "utf-8"),
b'')
tree = ET.XML(b""
b'text')
self.assertEqual(ET.tostring(tree, "utf-8"),
b'text')
tree = ET.Element("t\u00e4g")
self.assertEqual(ET.tostring(tree, "utf-8"), b'')
tree = ET.Element("tag")
tree.set("\u00e4ttr", "v\u00e4lue")
self.assertEqual(ET.tostring(tree, "utf-8"),
b'')
def test_bug_xmltoolkit54(self):
# problems handling internally defined entities
e = ET.XML("]>"
'&ldots;')
self.assertEqual(serialize(e, encoding="us-ascii"),
b'舰')
self.assertEqual(serialize(e), '\u8230')
def test_bug_xmltoolkit55(self):
# make sure we're reporting the first error, not the last
with self.assertRaises(ET.ParseError) as cm:
ET.XML(b""
b'&ldots;&ndots;&rdots;')
self.assertEqual(str(cm.exception),
'undefined entity &ldots;: line 1, column 36')
def test_bug_xmltoolkit60(self):
# Handle crash in stream source.
class ExceptionFile:
def read(self, x):
raise OSError
self.assertRaises(OSError, ET.parse, ExceptionFile())
def test_bug_xmltoolkit62(self):
# Don't crash when using custom entities.
ENTITIES = {'rsquo': '\u2019', 'lsquo': '\u2018'}
parser = ET.XMLParser()
parser.entity.update(ENTITIES)
parser.feed("""
A new cultivar of Begonia plant named ‘BCT9801BEG’.
""")
t = parser.close()
self.assertEqual(t.find('.//paragraph').text,
'A new cultivar of Begonia plant named \u2018BCT9801BEG\u2019.')
@unittest.skipIf(sys.gettrace(), "Skips under coverage.")
def test_bug_xmltoolkit63(self):
# Check reference leak.
def xmltoolkit63():
tree = ET.TreeBuilder()
tree.start("tag", {})
tree.data("text")
tree.end("tag")
xmltoolkit63()
count = sys.getrefcount(None)
for i in range(1000):
xmltoolkit63()
self.assertEqual(sys.getrefcount(None), count)
def test_bug_200708_newline(self):
# Preserve newlines in attributes.
e = ET.Element('SomeTag', text="def _f():\n return 3\n")
self.assertEqual(ET.tostring(e),
b'')
self.assertEqual(ET.XML(ET.tostring(e)).get("text"),
'def _f():\n return 3\n')
self.assertEqual(ET.tostring(ET.XML(ET.tostring(e))),
b'')
def test_bug_200708_close(self):
# Test default builder.
parser = ET.XMLParser() # default
parser.feed("some text")
self.assertEqual(parser.close().tag, 'element')
# Test custom builder.
class EchoTarget:
def close(self):
return ET.Element("element") # simulate root
parser = ET.XMLParser(target=EchoTarget())
parser.feed("some text")
self.assertEqual(parser.close().tag, 'element')
def test_bug_200709_default_namespace(self):
e = ET.Element("{default}elem")
s = ET.SubElement(e, "{default}elem")
self.assertEqual(serialize(e, default_namespace="default"), # 1
'')
e = ET.Element("{default}elem")
s = ET.SubElement(e, "{default}elem")
s = ET.SubElement(e, "{not-default}elem")
self.assertEqual(serialize(e, default_namespace="default"), # 2
''
''
''
'')
e = ET.Element("{default}elem")
s = ET.SubElement(e, "{default}elem")
s = ET.SubElement(e, "elem") # unprefixed name
with self.assertRaises(ValueError) as cm:
serialize(e, default_namespace="default") # 3
self.assertEqual(str(cm.exception),
'cannot use non-qualified names with default_namespace option')
def test_bug_200709_register_namespace(self):
e = ET.Element("{http://namespace.invalid/does/not/exist/}title")
self.assertEqual(ET.tostring(e),
b'')
ET.register_namespace("foo", "http://namespace.invalid/does/not/exist/")
e = ET.Element("{http://namespace.invalid/does/not/exist/}title")
self.assertEqual(ET.tostring(e),
b'')
# And the Dublin Core namespace is in the default list:
e = ET.Element("{http://purl.org/dc/elements/1.1/}title")
self.assertEqual(ET.tostring(e),
b'')
def test_bug_200709_element_comment(self):
# 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'))
self.assertEqual(a[0].tag, ET.Comment)
a = ET.Element('a')
a.append(ET.PI('foo'))
self.assertEqual(a[0].tag, ET.PI)
def test_bug_200709_element_insert(self):
a = ET.Element('a')
b = ET.SubElement(a, 'b')
c = ET.SubElement(a, 'c')
d = ET.Element('d')
a.insert(0, d)
self.assertEqual(summarize_list(a), ['d', 'b', 'c'])
a.insert(-1, d)
self.assertEqual(summarize_list(a), ['d', 'b', 'd', 'c'])
def test_bug_200709_iter_comment(self):
a = ET.Element('a')
b = ET.SubElement(a, 'b')
comment_b = ET.Comment("TEST-b")
b.append(comment_b)
self.assertEqual(summarize_list(a.iter(ET.Comment)), [ET.Comment])
# --------------------------------------------------------------------
# reported on bugs.python.org
def test_bug_1534630(self):
bob = ET.TreeBuilder()
e = bob.data("data")
e = bob.start("tag", {})
e = bob.end("tag")
e = bob.close()
self.assertEqual(serialize(e), '')
def test_issue6233(self):
e = ET.XML(b""
b't\xc3\xa3g')
self.assertEqual(ET.tostring(e, 'ascii'),
b"\n"
b'tãg')
e = ET.XML(b""
b't\xe3g')
self.assertEqual(ET.tostring(e, 'ascii'),
b"\n"
b'tãg')
def test_issue6565(self):
elem = ET.XML("")
self.assertEqual(summarize_list(elem), ['tag'])
newelem = ET.XML(SAMPLE_XML)
elem[:] = newelem[:]
self.assertEqual(summarize_list(elem), ['tag', 'tag', 'section'])
def test_issue10777(self):
# Registering a namespace twice caused a "dictionary changed size during
# iteration" bug.
ET.register_namespace('test10777', 'http://myuri/')
ET.register_namespace('test10777', 'http://myuri/')
def test_lost_text(self):
# Issue #25902: Borrowed text can disappear
class Text:
def __bool__(self):
e.text = 'changed'
return True
e = ET.Element('tag')
e.text = Text()
i = e.itertext()
t = next(i)
self.assertIsInstance(t, Text)
self.assertIsInstance(e.text, str)
self.assertEqual(e.text, 'changed')
def test_lost_tail(self):
# Issue #25902: Borrowed tail can disappear
class Text:
def __bool__(self):
e[0].tail = 'changed'
return True
e = ET.Element('root')
e.append(ET.Element('tag'))
e[0].tail = Text()
i = e.itertext()
t = next(i)
self.assertIsInstance(t, Text)
self.assertIsInstance(e[0].tail, str)
self.assertEqual(e[0].tail, 'changed')
def test_lost_elem(self):
# Issue #25902: Borrowed element can disappear
class Tag:
def __eq__(self, other):
e[0] = ET.Element('changed')
next(i)
return True
e = ET.Element('root')
e.append(ET.Element(Tag()))
e.append(ET.Element('tag'))
i = e.iter('tag')
try:
t = next(i)
except ValueError:
self.skipTest('generators are not reentrant')
self.assertIsInstance(t.tag, Tag)
self.assertIsInstance(e[0].tag, str)
self.assertEqual(e[0].tag, 'changed')
def check_expat224_utf8_bug(self, text):
xml = b'' % text
root = ET.XML(xml)
self.assertEqual(root.get('b'), text.decode('utf-8'))
def test_expat224_utf8_bug(self):
# bpo-31170: Expat 2.2.3 had a bug in its UTF-8 decoder.
# Check that Expat 2.2.4 fixed the bug.
#
# Test buffer bounds at odd and even positions.
text = b'\xc3\xa0' * 1024
self.check_expat224_utf8_bug(text)
text = b'x' + b'\xc3\xa0' * 1024
self.check_expat224_utf8_bug(text)
def test_expat224_utf8_bug_file(self):
with open(UTF8_BUG_XMLFILE, 'rb') as fp:
raw = fp.read()
root = ET.fromstring(raw)
xmlattr = root.get('b')
# "Parse" manually the XML file to extract the value of the 'b'
# attribute of the XML element
text = raw.decode('utf-8').strip()
text = text.replace('\r\n', ' ')
text = text[6:-4]
self.assertEqual(root.get('b'), text)
def test_39495_treebuilder_start(self):
self.assertRaises(TypeError, ET.TreeBuilder().start, "tag")
self.assertRaises(TypeError, ET.TreeBuilder().start, "tag", None)
# --------------------------------------------------------------------
class BasicElementTest(ElementTestCase, unittest.TestCase):
def test___init__(self):
tag = "foo"
attrib = { "zix": "wyp" }
element_foo = ET.Element(tag, attrib)
# traits of an element
self.assertIsInstance(element_foo, ET.Element)
self.assertIn("tag", dir(element_foo))
self.assertIn("attrib", dir(element_foo))
self.assertIn("text", dir(element_foo))
self.assertIn("tail", dir(element_foo))
# string attributes have expected values
self.assertEqual(element_foo.tag, tag)
self.assertIsNone(element_foo.text)
self.assertIsNone(element_foo.tail)
# attrib is a copy
self.assertIsNot(element_foo.attrib, attrib)
self.assertEqual(element_foo.attrib, attrib)
# attrib isn't linked
attrib["bar"] = "baz"
self.assertIsNot(element_foo.attrib, attrib)
self.assertNotEqual(element_foo.attrib, attrib)
def test_copy(self):
# Only run this test if Element.copy() is defined.
if "copy" not in dir(ET.Element):
raise unittest.SkipTest("Element.copy() not present")
element_foo = ET.Element("foo", { "zix": "wyp" })
element_foo.append(ET.Element("bar", { "baz": "qix" }))
with self.assertWarns(DeprecationWarning):
element_foo2 = element_foo.copy()
# elements are not the same
self.assertIsNot(element_foo2, element_foo)
# string attributes are equal
self.assertEqual(element_foo2.tag, element_foo.tag)
self.assertEqual(element_foo2.text, element_foo.text)
self.assertEqual(element_foo2.tail, element_foo.tail)
# number of children is the same
self.assertEqual(len(element_foo2), len(element_foo))
# children are the same
for (child1, child2) in itertools.zip_longest(element_foo, element_foo2):
self.assertIs(child1, child2)
# attrib is a copy
self.assertEqual(element_foo2.attrib, element_foo.attrib)
def test___copy__(self):
element_foo = ET.Element("foo", { "zix": "wyp" })
element_foo.append(ET.Element("bar", { "baz": "qix" }))
element_foo2 = copy.copy(element_foo)
# elements are not the same
self.assertIsNot(element_foo2, element_foo)
# string attributes are equal
self.assertEqual(element_foo2.tag, element_foo.tag)
self.assertEqual(element_foo2.text, element_foo.text)
self.assertEqual(element_foo2.tail, element_foo.tail)
# number of children is the same
self.assertEqual(len(element_foo2), len(element_foo))
# children are the same
for (child1, child2) in itertools.zip_longest(element_foo, element_foo2):
self.assertIs(child1, child2)
# attrib is a copy
self.assertEqual(element_foo2.attrib, element_foo.attrib)
def test___deepcopy__(self):
element_foo = ET.Element("foo", { "zix": "wyp" })
element_foo.append(ET.Element("bar", { "baz": "qix" }))
element_foo2 = copy.deepcopy(element_foo)
# elements are not the same
self.assertIsNot(element_foo2, element_foo)
# string attributes are equal
self.assertEqual(element_foo2.tag, element_foo.tag)
self.assertEqual(element_foo2.text, element_foo.text)
self.assertEqual(element_foo2.tail, element_foo.tail)
# number of children is the same
self.assertEqual(len(element_foo2), len(element_foo))
# children are not the same
for (child1, child2) in itertools.zip_longest(element_foo, element_foo2):
self.assertIsNot(child1, child2)
# attrib is a copy
self.assertIsNot(element_foo2.attrib, element_foo.attrib)
self.assertEqual(element_foo2.attrib, element_foo.attrib)
# attrib isn't linked
element_foo.attrib["bar"] = "baz"
self.assertIsNot(element_foo2.attrib, element_foo.attrib)
self.assertNotEqual(element_foo2.attrib, element_foo.attrib)
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')
e[:] = [ET.Element('bar')]
with self.assertRaises(TypeError):
e[0] = 'foo'
with self.assertRaises(TypeError):
e[:] = [ET.Element('bar'), 'foo']
if hasattr(e, '__setstate__'):
state = {
'tag': 'tag',
'_children': [None], # non-Element
'attrib': 'attr',
'tail': 'tail',
'text': 'text',
}
self.assertRaises(TypeError, e.__setstate__, state)
if hasattr(e, '__deepcopy__'):
class E(ET.Element):
def __deepcopy__(self, memo):
return None # non-Element
e[:] = [E('bar')]
self.assertRaises(TypeError, copy.deepcopy, e)
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')
e3.append(e1)
e2.append(e3)
e1.append(e2)
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
gc_collect() # For PyPy or other GCs.
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 proto in range(2, pickle.HIGHEST_PROTOCOL + 1):
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, proto)
self.assertEqual(e2.tag, 'foo')
self.assertEqual(e2.attrib['bar'], 42)
self.assertEqual(len(e2), 2)
self.assertEqualElements(e, e2)
def test_pickle_issue18997(self):
for proto in range(2, pickle.HIGHEST_PROTOCOL + 1):
for dumper, loader in product(self.modules, repeat=2):
XMLTEXT = """
4
"""
e1 = dumper.fromstring(XMLTEXT)
if hasattr(e1, '__getstate__'):
self.assertEqual(e1.__getstate__()['tag'], 'group')
e2 = self.pickleRoundTrip(e1, 'xml.etree.ElementTree',
dumper, loader, proto)
self.assertEqual(e2.tag, 'group')
self.assertEqual(e2[0].tag, 'dogs')
class BadElementTest(ElementTestCase, unittest.TestCase):
def test_extend_mutable_list(self):
class X:
@property
def __class__(self):
L[:] = [ET.Element('baz')]
return ET.Element
L = [X()]
e = ET.Element('foo')
try:
e.extend(L)
except TypeError:
pass
class Y(X, ET.Element):
pass
L = [Y('x')]
e = ET.Element('foo')
e.extend(L)
def test_extend_mutable_list2(self):
class X:
@property
def __class__(self):
del L[:]
return ET.Element
L = [X(), ET.Element('baz')]
e = ET.Element('foo')
try:
e.extend(L)
except TypeError:
pass
class Y(X, ET.Element):
pass
L = [Y('bar'), ET.Element('baz')]
e = ET.Element('foo')
e.extend(L)
def test_remove_with_mutating(self):
class X(ET.Element):
def __eq__(self, o):
del e[:]
return False
e = ET.Element('foo')
e.extend([X('bar')])
self.assertRaises(ValueError, e.remove, ET.Element('baz'))
e = ET.Element('foo')
e.extend([ET.Element('bar')])
self.assertRaises(ValueError, e.remove, X('baz'))
def test_recursive_repr(self):
# Issue #25455
e = ET.Element('foo')
with swap_attr(e, 'tag', e):
with self.assertRaises(RuntimeError):
repr(e) # Should not crash
def test_element_get_text(self):
# Issue #27863
class X(str):
def __del__(self):
try:
elem.text
except NameError:
pass
b = ET.TreeBuilder()
b.start('tag', {})
b.data('ABCD')
b.data(X('EFGH'))
b.data('IJKL')
b.end('tag')
elem = b.close()
self.assertEqual(elem.text, 'ABCDEFGHIJKL')
def test_element_get_tail(self):
# Issue #27863
class X(str):
def __del__(self):
try:
elem[0].tail
except NameError:
pass
b = ET.TreeBuilder()
b.start('root', {})
b.start('tag', {})
b.end('tag')
b.data('ABCD')
b.data(X('EFGH'))
b.data('IJKL')
b.end('root')
elem = b.close()
self.assertEqual(elem[0].tail, 'ABCDEFGHIJKL')
def test_subscr(self):
# Issue #27863
class X:
def __index__(self):
del e[:]
return 1
e = ET.Element('elem')
e.append(ET.Element('child'))
e[:X()] # shouldn't crash
e.append(ET.Element('child'))
e[0:10:X()] # shouldn't crash
def test_ass_subscr(self):
# Issue #27863
class X:
def __index__(self):
e[:] = []
return 1
e = ET.Element('elem')
for _ in range(10):
e.insert(0, ET.Element('child'))
e[0:10:X()] = [] # shouldn't crash
def test_treebuilder_start(self):
# Issue #27863
def element_factory(x, y):
return []
b = ET.TreeBuilder(element_factory=element_factory)
b.start('tag', {})
b.data('ABCD')
self.assertRaises(AttributeError, b.start, 'tag2', {})
del b
gc_collect()
def test_treebuilder_end(self):
# Issue #27863
def element_factory(x, y):
return []
b = ET.TreeBuilder(element_factory=element_factory)
b.start('tag', {})
b.data('ABCD')
self.assertRaises(AttributeError, b.end, 'tag')
del b
gc_collect()
class MutatingElementPath(str):
def __new__(cls, elem, *args):
self = str.__new__(cls, *args)
self.elem = elem
return self
def __eq__(self, o):
del self.elem[:]
return True
MutatingElementPath.__hash__ = str.__hash__
class BadElementPath(str):
def __eq__(self, o):
raise 1/0
BadElementPath.__hash__ = str.__hash__
class BadElementPathTest(ElementTestCase, unittest.TestCase):
def setUp(self):
super().setUp()
from xml.etree import ElementPath
self.path_cache = ElementPath._cache
ElementPath._cache = {}
def tearDown(self):
from xml.etree import ElementPath
ElementPath._cache = self.path_cache
super().tearDown()
def test_find_with_mutating(self):
e = ET.Element('foo')
e.extend([ET.Element('bar')])
e.find(MutatingElementPath(e, 'x'))
def test_find_with_error(self):
e = ET.Element('foo')
e.extend([ET.Element('bar')])
try:
e.find(BadElementPath('x'))
except ZeroDivisionError:
pass
def test_findtext_with_mutating(self):
e = ET.Element('foo')
e.extend([ET.Element('bar')])
e.findtext(MutatingElementPath(e, 'x'))
def test_findtext_with_error(self):
e = ET.Element('foo')
e.extend([ET.Element('bar')])
try:
e.findtext(BadElementPath('x'))
except ZeroDivisionError:
pass
def test_findall_with_mutating(self):
e = ET.Element('foo')
e.extend([ET.Element('bar')])
e.findall(MutatingElementPath(e, 'x'))
def test_findall_with_error(self):
e = ET.Element('foo')
e.extend([ET.Element('bar')])
try:
e.findall(BadElementPath('x'))
except ZeroDivisionError:
pass
class ElementTreeTypeTest(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_Element_subclass_find(self):
class MyElement(ET.Element):
pass
e = ET.Element('foo')
e.text = 'text'
sub = MyElement('bar')
sub.text = 'subtext'
e.append(sub)
self.assertEqual(e.findtext('bar'), 'subtext')
self.assertEqual(e.find('bar').tag, 'bar')
found = list(e.findall('bar'))
self.assertEqual(len(found), 1, found)
self.assertEqual(found[0].tag, 'bar')
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')
self.assertRaisesRegex(SyntaxError, 'XPath', e.find, './tag[0]')
self.assertRaisesRegex(SyntaxError, 'XPath', e.find, './tag[-1]')
self.assertRaisesRegex(SyntaxError, 'XPath', e.find, './tag[last()-0]')
self.assertRaisesRegex(SyntaxError, 'XPath', e.find, './tag[last()+1]')
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!="a"]')),
['tag'] * 2)
self.assertEqual(summarize_list(e.findall('.//tag[@class="b"]')),
['tag'] * 2)
self.assertEqual(summarize_list(e.findall('.//tag[@class!="b"]')),
['tag'])
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//*'))
self.assertEqual(summarize_list(e.findall(".//section[tag='subtext']")),
['section'])
self.assertEqual(summarize_list(e.findall(".//section[tag ='subtext']")),
['section'])
self.assertEqual(summarize_list(e.findall(".//section[tag= 'subtext']")),
['section'])
self.assertEqual(summarize_list(e.findall(".//section[tag = 'subtext']")),
['section'])
self.assertEqual(summarize_list(e.findall(".//section[ tag = 'subtext' ]")),
['section'])
# Negations of above tests. They match nothing because the sole section
# tag has subtext.
self.assertEqual(summarize_list(e.findall(".//section[tag!='subtext']")),
[])
self.assertEqual(summarize_list(e.findall(".//section[tag !='subtext']")),
[])
self.assertEqual(summarize_list(e.findall(".//section[tag!= 'subtext']")),
[])
self.assertEqual(summarize_list(e.findall(".//section[tag != 'subtext']")),
[])
self.assertEqual(summarize_list(e.findall(".//section[ tag != 'subtext' ]")),
[])
self.assertEqual(summarize_list(e.findall(".//tag[.='subtext']")),
['tag'])
self.assertEqual(summarize_list(e.findall(".//tag[. ='subtext']")),
['tag'])
self.assertEqual(summarize_list(e.findall('.//tag[.= "subtext"]')),
['tag'])
self.assertEqual(summarize_list(e.findall('.//tag[ . = "subtext" ]')),
['tag'])
self.assertEqual(summarize_list(e.findall(".//tag[. = 'subtext']")),
['tag'])
self.assertEqual(summarize_list(e.findall(".//tag[. = 'subtext ']")),
[])
self.assertEqual(summarize_list(e.findall(".//tag[.= ' subtext']")),
[])
# Negations of above tests.
# Matches everything but the tag containing subtext
self.assertEqual(summarize_list(e.findall(".//tag[.!='subtext']")),
['tag'] * 3)
self.assertEqual(summarize_list(e.findall(".//tag[. !='subtext']")),
['tag'] * 3)
self.assertEqual(summarize_list(e.findall('.//tag[.!= "subtext"]')),
['tag'] * 3)
self.assertEqual(summarize_list(e.findall('.//tag[ . != "subtext" ]')),
['tag'] * 3)
self.assertEqual(summarize_list(e.findall(".//tag[. != 'subtext']")),
['tag'] * 3)
# Matches all tags.
self.assertEqual(summarize_list(e.findall(".//tag[. != 'subtext ']")),
['tag'] * 4)
self.assertEqual(summarize_list(e.findall(".//tag[.!= ' subtext']")),
['tag'] * 4)
# duplicate section => 2x tag matches
e[1] = e[2]
self.assertEqual(summarize_list(e.findall(".//section[tag = 'subtext']")),
['section', 'section'])
self.assertEqual(summarize_list(e.findall(".//tag[. = 'subtext']")),
['tag', 'tag'])
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_findall_different_nsmaps(self):
root = ET.XML('''
''')
nsmap = {'xx': 'X'}
self.assertEqual(len(root.findall(".//xx:b", namespaces=nsmap)), 2)
self.assertEqual(len(root.findall(".//b", namespaces=nsmap)), 2)
nsmap = {'xx': 'Y'}
self.assertEqual(len(root.findall(".//xx:b", namespaces=nsmap)), 1)
self.assertEqual(len(root.findall(".//b", namespaces=nsmap)), 2)
nsmap = {'xx': 'X', '': 'Y'}
self.assertEqual(len(root.findall(".//xx:b", namespaces=nsmap)), 2)
self.assertEqual(len(root.findall(".//b", namespaces=nsmap)), 1)
def test_findall_wildcard(self):
root = ET.XML('''
''')
root.append(ET.Comment('test'))
self.assertEqual(summarize_list(root.findall("{*}b")),
['{X}b', 'b', '{Y}b'])
self.assertEqual(summarize_list(root.findall("{*}c")),
['c'])
self.assertEqual(summarize_list(root.findall("{X}*")),
['{X}b'])
self.assertEqual(summarize_list(root.findall("{Y}*")),
['{Y}b'])
self.assertEqual(summarize_list(root.findall("{}*")),
['b', 'c'])
self.assertEqual(summarize_list(root.findall("{}b")), # only for consistency
['b'])
self.assertEqual(summarize_list(root.findall("{}b")),
summarize_list(root.findall("b")))
self.assertEqual(summarize_list(root.findall("{*}*")),
['{X}b', 'b', 'c', '{Y}b'])
# This is an unfortunate difference, but that's how find('*') works.
self.assertEqual(summarize_list(root.findall("{*}*") + [root[-1]]),
summarize_list(root.findall("*")))
self.assertEqual(summarize_list(root.findall(".//{*}b")),
['{X}b', 'b', '{X}b', 'b', '{Y}b'])
self.assertEqual(summarize_list(root.findall(".//{*}c")),
['c', 'c'])
self.assertEqual(summarize_list(root.findall(".//{X}*")),
['{X}b', '{X}b'])
self.assertEqual(summarize_list(root.findall(".//{Y}*")),
['{Y}b'])
self.assertEqual(summarize_list(root.findall(".//{}*")),
['c', 'b', 'c', 'b'])
self.assertEqual(summarize_list(root.findall(".//{}b")), # only for consistency
['b', 'b'])
self.assertEqual(summarize_list(root.findall(".//{}b")),
summarize_list(root.findall(".//b")))
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
msg = ("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 '.+'")
with self.assertWarnsRegex(FutureWarning, msg):
it = ET.ElementTree(e).findall('//tag')
self.assertEqual(summarize_list(it), ['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')
# With an explicit parser too (issue #9708)
sourcefile = serialize(doc, to_string=False)
parser = ET.XMLParser(target=ET.TreeBuilder())
self.assertEqual(next(ET.iterparse(sourcefile, parser=parser))[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(summarize_list(doc.iter()), all_tags)
self.assertEqual(self._ilist(doc), all_tags)
self.assertEqual(self._ilist(doc, '*'), all_tags)
def test_copy(self):
a = ET.Element('a')
it = a.iter()
with self.assertRaises(TypeError):
copy.copy(it)
def test_pickle(self):
a = ET.Element('a')
it = a.iter()
for proto in range(pickle.HIGHEST_PROTOCOL + 1):
with self.assertRaises((TypeError, pickle.PicklingError)):
pickle.dumps(it, proto)
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_treebuilder_comment(self):
b = ET.TreeBuilder()
self.assertEqual(b.comment('ctext').tag, ET.Comment)
self.assertEqual(b.comment('ctext').text, 'ctext')
b = ET.TreeBuilder(comment_factory=ET.Comment)
self.assertEqual(b.comment('ctext').tag, ET.Comment)
self.assertEqual(b.comment('ctext').text, 'ctext')
b = ET.TreeBuilder(comment_factory=len)
self.assertEqual(b.comment('ctext'), len('ctext'))
def test_treebuilder_pi(self):
b = ET.TreeBuilder()
self.assertEqual(b.pi('target', None).tag, ET.PI)
self.assertEqual(b.pi('target', None).text, 'target')
b = ET.TreeBuilder(pi_factory=ET.PI)
self.assertEqual(b.pi('target').tag, ET.PI)
self.assertEqual(b.pi('target').text, "target")
self.assertEqual(b.pi('pitarget', ' text ').tag, ET.PI)
self.assertEqual(b.pi('pitarget', ' text ').text, "pitarget text ")
b = ET.TreeBuilder(pi_factory=lambda target, text: (len(target), text))
self.assertEqual(b.pi('target'), (len('target'), None))
self.assertEqual(b.pi('pitarget', ' text '), (len('pitarget'), ' text '))
def test_late_tail(self):
# Issue #37399: The tail of an ignored comment could overwrite the text before it.
class TreeBuilderSubclass(ET.TreeBuilder):
pass
xml = "texttail"
a = ET.fromstring(xml)
self.assertEqual(a.text, "texttail")
parser = ET.XMLParser(target=TreeBuilderSubclass())
parser.feed(xml)
a = parser.close()
self.assertEqual(a.text, "texttail")
xml = "texttail"
a = ET.fromstring(xml)
self.assertEqual(a.text, "texttail")
xml = "texttail"
parser = ET.XMLParser(target=TreeBuilderSubclass())
parser.feed(xml)
a = parser.close()
self.assertEqual(a.text, "texttail")
def test_late_tail_mix_pi_comments(self):
# Issue #37399: The tail of an ignored comment could overwrite the text before it.
# Test appending tails to comments/pis.
class TreeBuilderSubclass(ET.TreeBuilder):
pass
xml = "text \ntail"
parser = ET.XMLParser(target=ET.TreeBuilder(insert_comments=True))
parser.feed(xml)
a = parser.close()
self.assertEqual(a[0].text, ' comment ')
self.assertEqual(a[0].tail, '\ntail')
self.assertEqual(a.text, "text ")
parser = ET.XMLParser(target=TreeBuilderSubclass(insert_comments=True))
parser.feed(xml)
a = parser.close()
self.assertEqual(a[0].text, ' comment ')
self.assertEqual(a[0].tail, '\ntail')
self.assertEqual(a.text, "text ")
xml = "text\ntail"
parser = ET.XMLParser(target=ET.TreeBuilder(insert_pis=True))
parser.feed(xml)
a = parser.close()
self.assertEqual(a[0].text, 'pi data')
self.assertEqual(a[0].tail, 'tail')
self.assertEqual(a.text, "text\n")
parser = ET.XMLParser(target=TreeBuilderSubclass(insert_pis=True))
parser.feed(xml)
a = parser.close()
self.assertEqual(a[0].text, 'pi data')
self.assertEqual(a[0].tail, 'tail')
self.assertEqual(a.text, "text\n")
def test_treebuilder_elementfactory_none(self):
parser = ET.XMLParser(target=ET.TreeBuilder(element_factory=None))
parser.feed(self.sample1)
e = parser.close()
self._check_sample1_element(e)
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_subclass_comment_pi(self):
class MyTreeBuilder(ET.TreeBuilder):
def foobar(self, x):
return x * 2
tb = MyTreeBuilder(comment_factory=ET.Comment, pi_factory=ET.PI)
self.assertEqual(tb.foobar(10), 20)
parser = ET.XMLParser(target=tb)
parser.feed(self.sample1)
parser.feed('')
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):
# Mimic 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_Py
# 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'))
def test_builder_lookup_errors(self):
class RaisingBuilder:
def __init__(self, raise_in=None, what=ValueError):
self.raise_in = raise_in
self.what = what
def __getattr__(self, name):
if name == self.raise_in:
raise self.what(self.raise_in)
def handle(*args):
pass
return handle
ET.XMLParser(target=RaisingBuilder())
# cET also checks for 'close' and 'doctype', PyET does it only at need
for event in ('start', 'data', 'end', 'comment', 'pi'):
with self.assertRaisesRegex(ValueError, event):
ET.XMLParser(target=RaisingBuilder(event))
ET.XMLParser(target=RaisingBuilder(what=AttributeError))
for event in ('start', 'data', 'end', 'comment', 'pi'):
parser = ET.XMLParser(target=RaisingBuilder(event, what=AttributeError))
parser.feed(self.sample1)
self.assertIsNone(parser.close())
class XMLParserTest(unittest.TestCase):
sample1 = b'22'
sample2 = (b''
b'text')
sample3 = ('\n'
'$\xa3\u20ac\U0001017b')
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):
parser2 = ET.XMLParser(encoding='utf-8',
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_doctype_warning(self):
with warnings.catch_warnings():
warnings.simplefilter('error', DeprecationWarning)
parser = ET.XMLParser()
parser.feed(self.sample2)
parser.close()
def test_subclass_doctype(self):
_doctype = None
class MyParserWithDoctype(ET.XMLParser):
def doctype(self, *args, **kwargs):
nonlocal _doctype
_doctype = (args, kwargs)
parser = MyParserWithDoctype()
with self.assertWarnsRegex(RuntimeWarning, 'doctype'):
parser.feed(self.sample2)
parser.close()
self.assertIsNone(_doctype)
_doctype = _doctype2 = None
with warnings.catch_warnings():
warnings.simplefilter('error', DeprecationWarning)
warnings.simplefilter('error', RuntimeWarning)
class DoctypeParser:
def doctype(self, name, pubid, system):
nonlocal _doctype2
_doctype2 = (name, pubid, system)
parser = MyParserWithDoctype(target=DoctypeParser())
parser.feed(self.sample2)
parser.close()
self.assertIsNone(_doctype)
self.assertEqual(_doctype2,
('html', '-//W3C//DTD XHTML 1.0 Transitional//EN',
'http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd'))
def test_inherited_doctype(self):
'''Ensure that ordinary usage is not deprecated (Issue 19176)'''
with warnings.catch_warnings():
warnings.simplefilter('error', DeprecationWarning)
warnings.simplefilter('error', RuntimeWarning)
class MyParserWithoutDoctype(ET.XMLParser):
pass
parser = MyParserWithoutDoctype()
parser.feed(self.sample2)
parser.close()
def test_parse_string(self):
parser = ET.XMLParser(target=ET.TreeBuilder())
parser.feed(self.sample3)
e = parser.close()
self.assertEqual(e.tag, 'money')
self.assertEqual(e.attrib['value'], '$\xa3\u20ac\U0001017b')
self.assertEqual(e.text, '$\xa3\u20ac\U0001017b')
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])
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'])
self.assertEqual(self._elem_tags(e[3::sys.maxsize]), ['a3'])
self.assertEqual(self._elem_tags(e[3::sys.maxsize<<64]), ['a3'])
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'])
self.assertEqual(self._elem_tags(e[3::-sys.maxsize]), ['a3'])
self.assertEqual(self._elem_tags(e[3::-sys.maxsize-1]), ['a3'])
self.assertEqual(self._elem_tags(e[3::-sys.maxsize<<64]), ['a3'])
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'])
def test_setslice_single_index(self):
e = self._make_elem_with_children(4)
e[1] = ET.Element('b')
self.assertEqual(self._subelem_tags(e), ['a0', 'b', 'a2', 'a3'])
e[-2] = ET.Element('c')
self.assertEqual(self._subelem_tags(e), ['a0', 'b', 'c', 'a3'])
with self.assertRaises(IndexError):
e[5] = ET.Element('d')
with self.assertRaises(IndexError):
e[-5] = ET.Element('d')
self.assertEqual(self._subelem_tags(e), ['a0', 'b', 'c', 'a3'])
def test_setslice_range(self):
e = self._make_elem_with_children(4)
e[1:3] = [ET.Element('b%s' % i) for i in range(2)]
self.assertEqual(self._subelem_tags(e), ['a0', 'b0', 'b1', 'a3'])
e = self._make_elem_with_children(4)
e[1:3] = [ET.Element('b')]
self.assertEqual(self._subelem_tags(e), ['a0', 'b', 'a3'])
e = self._make_elem_with_children(4)
e[1:3] = [ET.Element('b%s' % i) for i in range(3)]
self.assertEqual(self._subelem_tags(e), ['a0', 'b0', 'b1', 'b2', 'a3'])
def test_setslice_steps(self):
e = self._make_elem_with_children(6)
e[1:5:2] = [ET.Element('b%s' % i) for i in range(2)]
self.assertEqual(self._subelem_tags(e), ['a0', 'b0', 'a2', 'b1', 'a4', 'a5'])
e = self._make_elem_with_children(6)
with self.assertRaises(ValueError):
e[1:5:2] = [ET.Element('b')]
with self.assertRaises(ValueError):
e[1:5:2] = [ET.Element('b%s' % i) for i in range(3)]
with self.assertRaises(ValueError):
e[1:5:2] = []
self.assertEqual(self._subelem_tags(e), ['a0', 'a1', 'a2', 'a3', 'a4', 'a5'])
e = self._make_elem_with_children(4)
e[1::sys.maxsize] = [ET.Element('b')]
self.assertEqual(self._subelem_tags(e), ['a0', 'b', 'a2', 'a3'])
e[1::sys.maxsize<<64] = [ET.Element('c')]
self.assertEqual(self._subelem_tags(e), ['a0', 'c', 'a2', 'a3'])
def test_setslice_negative_steps(self):
e = self._make_elem_with_children(4)
e[2:0:-1] = [ET.Element('b%s' % i) for i in range(2)]
self.assertEqual(self._subelem_tags(e), ['a0', 'b1', 'b0', 'a3'])
e = self._make_elem_with_children(4)
with self.assertRaises(ValueError):
e[2:0:-1] = [ET.Element('b')]
with self.assertRaises(ValueError):
e[2:0:-1] = [ET.Element('b%s' % i) for i in range(3)]
with self.assertRaises(ValueError):
e[2:0:-1] = []
self.assertEqual(self._subelem_tags(e), ['a0', 'a1', 'a2', 'a3'])
e = self._make_elem_with_children(4)
e[1::-sys.maxsize] = [ET.Element('b')]
self.assertEqual(self._subelem_tags(e), ['a0', 'b', 'a2', 'a3'])
e[1::-sys.maxsize-1] = [ET.Element('c')]
self.assertEqual(self._subelem_tags(e), ['a0', 'c', 'a2', 'a3'])
e[1::-sys.maxsize<<64] = [ET.Element('d')]
self.assertEqual(self._subelem_tags(e), ['a0', 'd', 'a2', 'a3'])
class IOTest(unittest.TestCase):
def test_encoding(self):
# Test encoding issues.
elem = ET.Element("tag")
elem.text = "abc"
self.assertEqual(serialize(elem), 'abc')
for enc in ("utf-8", "us-ascii"):
with self.subTest(enc):
self.assertEqual(serialize(elem, encoding=enc),
b'abc')
self.assertEqual(serialize(elem, encoding=enc.upper()),
b'abc')
for enc in ("iso-8859-1", "utf-16", "utf-32"):
with self.subTest(enc):
self.assertEqual(serialize(elem, encoding=enc),
("\n"
"abc" % enc).encode(enc))
upper = enc.upper()
self.assertEqual(serialize(elem, encoding=upper),
("\n"
"abc" % upper).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):
self.addCleanup(os_helper.unlink, TESTFN)
tree = ET.ElementTree(ET.XML('''\xf8'''))
tree.write(TESTFN)
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(), b'''ø''')
def test_write_to_filename_with_encoding(self):
self.addCleanup(os_helper.unlink, TESTFN)
tree = ET.ElementTree(ET.XML('''\xf8'''))
tree.write(TESTFN, encoding='utf-8')
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(), b'''\xc3\xb8''')
tree.write(TESTFN, encoding='ISO-8859-1')
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(), convlinesep(
b'''\n'''
b'''\xf8'''))
def test_write_to_filename_as_unicode(self):
self.addCleanup(os_helper.unlink, TESTFN)
with open(TESTFN, 'w') as f:
encoding = f.encoding
os_helper.unlink(TESTFN)
tree = ET.ElementTree(ET.XML('''\xf8'''))
tree.write(TESTFN, encoding='unicode')
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(), b"\xc3\xb8")
def test_write_to_text_file(self):
self.addCleanup(os_helper.unlink, TESTFN)
tree = ET.ElementTree(ET.XML('''\xf8'''))
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'''\xc3\xb8''')
with open(TESTFN, 'w', encoding='ascii', errors='xmlcharrefreplace') as f:
tree.write(f, encoding='unicode')
self.assertFalse(f.closed)
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(), b'''ø''')
with open(TESTFN, 'w', encoding='ISO-8859-1') as f:
tree.write(f, encoding='unicode')
self.assertFalse(f.closed)
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(), b'''\xf8''')
def test_write_to_binary_file(self):
self.addCleanup(os_helper.unlink, TESTFN)
tree = ET.ElementTree(ET.XML('''\xf8'''))
with open(TESTFN, 'wb') as f:
tree.write(f)
self.assertFalse(f.closed)
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(), b'''ø''')
def test_write_to_binary_file_with_encoding(self):
self.addCleanup(os_helper.unlink, TESTFN)
tree = ET.ElementTree(ET.XML('''\xf8'''))
with open(TESTFN, 'wb') as f:
tree.write(f, encoding='utf-8')
self.assertFalse(f.closed)
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(), b'''\xc3\xb8''')
with open(TESTFN, 'wb') as f:
tree.write(f, encoding='ISO-8859-1')
self.assertFalse(f.closed)
with open(TESTFN, 'rb') as f:
self.assertEqual(f.read(),
b'''\n'''
b'''\xf8''')
def test_write_to_binary_file_with_bom(self):
self.addCleanup(os_helper.unlink, TESTFN)
tree = ET.ElementTree(ET.XML('''\xf8'''))
# 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'''
'''\xf8'''.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'''
'''\xf8'''.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('''\xf8'''))
stream = io.StringIO()
tree.write(stream, encoding='unicode')
self.assertEqual(stream.getvalue(), '''\xf8''')
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('''\xf8'''))
raw = io.BytesIO()
tree.write(raw)
self.assertEqual(raw.getvalue(), b'''ø''')
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('''\xf8'''))
stream = io.StringIO()
writer = self.dummy()
writer.write = stream.write
tree.write(writer, encoding='unicode')
self.assertEqual(stream.getvalue(), '''\xf8''')
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('''\xf8'''))
raw = io.BytesIO()
writer = self.dummy()
writer.write = raw.write
tree.write(writer)
self.assertEqual(raw.getvalue(), b'''ø''')
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')))
def test_short_empty_elements(self):
root = ET.fromstring('abc')
self.assertEqual(
ET.tostring(root, 'unicode'),
'abc')
self.assertEqual(
ET.tostring(root, 'unicode', short_empty_elements=True),
'abc')
self.assertEqual(
ET.tostring(root, 'unicode', short_empty_elements=False),
'abc')
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):
# The type of methods defined in Python code is types.FunctionType,
# while the type of methods defined inside _elementtree is
#
self.assertIsInstance(pyET.Element.__init__, types.FunctionType)
self.assertIsInstance(pyET.XMLParser.__init__, types.FunctionType)
# --------------------------------------------------------------------
def c14n_roundtrip(xml, **options):
return pyET.canonicalize(xml, **options)
class C14NTest(unittest.TestCase):
maxDiff = None
#
# simple roundtrip tests (from c14n.py)
def test_simple_roundtrip(self):
# Basics
self.assertEqual(c14n_roundtrip(""), '')
self.assertEqual(c14n_roundtrip(""), # FIXME
'')
self.assertEqual(c14n_roundtrip(""),
'')
self.assertEqual(c14n_roundtrip(""),
'')
self.assertEqual(c14n_roundtrip(""),
'')
# C14N spec
self.assertEqual(c14n_roundtrip("Hello, world!"),
'Hello, world!')
self.assertEqual(c14n_roundtrip("2"),
'2')
self.assertEqual(c14n_roundtrip('"0" && value<"10" ?"valid":"error"]]>'),
'value>"0" && value<"10" ?"valid":"error"')
self.assertEqual(c14n_roundtrip('''valid'''),
'valid')
self.assertEqual(c14n_roundtrip(""),
'')
self.assertEqual(c14n_roundtrip(""),
'')
self.assertEqual(c14n_roundtrip(""),
'')
# fragments from PJ's tests
#self.assertEqual(c14n_roundtrip(""),
#'')
# Namespace issues
xml = ''
self.assertEqual(c14n_roundtrip(xml), xml)
xml = ''
self.assertEqual(c14n_roundtrip(xml), xml)
xml = ''
self.assertEqual(c14n_roundtrip(xml), xml)
def test_c14n_exclusion(self):
xml = textwrap.dedent("""\
abtext
btext
dtext
""")
self.assertEqual(
c14n_roundtrip(xml, strip_text=True),
''
'abtext'
'btext'
'dtext'
'')
self.assertEqual(
c14n_roundtrip(xml, strip_text=True, exclude_attrs=['{http://example.com/x}attr']),
''
'abtext'
'btext'
'dtext'
'')
self.assertEqual(
c14n_roundtrip(xml, strip_text=True, exclude_tags=['{http://example.com/x}d']),
''
'abtext'
'btext'
''
'')
self.assertEqual(
c14n_roundtrip(xml, strip_text=True, exclude_attrs=['{http://example.com/x}attr'],
exclude_tags=['{http://example.com/x}d']),
''
'abtext'
'btext'
''
'')
self.assertEqual(
c14n_roundtrip(xml, strip_text=True, exclude_tags=['a', 'b']),
''
'dtext'
'')
self.assertEqual(
c14n_roundtrip(xml, exclude_tags=['a', 'b']),
'\n'
' \n'
' \n'
' \n'
' dtext\n'
' \n'
'')
self.assertEqual(
c14n_roundtrip(xml, strip_text=True, exclude_tags=['{http://example.com/x}d', 'b']),
''
''
''
'')
self.assertEqual(
c14n_roundtrip(xml, exclude_tags=['{http://example.com/x}d', 'b']),
'\n'
' \n'
' \n'
' \n'
' \n'
' \n'
' \n'
' \n'
'')
#
# basic method=c14n tests from the c14n 2.0 specification. uses
# test files under xmltestdata/c14n-20.
# note that this uses generated C14N versions of the standard ET.write
# output, not roundtripped C14N (see above).
def test_xml_c14n2(self):
datadir = findfile("c14n-20", subdir="xmltestdata")
full_path = partial(os.path.join, datadir)
files = [filename[:-4] for filename in sorted(os.listdir(datadir))
if filename.endswith('.xml')]
input_files = [
filename for filename in files
if filename.startswith('in')
]
configs = {
filename: {
# sequential
option.tag.split('}')[-1]: ((option.text or '').strip(), option)
for option in ET.parse(full_path(filename) + ".xml").getroot()
}
for filename in files
if filename.startswith('c14n')
}
tests = {
input_file: [
(filename, configs[filename.rsplit('_', 1)[-1]])
for filename in files
if filename.startswith(f'out_{input_file}_')
and filename.rsplit('_', 1)[-1] in configs
]
for input_file in input_files
}
# Make sure we found all test cases.
self.assertEqual(30, len([
output_file for output_files in tests.values()
for output_file in output_files]))
def get_option(config, option_name, default=None):
return config.get(option_name, (default, ()))[0]
for input_file, output_files in tests.items():
for output_file, config in output_files:
keep_comments = get_option(
config, 'IgnoreComments') == 'true' # no, it's right :)
strip_text = get_option(
config, 'TrimTextNodes') == 'true'
rewrite_prefixes = get_option(
config, 'PrefixRewrite') == 'sequential'
if 'QNameAware' in config:
qattrs = [
f"{{{el.get('NS')}}}{el.get('Name')}"
for el in config['QNameAware'][1].findall(
'{http://www.w3.org/2010/xml-c14n2}QualifiedAttr')
]
qtags = [
f"{{{el.get('NS')}}}{el.get('Name')}"
for el in config['QNameAware'][1].findall(
'{http://www.w3.org/2010/xml-c14n2}Element')
]
else:
qtags = qattrs = None
# Build subtest description from config.
config_descr = ','.join(
f"{name}={value or ','.join(c.tag.split('}')[-1] for c in children)}"
for name, (value, children) in sorted(config.items())
)
with self.subTest(f"{output_file}({config_descr})"):
if input_file == 'inNsRedecl' and not rewrite_prefixes:
self.skipTest(
f"Redeclared namespace handling is not supported in {output_file}")
if input_file == 'inNsSuperfluous' and not rewrite_prefixes:
self.skipTest(
f"Redeclared namespace handling is not supported in {output_file}")
if 'QNameAware' in config and config['QNameAware'][1].find(
'{http://www.w3.org/2010/xml-c14n2}XPathElement') is not None:
self.skipTest(
f"QName rewriting in XPath text is not supported in {output_file}")
f = full_path(input_file + ".xml")
if input_file == 'inC14N5':
# Hack: avoid setting up external entity resolution in the parser.
with open(full_path('world.txt'), 'rb') as entity_file:
with open(f, 'rb') as f:
f = io.BytesIO(f.read().replace(b'&ent2;', entity_file.read()))
text = ET.canonicalize(
from_file=f,
with_comments=keep_comments,
strip_text=strip_text,
rewrite_prefixes=rewrite_prefixes,
qname_aware_tags=qtags, qname_aware_attrs=qattrs)
with open(full_path(output_file + ".xml"), 'r', encoding='utf8') as f:
expected = f.read()
if input_file == 'inC14N3':
# FIXME: cET resolves default attributes but ET does not!
expected = expected.replace(' attr="default"', '')
text = text.replace(' attr="default"', '')
self.assertEqual(expected, text)
# --------------------------------------------------------------------
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 = [
ModuleTest,
ElementSlicingTest,
BasicElementTest,
BadElementTest,
BadElementPathTest,
ElementTreeTest,
IOTest,
ParseErrorTest,
XIncludeTest,
ElementTreeTypeTest,
ElementFindTest,
ElementIterTest,
TreeBuilderTest,
XMLParserTest,
XMLPullParserTest,
BugsTest,
KeywordArgsTest,
C14NTest,
]
# 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,
])
# Provide default namespace mapping and path cache.
from xml.etree import ElementPath
nsmap = ET.register_namespace._namespace_map
# Copy the default namespace mapping
nsmap_copy = nsmap.copy()
# Copy the path cache (should be empty)
path_cache = ElementPath._cache
ElementPath._cache = path_cache.copy()
# Align the Comment/PI factories.
if hasattr(ET, '_set_factories'):
old_factories = ET._set_factories(ET.Comment, ET.PI)
else:
old_factories = None
try:
support.run_unittest(*test_classes)
finally:
from xml.etree import ElementPath
# Restore mapping and path cache
nsmap.clear()
nsmap.update(nsmap_copy)
ElementPath._cache = path_cache
if old_factories is not None:
ET._set_factories(*old_factories)
# don't interfere with subsequent tests
ET = pyET = None
if __name__ == '__main__':
test_main()