diff options
Diffstat (limited to 'Lib/test/test_xml_etree.py')
| -rw-r--r-- | Lib/test/test_xml_etree.py | 1014 |
1 files changed, 790 insertions, 224 deletions
diff --git a/Lib/test/test_xml_etree.py b/Lib/test/test_xml_etree.py index 3092f3b..9cebc3c 100644 --- a/Lib/test/test_xml_etree.py +++ b/Lib/test/test_xml_etree.py @@ -1,28 +1,34 @@ # xml.etree test. This file contains enough tests to make sure that # all included components work as they should. # Large parts are extracted from the upstream test suite. - -# IMPORTANT: the same doctests are run from "test_xml_etree_c" in -# order to ensure consistency between the C implementation and the -# Python implementation. +# +# PLEASE write all new tests using the standard unittest infrastructure and +# not doctest. +# +# IMPORTANT: the same tests are run from "test_xml_etree_c" in order +# to ensure consistency between the C implementation and the Python +# implementation. # # For this purpose, the module-level "ET" symbol is temporarily # monkey-patched when running the "test_xml_etree_c" test suite. # Don't re-import "xml.etree.ElementTree" module in the docstring, # except if the test is specific to the Python implementation. -import sys import html +import io +import sys import unittest +import weakref from test import support -from test.support import findfile +from test.support import TESTFN, findfile, unlink, import_fresh_module, gc_collect -from xml.etree import ElementTree as ET +pyET = None +ET = None SIMPLE_XMLFILE = findfile("simple.xml", subdir="xmltestdata") try: - SIMPLE_XMLFILE.encode("utf8") + SIMPLE_XMLFILE.encode("utf-8") except UnicodeEncodeError: raise unittest.SkipTest("filename is not encodable to utf8") SIMPLE_NS_XMLFILE = findfile("simple-ns.xml", subdir="xmltestdata") @@ -57,6 +63,22 @@ SAMPLE_XML_NS = """ </body> """ +SAMPLE_XML_NS_ELEMS = """ +<root> +<h:table xmlns:h="hello"> + <h:tr> + <h:td>Apples</h:td> + <h:td>Bananas</h:td> + </h:tr> +</h:table> + +<f:table xmlns:f="foo"> + <f:name>African Coffee Table</f:name> + <f:width>80</f:width> + <f:length>120</f:length> +</f:table> +</root> +""" def sanity(): """ @@ -188,10 +210,8 @@ def interface(): These methods return an iterable. See bug 6472. - >>> check_method(element.iter("tag").__next__) >>> check_method(element.iterfind("tag").__next__) >>> check_method(element.iterfind("*").__next__) - >>> check_method(tree.iter("tag").__next__) >>> check_method(tree.iterfind("tag").__next__) >>> check_method(tree.iterfind("*").__next__) @@ -270,42 +290,6 @@ def cdata(): '<tag>hello</tag>' """ -# Only with Python implementation -def simplefind(): - """ - Test find methods using the elementpath fallback. - - >>> from xml.etree import ElementTree - - >>> CurrentElementPath = ElementTree.ElementPath - >>> ElementTree.ElementPath = ElementTree._SimpleElementPath() - >>> elem = ElementTree.XML(SAMPLE_XML) - >>> elem.find("tag").tag - 'tag' - >>> ElementTree.ElementTree(elem).find("tag").tag - 'tag' - >>> elem.findtext("tag") - 'text' - >>> elem.findtext("tog") - >>> elem.findtext("tog", "default") - 'default' - >>> ElementTree.ElementTree(elem).findtext("tag") - 'text' - >>> summarize_list(elem.findall("tag")) - ['tag', 'tag'] - >>> summarize_list(elem.findall(".//tag")) - ['tag', 'tag', 'tag'] - - Path syntax doesn't work in this case. - - >>> elem.find("section/tag") - >>> elem.findtext("section/tag") - >>> summarize_list(elem.findall("section/tag")) - [] - - >>> ElementTree.ElementPath = CurrentElementPath - """ - def find(): """ Test find methods (including xpath syntax). @@ -460,17 +444,19 @@ def path_cache(): """ Check that the path cache behaves sanely. + >>> from xml.etree import ElementPath + >>> elem = ET.XML(SAMPLE_XML) >>> for i in range(10): ET.ElementTree(elem).find('./'+str(i)) - >>> cache_len_10 = len(ET.ElementPath._cache) + >>> cache_len_10 = len(ElementPath._cache) >>> for i in range(10): ET.ElementTree(elem).find('./'+str(i)) - >>> len(ET.ElementPath._cache) == cache_len_10 + >>> len(ElementPath._cache) == cache_len_10 True >>> for i in range(20): ET.ElementTree(elem).find('./'+str(i)) - >>> len(ET.ElementPath._cache) > cache_len_10 + >>> len(ElementPath._cache) > cache_len_10 True >>> for i in range(600): ET.ElementTree(elem).find('./'+str(i)) - >>> len(ET.ElementPath._cache) < 500 + >>> len(ElementPath._cache) < 500 True """ @@ -902,65 +888,6 @@ def check_encoding(encoding): """ ET.XML("<?xml version='1.0' encoding='%s'?><xml />" % encoding) -def encoding(): - r""" - Test encoding issues. - - >>> elem = ET.Element("tag") - >>> elem.text = "abc" - >>> serialize(elem) - '<tag>abc</tag>' - >>> serialize(elem, encoding="utf-8") - b'<tag>abc</tag>' - >>> serialize(elem, encoding="us-ascii") - b'<tag>abc</tag>' - >>> serialize(elem, encoding="iso-8859-1") - b"<?xml version='1.0' encoding='iso-8859-1'?>\n<tag>abc</tag>" - - >>> elem.text = "<&\"\'>" - >>> serialize(elem) - '<tag><&"\'></tag>' - >>> serialize(elem, encoding="utf-8") - b'<tag><&"\'></tag>' - >>> serialize(elem, encoding="us-ascii") # cdata characters - b'<tag><&"\'></tag>' - >>> serialize(elem, encoding="iso-8859-1") - b'<?xml version=\'1.0\' encoding=\'iso-8859-1\'?>\n<tag><&"\'></tag>' - - >>> elem.attrib["key"] = "<&\"\'>" - >>> elem.text = None - >>> serialize(elem) - '<tag key="<&"\'>" />' - >>> serialize(elem, encoding="utf-8") - b'<tag key="<&"\'>" />' - >>> serialize(elem, encoding="us-ascii") - b'<tag key="<&"\'>" />' - >>> serialize(elem, encoding="iso-8859-1") - b'<?xml version=\'1.0\' encoding=\'iso-8859-1\'?>\n<tag key="<&"\'>" />' - - >>> elem.text = '\xe5\xf6\xf6<>' - >>> elem.attrib.clear() - >>> serialize(elem) - '<tag>\xe5\xf6\xf6<></tag>' - >>> serialize(elem, encoding="utf-8") - b'<tag>\xc3\xa5\xc3\xb6\xc3\xb6<></tag>' - >>> serialize(elem, encoding="us-ascii") - b'<tag>åöö<></tag>' - >>> serialize(elem, encoding="iso-8859-1") - b"<?xml version='1.0' encoding='iso-8859-1'?>\n<tag>\xe5\xf6\xf6<></tag>" - - >>> elem.attrib["key"] = '\xe5\xf6\xf6<>' - >>> elem.text = None - >>> serialize(elem) - '<tag key="\xe5\xf6\xf6<>" />' - >>> serialize(elem, encoding="utf-8") - b'<tag key="\xc3\xa5\xc3\xb6\xc3\xb6<>" />' - >>> serialize(elem, encoding="us-ascii") - b'<tag key="åöö<>" />' - >>> serialize(elem, encoding="iso-8859-1") - b'<?xml version=\'1.0\' encoding=\'iso-8859-1\'?>\n<tag key="\xe5\xf6\xf6<>" />' - """ - def methods(): r""" Test serialization methods. @@ -979,36 +906,6 @@ def methods(): '1 < 2\n' """ -def iterators(): - """ - Test iterators. - - >>> e = ET.XML("<html><body>this is a <i>paragraph</i>.</body>..</html>") - >>> summarize_list(e.iter()) - ['html', 'body', 'i'] - >>> summarize_list(e.find("body").iter()) - ['body', 'i'] - >>> summarize(next(e.iter())) - 'html' - >>> "".join(e.itertext()) - 'this is a paragraph...' - >>> "".join(e.find("body").itertext()) - 'this is a paragraph.' - >>> next(e.itertext()) - 'this is a ' - - Method iterparse should return an iterator. See bug 6472. - - >>> sourcefile = serialize(e, to_string=False) - >>> next(ET.iterparse(sourcefile)) # doctest: +ELLIPSIS - ('end', <Element 'i' at 0x...>) - - >>> tree = ET.ElementTree(None) - >>> tree.iter() - Traceback (most recent call last): - AttributeError: 'NoneType' object has no attribute 'iter' - """ - ENTITY_XML = """\ <!DOCTYPE points [ <!ENTITY % user-entities SYSTEM 'user-entities.xml'> @@ -1049,26 +946,6 @@ def entity(): '<document>text</document>' """ -def error(xml): - """ - - Test error handling. - - >>> issubclass(ET.ParseError, SyntaxError) - True - >>> error("foo").position - (1, 0) - >>> error("<tag>&foo;</tag>").position - (1, 5) - >>> error("foobar<").position - (1, 6) - - """ - try: - ET.XML(xml) - except ET.ParseError: - return sys.exc_info()[1] - def namespace(): """ Test namespace issues. @@ -1256,8 +1133,8 @@ def processinginstruction(): >>> ET.tostring(ET.PI('test', '<testing&>')) b'<?test <testing&>?>' - >>> ET.tostring(ET.PI('test', '<testing&>\xe3'), 'latin1') - b"<?xml version='1.0' encoding='latin1'?>\\n<?test <testing&>\\xe3?>" + >>> ET.tostring(ET.PI('test', '<testing&>\xe3'), 'latin-1') + b"<?xml version='1.0' encoding='latin-1'?>\\n<?test <testing&>\\xe3?>" """ # @@ -1336,21 +1213,20 @@ XINCLUDE["default.xml"] = """\ </document> """.format(html.escape(SIMPLE_XMLFILE, True)) + def xinclude_loader(href, parse="xml", encoding=None): try: data = XINCLUDE[href] except KeyError: - raise IOError("resource not found") + raise OSError("resource not found") if parse == "xml": - from xml.etree.ElementTree import XML - return XML(data) + data = ET.XML(data) return data def xinclude(): r""" Basic inclusion example (XInclude C.1) - >>> from xml.etree import ElementTree as ET >>> from xml.etree import ElementInclude >>> document = xinclude_loader("C1.xml") @@ -1405,26 +1281,10 @@ def xinclude(): >>> document = xinclude_loader("C5.xml") >>> ElementInclude.include(document, xinclude_loader) Traceback (most recent call last): - IOError: resource not found + OSError: resource not found >>> # print(serialize(document)) # C5 """ -def xinclude_default(): - """ - >>> from xml.etree import ElementInclude - - >>> document = xinclude_loader("default.xml") - >>> ElementInclude.include(document) - >>> print(serialize(document)) # default - <document> - <p>Example.</p> - <root> - <element key="value">text</element> - <element>text</element>tail - <empty-element /> - </root> - </document> - """ # # badly formatted xi:include tags @@ -1612,7 +1472,7 @@ def bug_xmltoolkit55(): class ExceptionFile: def read(self, x): - raise IOError + raise OSError def xmltoolkit60(): """ @@ -1620,7 +1480,7 @@ def xmltoolkit60(): Handle crash in stream source. >>> tree = ET.parse(ExceptionFile()) Traceback (most recent call last): - IOError + OSError """ @@ -1851,25 +1711,706 @@ def check_issue10777(): >>> ET.register_namespace('test10777', 'http://myuri/') """ -def check_html_empty_elems_serialization(self): - # issue 15970 - # from http://www.w3.org/TR/html401/index/elements.html - """ +# -------------------------------------------------------------------- - >>> empty_elems = ['AREA', 'BASE', 'BASEFONT', 'BR', 'COL', 'FRAME', 'HR', - ... 'IMG', 'INPUT', 'ISINDEX', 'LINK', 'META', 'PARAM'] - >>> elems = ''.join('<%s />' % elem for elem in empty_elems) - >>> serialize(ET.XML('<html>%s</html>' % elems), method='html') - '<html><AREA><BASE><BASEFONT><BR><COL><FRAME><HR><IMG><INPUT><ISINDEX><LINK><META><PARAM></html>' - >>> serialize(ET.XML('<html>%s</html>' % elems.lower()), method='html') - '<html><area><base><basefont><br><col><frame><hr><img><input><isindex><link><meta><param></html>' - >>> elems = ''.join('<%s></%s>' % (elem, elem) for elem in empty_elems) - >>> serialize(ET.XML('<html>%s</html>' % elems), method='html') - '<html><AREA><BASE><BASEFONT><BR><COL><FRAME><HR><IMG><INPUT><ISINDEX><LINK><META><PARAM></html>' - >>> serialize(ET.XML('<html>%s</html>' % elems.lower()), method='html') - '<html><area><base><basefont><br><col><frame><hr><img><input><isindex><link><meta><param></html>' - """ +class BasicElementTest(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) + + +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></%s>' % (elem,elem)), + method='html') + self.assertEqual(serialized, expected) + +class ElementIterTest(unittest.TestCase): + def _ilist(self, elem, tag=None): + return summarize_list(elem.iter(tag)) + + def test_basic(self): + doc = ET.XML("<html><body>this is a <i>paragraph</i>.</body>..</html>") + 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) + + 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(''' + <document> + <house> + <room>bedroom1</room> + <room>bedroom2</room> + </house> + <shed>nothing here + </shed> + <house> + <room>bedroom8</room> + </house> + </document>''') + + self.assertEqual(self._ilist(doc, 'room'), ['room'] * 3) + self.assertEqual(self._ilist(doc, 'house'), ['house'] * 2) + + # 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 = ('<!DOCTYPE html PUBLIC' + ' "-//W3C//DTD XHTML 1.0 Transitional//EN"' + ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">' + '<html>text<div>subtext</div>tail</html>') + + sample2 = '''<toplevel>sometext</toplevel>''' + + 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(), '''<document> + <p>Example.</p> + <root> + <element key="value">text</element> + <element>text</element>tail + <empty-element /> +</root> +</document>''') + + +class XMLParserTest(unittest.TestCase): + sample1 = '<file><line>22</line></file>' + sample2 = ('<!DOCTYPE html PUBLIC' + ' "-//W3C//DTD XHTML 1.0 Transitional//EN"' + ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">' + '<html>text</html>') + + 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), '<tag>abc</tag>') + self.assertEqual(serialize(elem, encoding="utf-8"), + b'<tag>abc</tag>') + self.assertEqual(serialize(elem, encoding="us-ascii"), + b'<tag>abc</tag>') + for enc in ("iso-8859-1", "utf-16", "utf-32"): + self.assertEqual(serialize(elem, encoding=enc), + ("<?xml version='1.0' encoding='%s'?>\n" + "<tag>abc</tag>" % enc).encode(enc)) + + elem = ET.Element("tag") + elem.text = "<&\"\'>" + self.assertEqual(serialize(elem), '<tag><&"\'></tag>') + self.assertEqual(serialize(elem, encoding="utf-8"), + b'<tag><&"\'></tag>') + self.assertEqual(serialize(elem, encoding="us-ascii"), + b'<tag><&"\'></tag>') + for enc in ("iso-8859-1", "utf-16", "utf-32"): + self.assertEqual(serialize(elem, encoding=enc), + ("<?xml version='1.0' encoding='%s'?>\n" + "<tag><&\"'></tag>" % enc).encode(enc)) + + elem = ET.Element("tag") + elem.attrib["key"] = "<&\"\'>" + self.assertEqual(serialize(elem), '<tag key="<&"\'>" />') + self.assertEqual(serialize(elem, encoding="utf-8"), + b'<tag key="<&"\'>" />') + self.assertEqual(serialize(elem, encoding="us-ascii"), + b'<tag key="<&"\'>" />') + for enc in ("iso-8859-1", "utf-16", "utf-32"): + self.assertEqual(serialize(elem, encoding=enc), + ("<?xml version='1.0' encoding='%s'?>\n" + "<tag key=\"<&"'>\" />" % enc).encode(enc)) + + elem = ET.Element("tag") + elem.text = '\xe5\xf6\xf6<>' + self.assertEqual(serialize(elem), '<tag>\xe5\xf6\xf6<></tag>') + self.assertEqual(serialize(elem, encoding="utf-8"), + b'<tag>\xc3\xa5\xc3\xb6\xc3\xb6<></tag>') + self.assertEqual(serialize(elem, encoding="us-ascii"), + b'<tag>åöö<></tag>') + for enc in ("iso-8859-1", "utf-16", "utf-32"): + self.assertEqual(serialize(elem, encoding=enc), + ("<?xml version='1.0' encoding='%s'?>\n" + "<tag>åöö<></tag>" % enc).encode(enc)) + + elem = ET.Element("tag") + elem.attrib["key"] = '\xe5\xf6\xf6<>' + self.assertEqual(serialize(elem), '<tag key="\xe5\xf6\xf6<>" />') + self.assertEqual(serialize(elem, encoding="utf-8"), + b'<tag key="\xc3\xa5\xc3\xb6\xc3\xb6<>" />') + self.assertEqual(serialize(elem, encoding="us-ascii"), + b'<tag key="åöö<>" />') + for enc in ("iso-8859-1", "utf-16", "utf-16le", "utf-16be", "utf-32"): + self.assertEqual(serialize(elem, encoding=enc), + ("<?xml version='1.0' encoding='%s'?>\n" + "<tag key=\"åöö<>\" />" % enc).encode(enc)) + + def test_write_to_filename(self): + tree = ET.ElementTree(ET.XML('''<site />''')) + tree.write(TESTFN) + with open(TESTFN, 'rb') as f: + self.assertEqual(f.read(), b'''<site />''') + + def test_write_to_text_file(self): + tree = ET.ElementTree(ET.XML('''<site />''')) + 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'''<site />''') + + def test_write_to_binary_file(self): + tree = ET.ElementTree(ET.XML('''<site />''')) + with open(TESTFN, 'wb') as f: + tree.write(f) + self.assertFalse(f.closed) + with open(TESTFN, 'rb') as f: + self.assertEqual(f.read(), b'''<site />''') + + def test_write_to_binary_file_with_bom(self): + tree = ET.ElementTree(ET.XML('''<site />''')) + # 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(), + '''<?xml version='1.0' encoding='utf-16'?>\n''' + '''<site />'''.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(), + '''<?xml version='1.0' encoding='utf-16'?>\n''' + '''<site />'''.encode("utf-16")) + + def test_read_from_stringio(self): + tree = ET.ElementTree() + stream = io.StringIO('''<?xml version="1.0"?><site></site>''') + tree.parse(stream) + self.assertEqual(tree.getroot().tag, 'site') + + def test_write_to_stringio(self): + tree = ET.ElementTree(ET.XML('''<site />''')) + stream = io.StringIO() + tree.write(stream, encoding='unicode') + self.assertEqual(stream.getvalue(), '''<site />''') + + def test_read_from_bytesio(self): + tree = ET.ElementTree() + raw = io.BytesIO(b'''<?xml version="1.0"?><site></site>''') + tree.parse(raw) + self.assertEqual(tree.getroot().tag, 'site') + + def test_write_to_bytesio(self): + tree = ET.ElementTree(ET.XML('''<site />''')) + raw = io.BytesIO() + tree.write(raw) + self.assertEqual(raw.getvalue(), b'''<site />''') + + class dummy: + pass + + def test_read_from_user_text_reader(self): + stream = io.StringIO('''<?xml version="1.0"?><site></site>''') + 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('''<site />''')) + stream = io.StringIO() + writer = self.dummy() + writer.write = stream.write + tree.write(writer, encoding='unicode') + self.assertEqual(stream.getvalue(), '''<site />''') + + def test_read_from_user_binary_reader(self): + raw = io.BytesIO(b'''<?xml version="1.0"?><site></site>''') + 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('''<site />''')) + raw = io.BytesIO() + writer = self.dummy() + writer.write = raw.write + tree.write(writer) + self.assertEqual(raw.getvalue(), b'''<site />''') + + def test_write_to_user_binary_writer_with_bom(self): + tree = ET.ElementTree(ET.XML('''<site />''')) + 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(), + '''<?xml version='1.0' encoding='utf-16'?>\n''' + '''<site />'''.encode("utf-16")) + + def test_tostringlist_invariant(self): + root = ET.fromstring('<tag>foo</tag>') + 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('<tag>&foo;</tag>').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("<a>foo</a>") + 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") + +# -------------------------------------------------------------------- + +@unittest.skipUnless(pyET, 'only for the Python version') +class NoAcceleratorTest(unittest.TestCase): + # Test that the C accelerator was not imported for pyET + def test_correct_import_pyET(self): + self.assertEqual(pyET.Element.__module__, 'xml.etree.ElementTree') + self.assertEqual(pyET.SubElement.__module__, 'xml.etree.ElementTree') # -------------------------------------------------------------------- @@ -1892,44 +2433,69 @@ class CleanContext(object): ("This method will be removed in future versions. " "Use .+ instead.", DeprecationWarning), ("This method will be removed in future versions. " - "Use .+ instead.", PendingDeprecationWarning), - # XMLParser.doctype() is deprecated. - ("This method of XMLParser is deprecated. Define doctype.. " - "method on the TreeBuilder target.", DeprecationWarning)) + "Use .+ instead.", PendingDeprecationWarning)) self.checkwarnings = support.check_warnings(*deprecations, quiet=quiet) def __enter__(self): - from xml.etree import ElementTree - self._nsmap = ElementTree._namespace_map - self._path_cache = ElementTree.ElementPath._cache + from xml.etree import ElementPath + self._nsmap = ET.register_namespace._namespace_map # Copy the default namespace mapping - ElementTree._namespace_map = self._nsmap.copy() + self._nsmap_copy = self._nsmap.copy() # Copy the path cache (should be empty) - ElementTree.ElementPath._cache = self._path_cache.copy() + self._path_cache = ElementPath._cache + ElementPath._cache = self._path_cache.copy() self.checkwarnings.__enter__() def __exit__(self, *args): - from xml.etree import ElementTree + from xml.etree import ElementPath # Restore mapping and path cache - ElementTree._namespace_map = self._nsmap - ElementTree.ElementPath._cache = self._path_cache + self._nsmap.clear() + self._nsmap.update(self._nsmap_copy) + ElementPath._cache = self._path_cache self.checkwarnings.__exit__(*args) -def test_main(module_name='xml.etree.ElementTree'): - from test import test_xml_etree +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. + if module is None: + global pyET + pyET = import_fresh_module('xml.etree.ElementTree', + blocked=['_elementtree']) + module = pyET + + global ET + ET = module + + test_classes = [ + ElementSlicingTest, + BasicElementTest, + IOTest, + ParseErrorTest, + XincludeTest, + ElementTreeTest, + 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: + test_classes.extend([ + NoAcceleratorTest, + ]) - use_py_module = (module_name == 'xml.etree.ElementTree') - - # The same doctests are used for both the Python and the C implementations - assert test_xml_etree.ET.__name__ == module_name + try: + support.run_unittest(*test_classes) - # XXX the C module should give the same warnings as the Python module - with CleanContext(quiet=not use_py_module): - support.run_doctest(test_xml_etree, verbosity=True) + # XXX the C module should give the same warnings as the Python module + with CleanContext(quiet=(module is not pyET)): + support.run_doctest(sys.modules[__name__], verbosity=True) + finally: + # don't interfere with subsequent tests + ET = pyET = None - # The module should not be changed by the tests - assert test_xml_etree.ET.__name__ == module_name if __name__ == '__main__': test_main() |