diff options
Diffstat (limited to 'Lib/test/test_xml_etree.py')
| -rw-r--r-- | Lib/test/test_xml_etree.py | 3754 |
1 files changed, 2066 insertions, 1688 deletions
diff --git a/Lib/test/test_xml_etree.py b/Lib/test/test_xml_etree.py index f61292f..f3683f2 100644 --- a/Lib/test/test_xml_etree.py +++ b/Lib/test/test_xml_etree.py @@ -1,28 +1,32 @@ -# 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. +# 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 operator +import pickle +import sys import unittest +import weakref +from itertools import product from test import support -from test.support import findfile +from test.support import TESTFN, findfile, import_fresh_module, gc_collect -from xml.etree import ElementTree as ET +# pyET is the pure-Python implementation. +# +# ET is pyET in test_xml_etree and is the C accelerated version in +# test_xml_etree_c. +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") @@ -57,22 +61,44 @@ 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> +""" + +ENTITY_XML = """\ +<!DOCTYPE points [ +<!ENTITY % user-entities SYSTEM 'user-entities.xml'> +%user-entities; +]> +<document>&entity;</document> +""" -def sanity(): - """ - Import sanity. - >>> from xml.etree import ElementTree - >>> from xml.etree import ElementInclude - >>> from xml.etree import ElementPath - """ +class ModuleTest(unittest.TestCase): + # TODO: this should be removed once we get rid of the global module vars + + def test_sanity(self): + # Import sanity. + + from xml.etree import ElementTree + from xml.etree import ElementInclude + from xml.etree import ElementPath -def check_method(method): - if not hasattr(method, '__call__'): - print(method, "not callable") def serialize(elem, to_string=True, encoding='unicode', **options): - import io if encoding != 'unicode': file = io.BytesIO() else: @@ -85,1182 +111,797 @@ def serialize(elem, to_string=True, encoding='unicode', **options): file.seek(0) return file -def summarize(elem): - if elem.tag == ET.Comment: - return "<Comment>" - return elem.tag - def summarize_list(seq): - return [summarize(elem) for elem in seq] - -def normalize_crlf(tree): - for elem in tree.iter(): - if elem.text: - elem.text = elem.text.replace("\r\n", "\n") - if elem.tail: - elem.tail = elem.tail.replace("\r\n", "\n") - -def normalize_exception(func, *args, **kwargs): - # Ignore the exception __module__ - try: - func(*args, **kwargs) - except Exception as err: - print("Traceback (most recent call last):") - print("{}: {}".format(err.__class__.__name__, err)) - -def check_string(string): - len(string) - for char in string: - if len(char) != 1: - print("expected one-character string, got %r" % char) - new_string = string + "" - new_string = string + " " - string[:0] - -def check_mapping(mapping): - len(mapping) - keys = mapping.keys() - items = mapping.items() - for key in keys: - item = mapping[key] - mapping["key"] = "value" - if mapping["key"] != "value": - print("expected value string, got %r" % mapping["key"]) - -def check_element(element): - if not ET.iselement(element): - print("not an element") - if not hasattr(element, "tag"): - print("no tag member") - if not hasattr(element, "attrib"): - print("no attrib member") - if not hasattr(element, "text"): - print("no text member") - if not hasattr(element, "tail"): - print("no tail member") - - check_string(element.tag) - check_mapping(element.attrib) - if element.text is not None: - check_string(element.text) - if element.tail is not None: - check_string(element.tail) - for elem in element: - check_element(elem) + return [elem.tag for elem in seq] + + +class ElementTestCase: + @classmethod + def setUpClass(cls): + cls.modules = {pyET, ET} + + def pickleRoundTrip(self, obj, name, dumper, loader): + save_m = sys.modules[name] + try: + sys.modules[name] = dumper + temp = pickle.dumps(obj) + sys.modules[name] = loader + result = pickle.loads(temp) + except pickle.PicklingError as pe: + # pyET must be second, because pyET may be (equal to) ET. + human = dict([(ET, "cET"), (pyET, "pyET")]) + raise support.TestFailed("Failed to round-trip %r from %r to %r" + % (obj, + human.get(dumper, dumper), + human.get(loader, loader))) from pe + finally: + sys.modules[name] = save_m + return result + + def assertEqualElements(self, alice, bob): + self.assertIsInstance(alice, (ET.Element, pyET.Element)) + self.assertIsInstance(bob, (ET.Element, pyET.Element)) + self.assertEqual(len(list(alice)), len(list(bob))) + for x, y in zip(alice, bob): + self.assertEqualElements(x, y) + properties = operator.attrgetter('tag', 'tail', 'text', 'attrib') + self.assertEqual(properties(alice), properties(bob)) # -------------------------------------------------------------------- # element tree tests -def interface(): - """ - Test element tree interface. - - >>> element = ET.Element("tag") - >>> check_element(element) - >>> tree = ET.ElementTree(element) - >>> check_element(tree.getroot()) - - >>> element = ET.Element("t\\xe4g", key="value") - >>> tree = ET.ElementTree(element) - >>> repr(element) # doctest: +ELLIPSIS - "<Element 't\\xe4g' at 0x...>" - >>> element = ET.Element("tag", key="value") - - Make sure all standard element methods exist. - - >>> check_method(element.append) - >>> check_method(element.extend) - >>> check_method(element.insert) - >>> check_method(element.remove) - >>> check_method(element.getchildren) - >>> check_method(element.find) - >>> check_method(element.iterfind) - >>> check_method(element.findall) - >>> check_method(element.findtext) - >>> check_method(element.clear) - >>> check_method(element.get) - >>> check_method(element.set) - >>> check_method(element.keys) - >>> check_method(element.items) - >>> check_method(element.iter) - >>> check_method(element.itertext) - >>> check_method(element.getiterator) - - 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__) - - These aliases are provided: - - >>> assert ET.XML == ET.fromstring - >>> assert ET.PI == ET.ProcessingInstruction - >>> assert ET.XMLParser == ET.XMLTreeBuilder - """ - -def simpleops(): - """ - Basic method sanity checks. - - >>> elem = ET.XML("<body><tag/></body>") - >>> serialize(elem) - '<body><tag /></body>' - >>> e = ET.Element("tag2") - >>> elem.append(e) - >>> serialize(elem) - '<body><tag /><tag2 /></body>' - >>> elem.remove(e) - >>> serialize(elem) - '<body><tag /></body>' - >>> elem.insert(0, e) - >>> serialize(elem) - '<body><tag2 /><tag /></body>' - >>> elem.remove(e) - >>> elem.extend([e]) - >>> serialize(elem) - '<body><tag /><tag2 /></body>' - >>> elem.remove(e) - - >>> element = ET.Element("tag", key="value") - >>> serialize(element) # 1 - '<tag key="value" />' - >>> subelement = ET.Element("subtag") - >>> element.append(subelement) - >>> serialize(element) # 2 - '<tag key="value"><subtag /></tag>' - >>> element.insert(0, subelement) - >>> serialize(element) # 3 - '<tag key="value"><subtag /><subtag /></tag>' - >>> element.remove(subelement) - >>> serialize(element) # 4 - '<tag key="value"><subtag /></tag>' - >>> element.remove(subelement) - >>> serialize(element) # 5 - '<tag key="value" />' - >>> element.remove(subelement) - Traceback (most recent call last): - ValueError: list.remove(x): x not in list - >>> serialize(element) # 6 - '<tag key="value" />' - >>> element[0:0] = [subelement, subelement, subelement] - >>> serialize(element[1]) - '<subtag />' - >>> element[1:9] == [element[1], element[2]] - True - >>> element[:9:2] == [element[0], element[2]] - True - >>> del element[1:2] - >>> serialize(element) - '<tag key="value"><subtag /><subtag /></tag>' - """ - -def cdata(): - """ - Test CDATA handling (etc). - - >>> serialize(ET.XML("<tag>hello</tag>")) - '<tag>hello</tag>' - >>> serialize(ET.XML("<tag>hello</tag>")) - '<tag>hello</tag>' - >>> serialize(ET.XML("<tag><![CDATA[hello]]></tag>")) - '<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). - - >>> elem = ET.XML(SAMPLE_XML) - >>> elem.find("tag").tag - 'tag' - >>> ET.ElementTree(elem).find("tag").tag - 'tag' - >>> elem.find("section/tag").tag - 'tag' - >>> elem.find("./tag").tag - 'tag' - >>> ET.ElementTree(elem).find("./tag").tag - 'tag' - >>> ET.ElementTree(elem).find("/tag").tag - 'tag' - >>> elem[2] = ET.XML(SAMPLE_SECTION) - >>> elem.find("section/nexttag").tag - 'nexttag' - >>> ET.ElementTree(elem).find("section/tag").tag - 'tag' - >>> ET.ElementTree(elem).find("tog") - >>> ET.ElementTree(elem).find("tog/foo") - >>> elem.findtext("tag") - 'text' - >>> elem.findtext("section/nexttag") - '' - >>> elem.findtext("section/nexttag", "default") - '' - >>> elem.findtext("tog") - >>> elem.findtext("tog", "default") - 'default' - >>> ET.ElementTree(elem).findtext("tag") - 'text' - >>> ET.ElementTree(elem).findtext("tog/foo") - >>> ET.ElementTree(elem).findtext("tog/foo", "default") - 'default' - >>> ET.ElementTree(elem).findtext("./tag") - 'text' - >>> ET.ElementTree(elem).findtext("/tag") - 'text' - >>> elem.findtext("section/tag") - 'subtext' - >>> ET.ElementTree(elem).findtext("section/tag") - 'subtext' - >>> ET.XML('<root><empty /></root>').findtext('empty') - '' - >>> summarize_list(elem.findall(".")) - ['body'] - >>> summarize_list(elem.findall("tag")) - ['tag', 'tag'] - >>> summarize_list(elem.findall("tog")) - [] - >>> summarize_list(elem.findall("tog/foo")) - [] - >>> summarize_list(elem.findall("*")) - ['tag', 'tag', 'section'] - >>> summarize_list(elem.findall(".//tag")) - ['tag', 'tag', 'tag', 'tag'] - >>> summarize_list(elem.findall("section/tag")) - ['tag'] - >>> summarize_list(elem.findall("section//tag")) - ['tag', 'tag'] - >>> summarize_list(elem.findall("section/*")) - ['tag', 'nexttag', 'nextsection'] - >>> summarize_list(elem.findall("section//*")) - ['tag', 'nexttag', 'nextsection', 'tag'] - >>> summarize_list(elem.findall("section/.//*")) - ['tag', 'nexttag', 'nextsection', 'tag'] - >>> summarize_list(elem.findall("*/*")) - ['tag', 'nexttag', 'nextsection'] - >>> summarize_list(elem.findall("*//*")) - ['tag', 'nexttag', 'nextsection', 'tag'] - >>> summarize_list(elem.findall("*/tag")) - ['tag'] - >>> summarize_list(elem.findall("*/./tag")) - ['tag'] - >>> summarize_list(elem.findall("./tag")) - ['tag', 'tag'] - >>> summarize_list(elem.findall(".//tag")) - ['tag', 'tag', 'tag', 'tag'] - >>> summarize_list(elem.findall("././tag")) - ['tag', 'tag'] - >>> summarize_list(elem.findall(".//tag[@class]")) - ['tag', 'tag', 'tag'] - >>> summarize_list(elem.findall(".//tag[@class='a']")) - ['tag'] - >>> summarize_list(elem.findall(".//tag[@class='b']")) - ['tag', 'tag'] - >>> summarize_list(elem.findall(".//tag[@id]")) - ['tag'] - >>> summarize_list(elem.findall(".//section[tag]")) - ['section'] - >>> summarize_list(elem.findall(".//section[element]")) - [] - >>> summarize_list(elem.findall("../tag")) - [] - >>> summarize_list(elem.findall("section/../tag")) - ['tag', 'tag'] - >>> summarize_list(ET.ElementTree(elem).findall("./tag")) - ['tag', 'tag'] - - Following example is invalid in 1.2. - A leading '*' is assumed in 1.3. - - >>> elem.findall("section//") == elem.findall("section//*") - True - - ET's Path module handles this case incorrectly; this gives - a warning in 1.3, and the behaviour will be modified in 1.4. - - >>> summarize_list(ET.ElementTree(elem).findall("/tag")) - ['tag', 'tag'] - - >>> elem = ET.XML(SAMPLE_XML_NS) - >>> summarize_list(elem.findall("tag")) - [] - >>> summarize_list(elem.findall("{http://effbot.org/ns}tag")) - ['{http://effbot.org/ns}tag', '{http://effbot.org/ns}tag'] - >>> summarize_list(elem.findall(".//{http://effbot.org/ns}tag")) - ['{http://effbot.org/ns}tag', '{http://effbot.org/ns}tag', '{http://effbot.org/ns}tag'] - """ - -def file_init(): - """ - >>> import io - - >>> stringfile = io.BytesIO(SAMPLE_XML.encode("utf-8")) - >>> tree = ET.ElementTree(file=stringfile) - >>> tree.find("tag").tag - 'tag' - >>> tree.find("section/tag").tag - 'tag' - - >>> tree = ET.ElementTree(file=SIMPLE_XMLFILE) - >>> tree.find("element").tag - 'element' - >>> tree.find("element/../empty-element").tag - 'empty-element' - """ - -def bad_find(): - """ - Check bad or unsupported path expressions. - - >>> elem = ET.XML(SAMPLE_XML) - >>> elem.findall("/tag") - Traceback (most recent call last): - SyntaxError: cannot use absolute path on element - """ - -def path_cache(): - """ - Check that the path cache behaves sanely. - - >>> elem = ET.XML(SAMPLE_XML) - >>> for i in range(10): ET.ElementTree(elem).find('./'+str(i)) - >>> cache_len_10 = len(ET.ElementPath._cache) - >>> for i in range(10): ET.ElementTree(elem).find('./'+str(i)) - >>> len(ET.ElementPath._cache) == cache_len_10 - True - >>> for i in range(20): ET.ElementTree(elem).find('./'+str(i)) - >>> len(ET.ElementPath._cache) > cache_len_10 - True - >>> for i in range(600): ET.ElementTree(elem).find('./'+str(i)) - >>> len(ET.ElementPath._cache) < 500 - True - """ - -def copy(): - """ - Test copy handling (etc). - - >>> import copy - >>> e1 = ET.XML("<tag>hello<foo/></tag>") - >>> e2 = copy.copy(e1) - >>> e3 = copy.deepcopy(e1) - >>> e1.find("foo").tag = "bar" - >>> serialize(e1) - '<tag>hello<bar /></tag>' - >>> serialize(e2) - '<tag>hello<bar /></tag>' - >>> serialize(e3) - '<tag>hello<foo /></tag>' - - """ - -def attrib(): - """ - Test attribute handling. - - >>> elem = ET.Element("tag") - >>> elem.get("key") # 1.1 - >>> elem.get("key", "default") # 1.2 - 'default' - >>> elem.set("key", "value") - >>> elem.get("key") # 1.3 - 'value' - - >>> elem = ET.Element("tag", key="value") - >>> elem.get("key") # 2.1 - 'value' - >>> elem.attrib # 2.2 - {'key': 'value'} - - >>> attrib = {"key": "value"} - >>> elem = ET.Element("tag", attrib) - >>> attrib.clear() # check for aliasing issues - >>> elem.get("key") # 3.1 - 'value' - >>> elem.attrib # 3.2 - {'key': 'value'} - - >>> attrib = {"key": "value"} - >>> elem = ET.Element("tag", **attrib) - >>> attrib.clear() # check for aliasing issues - >>> elem.get("key") # 4.1 - 'value' - >>> elem.attrib # 4.2 - {'key': 'value'} - - >>> elem = ET.Element("tag", {"key": "other"}, key="value") - >>> elem.get("key") # 5.1 - 'value' - >>> elem.attrib # 5.2 - {'key': 'value'} - - >>> elem = ET.Element('test') - >>> elem.text = "aa" - >>> elem.set('testa', 'testval') - >>> elem.set('testb', 'test2') - >>> ET.tostring(elem) - b'<test testa="testval" testb="test2">aa</test>' - >>> sorted(elem.keys()) - ['testa', 'testb'] - >>> sorted(elem.items()) - [('testa', 'testval'), ('testb', 'test2')] - >>> elem.attrib['testb'] - 'test2' - >>> elem.attrib['testb'] = 'test1' - >>> elem.attrib['testc'] = 'test2' - >>> ET.tostring(elem) - b'<test testa="testval" testb="test1" testc="test2">aa</test>' - """ - -def makeelement(): - """ - Test makeelement handling. - - >>> elem = ET.Element("tag") - >>> attrib = {"key": "value"} - >>> subelem = elem.makeelement("subtag", attrib) - >>> if subelem.attrib is attrib: - ... print("attrib aliasing") - >>> elem.append(subelem) - >>> serialize(elem) - '<tag><subtag key="value" /></tag>' - - >>> elem.clear() - >>> serialize(elem) - '<tag />' - >>> elem.append(subelem) - >>> serialize(elem) - '<tag><subtag key="value" /></tag>' - >>> elem.extend([subelem, subelem]) - >>> serialize(elem) - '<tag><subtag key="value" /><subtag key="value" /><subtag key="value" /></tag>' - >>> elem[:] = [subelem] - >>> serialize(elem) - '<tag><subtag key="value" /></tag>' - >>> elem[:] = tuple([subelem]) - >>> serialize(elem) - '<tag><subtag key="value" /></tag>' - - """ - -def parsefile(): - """ - Test parsing from file. - - >>> tree = ET.parse(SIMPLE_XMLFILE) - >>> normalize_crlf(tree) - >>> tree.write(sys.stdout, encoding='unicode') - <root> - <element key="value">text</element> - <element>text</element>tail - <empty-element /> - </root> - >>> tree = ET.parse(SIMPLE_NS_XMLFILE) - >>> normalize_crlf(tree) - >>> tree.write(sys.stdout, encoding='unicode') - <ns0:root xmlns:ns0="namespace"> - <ns0:element key="value">text</ns0:element> - <ns0:element>text</ns0:element>tail - <ns0:empty-element /> - </ns0:root> - - >>> with open(SIMPLE_XMLFILE) as f: - ... data = f.read() - - >>> parser = ET.XMLParser() - >>> parser.version # doctest: +ELLIPSIS - 'Expat ...' - >>> parser.feed(data) - >>> print(serialize(parser.close())) - <root> - <element key="value">text</element> - <element>text</element>tail - <empty-element /> - </root> - - >>> parser = ET.XMLTreeBuilder() # 1.2 compatibility - >>> parser.feed(data) - >>> print(serialize(parser.close())) - <root> - <element key="value">text</element> - <element>text</element>tail - <empty-element /> - </root> - - >>> target = ET.TreeBuilder() - >>> parser = ET.XMLParser(target=target) - >>> parser.feed(data) - >>> print(serialize(parser.close())) - <root> - <element key="value">text</element> - <element>text</element>tail - <empty-element /> - </root> - """ - -def parseliteral(): - """ - >>> element = ET.XML("<html><body>text</body></html>") - >>> ET.ElementTree(element).write(sys.stdout, encoding='unicode') - <html><body>text</body></html> - >>> element = ET.fromstring("<html><body>text</body></html>") - >>> ET.ElementTree(element).write(sys.stdout, encoding='unicode') - <html><body>text</body></html> - >>> sequence = ["<html><body>", "text</bo", "dy></html>"] - >>> element = ET.fromstringlist(sequence) - >>> ET.tostring(element) - b'<html><body>text</body></html>' - >>> b"".join(ET.tostringlist(element)) - b'<html><body>text</body></html>' - >>> ET.tostring(element, "ascii") - b"<?xml version='1.0' encoding='ascii'?>\\n<html><body>text</body></html>" - >>> _, ids = ET.XMLID("<html><body>text</body></html>") - >>> len(ids) - 0 - >>> _, ids = ET.XMLID("<html><body id='body'>text</body></html>") - >>> len(ids) - 1 - >>> ids["body"].tag - 'body' - """ - -def iterparse(): - """ - Test iterparse interface. - - >>> iterparse = ET.iterparse - - >>> context = iterparse(SIMPLE_XMLFILE) - >>> action, elem = next(context) - >>> print(action, elem.tag) - end element - >>> for action, elem in context: - ... print(action, elem.tag) - end element - end empty-element - end root - >>> context.root.tag - 'root' - - >>> context = iterparse(SIMPLE_NS_XMLFILE) - >>> for action, elem in context: - ... print(action, elem.tag) - end {namespace}element - end {namespace}element - end {namespace}empty-element - end {namespace}root - - >>> events = () - >>> context = iterparse(SIMPLE_XMLFILE, events) - >>> for action, elem in context: - ... print(action, elem.tag) - - >>> events = () - >>> context = iterparse(SIMPLE_XMLFILE, events=events) - >>> for action, elem in context: - ... print(action, elem.tag) - - >>> events = ("start", "end") - >>> context = iterparse(SIMPLE_XMLFILE, events) - >>> for action, elem in context: - ... print(action, elem.tag) - start root - start element - end element - start element - end element - start empty-element - end empty-element - end root - - >>> events = ("start", "end", "start-ns", "end-ns") - >>> context = iterparse(SIMPLE_NS_XMLFILE, events) - >>> for action, elem in context: - ... if action in ("start", "end"): - ... print(action, elem.tag) - ... else: - ... print(action, elem) - start-ns ('', 'namespace') - start {namespace}root - start {namespace}element - end {namespace}element - start {namespace}element - end {namespace}element - start {namespace}empty-element - end {namespace}empty-element - end {namespace}root - end-ns None - - >>> events = ("start", "end", "bogus") - >>> with open(SIMPLE_XMLFILE, "rb") as f: - ... iterparse(f, events) - Traceback (most recent call last): - ValueError: unknown event 'bogus' - - >>> import io - - >>> source = io.BytesIO( - ... b"<?xml version='1.0' encoding='iso-8859-1'?>\\n" - ... b"<body xmlns='http://éffbot.org/ns'\\n" - ... b" xmlns:cl\\xe9='http://effbot.org/ns'>text</body>\\n") - >>> events = ("start-ns",) - >>> context = iterparse(source, events) - >>> for action, elem in context: - ... print(action, elem) - start-ns ('', 'http://\\xe9ffbot.org/ns') - start-ns ('cl\\xe9', 'http://effbot.org/ns') - - >>> source = io.StringIO("<document />junk") - >>> try: - ... for action, elem in iterparse(source): - ... print(action, elem.tag) - ... except ET.ParseError as v: - ... print(v) - end document - junk after document element: line 1, column 12 - """ - -def writefile(): - """ - >>> elem = ET.Element("tag") - >>> elem.text = "text" - >>> serialize(elem) - '<tag>text</tag>' - >>> ET.SubElement(elem, "subtag").text = "subtext" - >>> serialize(elem) - '<tag>text<subtag>subtext</subtag></tag>' - - Test tag suppression - >>> elem.tag = None - >>> serialize(elem) - 'text<subtag>subtext</subtag>' - >>> elem.insert(0, ET.Comment("comment")) - >>> serialize(elem) # assumes 1.3 - 'text<!--comment--><subtag>subtext</subtag>' - >>> elem[0] = ET.PI("key", "value") - >>> serialize(elem) - 'text<?key value?><subtag>subtext</subtag>' - """ - -def custom_builder(): - """ - Test parser w. custom builder. - - >>> with open(SIMPLE_XMLFILE) as f: - ... data = f.read() - >>> class Builder: - ... def start(self, tag, attrib): - ... print("start", tag) - ... def end(self, tag): - ... print("end", tag) - ... def data(self, text): - ... pass - >>> builder = Builder() - >>> parser = ET.XMLParser(target=builder) - >>> parser.feed(data) - start root - start element - end element - start element - end element - start empty-element - end empty-element - end root - - >>> with open(SIMPLE_NS_XMLFILE) as f: - ... data = f.read() - >>> class Builder: - ... def start(self, tag, attrib): - ... print("start", tag) - ... def end(self, tag): - ... print("end", tag) - ... def data(self, text): - ... pass - ... def pi(self, target, data): - ... print("pi", target, repr(data)) - ... def comment(self, data): - ... print("comment", repr(data)) - >>> builder = Builder() - >>> parser = ET.XMLParser(target=builder) - >>> parser.feed(data) - pi pi 'data' - comment ' comment ' - start {namespace}root - start {namespace}element - end {namespace}element - start {namespace}element - end {namespace}element - start {namespace}empty-element - end {namespace}empty-element - end {namespace}root - - """ - -def getchildren(): - """ - Test Element.getchildren() - - >>> with open(SIMPLE_XMLFILE, "rb") as f: - ... tree = ET.parse(f) - >>> for elem in tree.getroot().iter(): - ... summarize_list(elem.getchildren()) - ['element', 'element', 'empty-element'] - [] - [] - [] - >>> for elem in tree.getiterator(): - ... summarize_list(elem.getchildren()) - ['element', 'element', 'empty-element'] - [] - [] - [] - - >>> elem = ET.XML(SAMPLE_XML) - >>> len(elem.getchildren()) - 3 - >>> len(elem[2].getchildren()) - 1 - >>> elem[:] == elem.getchildren() - True - >>> child1 = elem[0] - >>> child2 = elem[2] - >>> del elem[1:2] - >>> len(elem.getchildren()) - 2 - >>> child1 == elem[0] - True - >>> child2 == elem[1] - True - >>> elem[0:2] = [child2, child1] - >>> child2 == elem[0] - True - >>> child1 == elem[1] - True - >>> child1 == elem[0] - False - >>> elem.clear() - >>> elem.getchildren() - [] - """ - -def writestring(): - """ - >>> elem = ET.XML("<html><body>text</body></html>") - >>> ET.tostring(elem) - b'<html><body>text</body></html>' - >>> elem = ET.fromstring("<html><body>text</body></html>") - >>> ET.tostring(elem) - b'<html><body>text</body></html>' - """ - -def check_encoding(encoding): - """ - >>> check_encoding("ascii") - >>> check_encoding("us-ascii") - >>> check_encoding("iso-8859-1") - >>> check_encoding("iso-8859-15") - >>> check_encoding("cp437") - >>> check_encoding("mac-roman") - """ - 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. - - >>> e = ET.XML("<html><link/><script>1 < 2</script></html>") - >>> e.tail = "\n" - >>> serialize(e) - '<html><link /><script>1 < 2</script></html>\n' - >>> serialize(e, method=None) - '<html><link /><script>1 < 2</script></html>\n' - >>> serialize(e, method="xml") - '<html><link /><script>1 < 2</script></html>\n' - >>> serialize(e, method="html") - '<html><link><script>1 < 2</script></html>\n' - >>> serialize(e, method="text") - '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' - """ +class ElementTreeTest(unittest.TestCase): + + def serialize_check(self, elem, expected): + self.assertEqual(serialize(elem), expected) + + def test_interface(self): + # Test element tree interface. + + def check_string(string): + len(string) + for char in string: + self.assertEqual(len(char), 1, + msg="expected one-character string, got %r" % char) + new_string = string + "" + new_string = string + " " + string[:0] + + def check_mapping(mapping): + len(mapping) + keys = mapping.keys() + items = mapping.items() + for key in keys: + item = mapping[key] + mapping["key"] = "value" + self.assertEqual(mapping["key"], "value", + msg="expected value string, got %r" % mapping["key"]) + + def check_element(element): + self.assertTrue(ET.iselement(element), msg="not an element") + self.assertTrue(hasattr(element, "tag"), msg="no tag member") + self.assertTrue(hasattr(element, "attrib"), msg="no attrib member") + self.assertTrue(hasattr(element, "text"), msg="no text member") + self.assertTrue(hasattr(element, "tail"), msg="no tail member") + + check_string(element.tag) + check_mapping(element.attrib) + if element.text is not None: + check_string(element.text) + if element.tail is not None: + check_string(element.tail) + for elem in element: + check_element(elem) + + element = ET.Element("tag") + check_element(element) + tree = ET.ElementTree(element) + check_element(tree.getroot()) + element = ET.Element("t\xe4g", key="value") + tree = ET.ElementTree(element) + self.assertRegex(repr(element), r"^<Element 't\xe4g' at 0x.*>$") + element = ET.Element("tag", key="value") + + # Make sure all standard element methods exist. + + def check_method(method): + self.assertTrue(hasattr(method, '__call__'), + msg="%s not callable" % method) + + check_method(element.append) + check_method(element.extend) + check_method(element.insert) + check_method(element.remove) + check_method(element.getchildren) + check_method(element.find) + check_method(element.iterfind) + check_method(element.findall) + check_method(element.findtext) + check_method(element.clear) + check_method(element.get) + check_method(element.set) + check_method(element.keys) + check_method(element.items) + check_method(element.iter) + check_method(element.itertext) + check_method(element.getiterator) + + # These methods return an iterable. See bug 6472. + + def check_iter(it): + check_method(it.__next__) + + check_iter(element.iterfind("tag")) + check_iter(element.iterfind("*")) + check_iter(tree.iterfind("tag")) + check_iter(tree.iterfind("*")) + + # These aliases are provided: + + self.assertEqual(ET.XML, ET.fromstring) + self.assertEqual(ET.PI, ET.ProcessingInstruction) + self.assertEqual(ET.XMLParser, ET.XMLTreeBuilder) + + def test_simpleops(self): + # Basic method sanity checks. + + elem = ET.XML("<body><tag/></body>") + self.serialize_check(elem, '<body><tag /></body>') + e = ET.Element("tag2") + elem.append(e) + self.serialize_check(elem, '<body><tag /><tag2 /></body>') + elem.remove(e) + self.serialize_check(elem, '<body><tag /></body>') + elem.insert(0, e) + self.serialize_check(elem, '<body><tag2 /><tag /></body>') + elem.remove(e) + elem.extend([e]) + self.serialize_check(elem, '<body><tag /><tag2 /></body>') + elem.remove(e) + + element = ET.Element("tag", key="value") + self.serialize_check(element, '<tag key="value" />') # 1 + subelement = ET.Element("subtag") + element.append(subelement) + self.serialize_check(element, '<tag key="value"><subtag /></tag>') # 2 + element.insert(0, subelement) + self.serialize_check(element, + '<tag key="value"><subtag /><subtag /></tag>') # 3 + element.remove(subelement) + self.serialize_check(element, '<tag key="value"><subtag /></tag>') # 4 + element.remove(subelement) + self.serialize_check(element, '<tag key="value" />') # 5 + with self.assertRaises(ValueError) as cm: + element.remove(subelement) + self.assertEqual(str(cm.exception), 'list.remove(x): x not in list') + self.serialize_check(element, '<tag key="value" />') # 6 + element[0:0] = [subelement, subelement, subelement] + self.serialize_check(element[1], '<subtag />') + self.assertEqual(element[1:9], [element[1], element[2]]) + self.assertEqual(element[:9:2], [element[0], element[2]]) + del element[1:2] + self.serialize_check(element, + '<tag key="value"><subtag /><subtag /></tag>') + + def test_cdata(self): + # Test CDATA handling (etc). + + self.serialize_check(ET.XML("<tag>hello</tag>"), + '<tag>hello</tag>') + self.serialize_check(ET.XML("<tag>hello</tag>"), + '<tag>hello</tag>') + self.serialize_check(ET.XML("<tag><![CDATA[hello]]></tag>"), + '<tag>hello</tag>') + + def test_file_init(self): + stringfile = io.BytesIO(SAMPLE_XML.encode("utf-8")) + tree = ET.ElementTree(file=stringfile) + self.assertEqual(tree.find("tag").tag, 'tag') + self.assertEqual(tree.find("section/tag").tag, 'tag') + + tree = ET.ElementTree(file=SIMPLE_XMLFILE) + self.assertEqual(tree.find("element").tag, 'element') + self.assertEqual(tree.find("element/../empty-element").tag, + 'empty-element') + + def test_path_cache(self): + # 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(ElementPath._cache) + for i in range(10): ET.ElementTree(elem).find('./'+str(i)) + self.assertEqual(len(ElementPath._cache), cache_len_10) + for i in range(20): ET.ElementTree(elem).find('./'+str(i)) + self.assertGreater(len(ElementPath._cache), cache_len_10) + for i in range(600): ET.ElementTree(elem).find('./'+str(i)) + self.assertLess(len(ElementPath._cache), 500) + + def test_copy(self): + # Test copy handling (etc). + + import copy + e1 = ET.XML("<tag>hello<foo/></tag>") + e2 = copy.copy(e1) + e3 = copy.deepcopy(e1) + e1.find("foo").tag = "bar" + self.serialize_check(e1, '<tag>hello<bar /></tag>') + self.serialize_check(e2, '<tag>hello<bar /></tag>') + self.serialize_check(e3, '<tag>hello<foo /></tag>') + + def test_attrib(self): + # Test attribute handling. + + elem = ET.Element("tag") + elem.get("key") # 1.1 + self.assertEqual(elem.get("key", "default"), 'default') # 1.2 + + elem.set("key", "value") + self.assertEqual(elem.get("key"), 'value') # 1.3 + + elem = ET.Element("tag", key="value") + self.assertEqual(elem.get("key"), 'value') # 2.1 + self.assertEqual(elem.attrib, {'key': 'value'}) # 2.2 + + attrib = {"key": "value"} + elem = ET.Element("tag", attrib) + attrib.clear() # check for aliasing issues + self.assertEqual(elem.get("key"), 'value') # 3.1 + self.assertEqual(elem.attrib, {'key': 'value'}) # 3.2 + + attrib = {"key": "value"} + elem = ET.Element("tag", **attrib) + attrib.clear() # check for aliasing issues + self.assertEqual(elem.get("key"), 'value') # 4.1 + self.assertEqual(elem.attrib, {'key': 'value'}) # 4.2 + + elem = ET.Element("tag", {"key": "other"}, key="value") + self.assertEqual(elem.get("key"), 'value') # 5.1 + self.assertEqual(elem.attrib, {'key': 'value'}) # 5.2 + + elem = ET.Element('test') + elem.text = "aa" + elem.set('testa', 'testval') + elem.set('testb', 'test2') + self.assertEqual(ET.tostring(elem), + b'<test testa="testval" testb="test2">aa</test>') + self.assertEqual(sorted(elem.keys()), ['testa', 'testb']) + self.assertEqual(sorted(elem.items()), + [('testa', 'testval'), ('testb', 'test2')]) + self.assertEqual(elem.attrib['testb'], 'test2') + elem.attrib['testb'] = 'test1' + elem.attrib['testc'] = 'test2' + self.assertEqual(ET.tostring(elem), + b'<test testa="testval" testb="test1" testc="test2">aa</test>') + + def test_makeelement(self): + # Test makeelement handling. + + elem = ET.Element("tag") + attrib = {"key": "value"} + subelem = elem.makeelement("subtag", attrib) + self.assertIsNot(subelem.attrib, attrib, msg="attrib aliasing") + elem.append(subelem) + self.serialize_check(elem, '<tag><subtag key="value" /></tag>') + + elem.clear() + self.serialize_check(elem, '<tag />') + elem.append(subelem) + self.serialize_check(elem, '<tag><subtag key="value" /></tag>') + elem.extend([subelem, subelem]) + self.serialize_check(elem, + '<tag><subtag key="value" /><subtag key="value" /><subtag key="value" /></tag>') + elem[:] = [subelem] + self.serialize_check(elem, '<tag><subtag key="value" /></tag>') + elem[:] = tuple([subelem]) + self.serialize_check(elem, '<tag><subtag key="value" /></tag>') + + def test_parsefile(self): + # Test parsing from file. + + tree = ET.parse(SIMPLE_XMLFILE) + stream = io.StringIO() + tree.write(stream, encoding='unicode') + self.assertEqual(stream.getvalue(), + '<root>\n' + ' <element key="value">text</element>\n' + ' <element>text</element>tail\n' + ' <empty-element />\n' + '</root>') + tree = ET.parse(SIMPLE_NS_XMLFILE) + stream = io.StringIO() + tree.write(stream, encoding='unicode') + self.assertEqual(stream.getvalue(), + '<ns0:root xmlns:ns0="namespace">\n' + ' <ns0:element key="value">text</ns0:element>\n' + ' <ns0:element>text</ns0:element>tail\n' + ' <ns0:empty-element />\n' + '</ns0:root>') + + with open(SIMPLE_XMLFILE) as f: + data = f.read() + + parser = ET.XMLParser() + self.assertRegex(parser.version, r'^Expat ') + parser.feed(data) + self.serialize_check(parser.close(), + '<root>\n' + ' <element key="value">text</element>\n' + ' <element>text</element>tail\n' + ' <empty-element />\n' + '</root>') + + parser = ET.XMLTreeBuilder() # 1.2 compatibility + parser.feed(data) + self.serialize_check(parser.close(), + '<root>\n' + ' <element key="value">text</element>\n' + ' <element>text</element>tail\n' + ' <empty-element />\n' + '</root>') + + target = ET.TreeBuilder() + parser = ET.XMLParser(target=target) + parser.feed(data) + self.serialize_check(parser.close(), + '<root>\n' + ' <element key="value">text</element>\n' + ' <element>text</element>tail\n' + ' <empty-element />\n' + '</root>') + + def test_parseliteral(self): + element = ET.XML("<html><body>text</body></html>") + self.assertEqual(ET.tostring(element, encoding='unicode'), + '<html><body>text</body></html>') + element = ET.fromstring("<html><body>text</body></html>") + self.assertEqual(ET.tostring(element, encoding='unicode'), + '<html><body>text</body></html>') + sequence = ["<html><body>", "text</bo", "dy></html>"] + element = ET.fromstringlist(sequence) + self.assertEqual(ET.tostring(element), + b'<html><body>text</body></html>') + self.assertEqual(b"".join(ET.tostringlist(element)), + b'<html><body>text</body></html>') + self.assertEqual(ET.tostring(element, "ascii"), + b"<?xml version='1.0' encoding='ascii'?>\n" + b"<html><body>text</body></html>") + _, ids = ET.XMLID("<html><body>text</body></html>") + self.assertEqual(len(ids), 0) + _, ids = ET.XMLID("<html><body id='body'>text</body></html>") + self.assertEqual(len(ids), 1) + self.assertEqual(ids["body"].tag, 'body') + + def test_iterparse(self): + # Test iterparse interface. + + iterparse = ET.iterparse + + context = iterparse(SIMPLE_XMLFILE) + action, elem = next(context) + self.assertEqual((action, elem.tag), ('end', 'element')) + self.assertEqual([(action, elem.tag) for action, elem in context], [ + ('end', 'element'), + ('end', 'empty-element'), + ('end', 'root'), + ]) + self.assertEqual(context.root.tag, 'root') + + context = iterparse(SIMPLE_NS_XMLFILE) + self.assertEqual([(action, elem.tag) for action, elem in context], [ + ('end', '{namespace}element'), + ('end', '{namespace}element'), + ('end', '{namespace}empty-element'), + ('end', '{namespace}root'), + ]) + + events = () + context = iterparse(SIMPLE_XMLFILE, events) + self.assertEqual([(action, elem.tag) for action, elem in context], []) + + events = () + context = iterparse(SIMPLE_XMLFILE, events=events) + self.assertEqual([(action, elem.tag) for action, elem in context], []) + + events = ("start", "end") + context = iterparse(SIMPLE_XMLFILE, events) + self.assertEqual([(action, elem.tag) for action, elem in context], [ + ('start', 'root'), + ('start', 'element'), + ('end', 'element'), + ('start', 'element'), + ('end', 'element'), + ('start', 'empty-element'), + ('end', 'empty-element'), + ('end', 'root'), + ]) + + events = ("start", "end", "start-ns", "end-ns") + context = iterparse(SIMPLE_NS_XMLFILE, events) + self.assertEqual([(action, elem.tag) if action in ("start", "end") + else (action, elem) + for action, elem in context], [ + ('start-ns', ('', 'namespace')), + ('start', '{namespace}root'), + ('start', '{namespace}element'), + ('end', '{namespace}element'), + ('start', '{namespace}element'), + ('end', '{namespace}element'), + ('start', '{namespace}empty-element'), + ('end', '{namespace}empty-element'), + ('end', '{namespace}root'), + ('end-ns', None), + ]) + + events = ("start", "end", "bogus") + with self.assertRaises(ValueError) as cm: + with open(SIMPLE_XMLFILE, "rb") as f: + iterparse(f, events) + self.assertEqual(str(cm.exception), "unknown event 'bogus'") + + source = io.BytesIO( + b"<?xml version='1.0' encoding='iso-8859-1'?>\n" + b"<body xmlns='http://éffbot.org/ns'\n" + b" xmlns:cl\xe9='http://effbot.org/ns'>text</body>\n") + events = ("start-ns",) + context = iterparse(source, events) + self.assertEqual([(action, elem) for action, elem in context], [ + ('start-ns', ('', 'http://\xe9ffbot.org/ns')), + ('start-ns', ('cl\xe9', 'http://effbot.org/ns')), + ]) + + source = io.StringIO("<document />junk") + it = iterparse(source) + action, elem = next(it) + self.assertEqual((action, elem.tag), ('end', 'document')) + with self.assertRaises(ET.ParseError) as cm: + next(it) + self.assertEqual(str(cm.exception), + 'junk after document element: line 1, column 12') + + def test_writefile(self): + elem = ET.Element("tag") + elem.text = "text" + self.serialize_check(elem, '<tag>text</tag>') + ET.SubElement(elem, "subtag").text = "subtext" + self.serialize_check(elem, '<tag>text<subtag>subtext</subtag></tag>') + + # Test tag suppression + elem.tag = None + self.serialize_check(elem, 'text<subtag>subtext</subtag>') + elem.insert(0, ET.Comment("comment")) + self.serialize_check(elem, + 'text<!--comment--><subtag>subtext</subtag>') # assumes 1.3 + + elem[0] = ET.PI("key", "value") + self.serialize_check(elem, 'text<?key value?><subtag>subtext</subtag>') + + def test_custom_builder(self): + # Test parser w. custom builder. + + with open(SIMPLE_XMLFILE) as f: + data = f.read() + class Builder(list): + def start(self, tag, attrib): + self.append(("start", tag)) + def end(self, tag): + self.append(("end", tag)) + def data(self, text): + pass + builder = Builder() + parser = ET.XMLParser(target=builder) + parser.feed(data) + self.assertEqual(builder, [ + ('start', 'root'), + ('start', 'element'), + ('end', 'element'), + ('start', 'element'), + ('end', 'element'), + ('start', 'empty-element'), + ('end', 'empty-element'), + ('end', 'root'), + ]) + + with open(SIMPLE_NS_XMLFILE) as f: + data = f.read() + class Builder(list): + def start(self, tag, attrib): + self.append(("start", tag)) + def end(self, tag): + self.append(("end", tag)) + def data(self, text): + pass + def pi(self, target, data): + self.append(("pi", target, data)) + def comment(self, data): + self.append(("comment", data)) + builder = Builder() + parser = ET.XMLParser(target=builder) + parser.feed(data) + self.assertEqual(builder, [ + ('pi', 'pi', 'data'), + ('comment', ' comment '), + ('start', '{namespace}root'), + ('start', '{namespace}element'), + ('end', '{namespace}element'), + ('start', '{namespace}element'), + ('end', '{namespace}element'), + ('start', '{namespace}empty-element'), + ('end', '{namespace}empty-element'), + ('end', '{namespace}root'), + ]) + + + def test_getchildren(self): + # Test Element.getchildren() + + with open(SIMPLE_XMLFILE, "rb") as f: + tree = ET.parse(f) + self.assertEqual([summarize_list(elem.getchildren()) + for elem in tree.getroot().iter()], [ + ['element', 'element', 'empty-element'], + [], + [], + [], + ]) + self.assertEqual([summarize_list(elem.getchildren()) + for elem in tree.getiterator()], [ + ['element', 'element', 'empty-element'], + [], + [], + [], + ]) + + elem = ET.XML(SAMPLE_XML) + self.assertEqual(len(elem.getchildren()), 3) + self.assertEqual(len(elem[2].getchildren()), 1) + self.assertEqual(elem[:], elem.getchildren()) + child1 = elem[0] + child2 = elem[2] + del elem[1:2] + self.assertEqual(len(elem.getchildren()), 2) + self.assertEqual(child1, elem[0]) + self.assertEqual(child2, elem[1]) + elem[0:2] = [child2, child1] + self.assertEqual(child2, elem[0]) + self.assertEqual(child1, elem[1]) + self.assertNotEqual(child1, elem[0]) + elem.clear() + self.assertEqual(elem.getchildren(), []) + + def test_writestring(self): + elem = ET.XML("<html><body>text</body></html>") + self.assertEqual(ET.tostring(elem), b'<html><body>text</body></html>') + elem = ET.fromstring("<html><body>text</body></html>") + self.assertEqual(ET.tostring(elem), b'<html><body>text</body></html>') + + def test_encoding(encoding): + def check(encoding): + ET.XML("<?xml version='1.0' encoding='%s'?><xml />" % encoding) + check("ascii") + check("us-ascii") + check("iso-8859-1") + check("iso-8859-15") + check("cp437") + check("mac-roman") + + def test_methods(self): + # Test serialization methods. + + e = ET.XML("<html><link/><script>1 < 2</script></html>") + e.tail = "\n" + self.assertEqual(serialize(e), + '<html><link /><script>1 < 2</script></html>\n') + self.assertEqual(serialize(e, method=None), + '<html><link /><script>1 < 2</script></html>\n') + self.assertEqual(serialize(e, method="xml"), + '<html><link /><script>1 < 2</script></html>\n') + self.assertEqual(serialize(e, method="html"), + '<html><link><script>1 < 2</script></html>\n') + self.assertEqual(serialize(e, method="text"), '1 < 2\n') + + def test_entity(self): + # Test entity handling. + + # 1) good entities + + e = ET.XML("<document title='舰'>test</document>") + self.assertEqual(serialize(e, encoding="us-ascii"), + b'<document title="舰">test</document>') + self.serialize_check(e, '<document title="\u8230">test</document>') + + # 2) bad entities + + with self.assertRaises(ET.ParseError) as cm: + ET.XML("<document>&entity;</document>") + self.assertEqual(str(cm.exception), + 'undefined entity: line 1, column 10') + + with self.assertRaises(ET.ParseError) as cm: + ET.XML(ENTITY_XML) + self.assertEqual(str(cm.exception), + 'undefined entity &entity;: line 5, column 10') + + # 3) custom entity + + parser = ET.XMLParser() + parser.entity["entity"] = "text" + parser.feed(ENTITY_XML) + root = parser.close() + self.serialize_check(root, '<document>text</document>') + + def test_namespace(self): + # Test namespace issues. + + # 1) xml namespace + + elem = ET.XML("<tag xml:lang='en' />") + self.serialize_check(elem, '<tag xml:lang="en" />') # 1.1 + + # 2) other "well-known" namespaces + + elem = ET.XML("<rdf:RDF xmlns:rdf='http://www.w3.org/1999/02/22-rdf-syntax-ns#' />") + self.serialize_check(elem, + '<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" />') # 2.1 + + elem = ET.XML("<html:html xmlns:html='http://www.w3.org/1999/xhtml' />") + self.serialize_check(elem, + '<html:html xmlns:html="http://www.w3.org/1999/xhtml" />') # 2.2 + + elem = ET.XML("<soap:Envelope xmlns:soap='http://schemas.xmlsoap.org/soap/envelope' />") + self.serialize_check(elem, + '<ns0:Envelope xmlns:ns0="http://schemas.xmlsoap.org/soap/envelope" />') # 2.3 + + # 3) unknown namespaces + elem = ET.XML(SAMPLE_XML_NS) + self.serialize_check(elem, + '<ns0:body xmlns:ns0="http://effbot.org/ns">\n' + ' <ns0:tag>text</ns0:tag>\n' + ' <ns0:tag />\n' + ' <ns0:section>\n' + ' <ns0:tag>subtext</ns0:tag>\n' + ' </ns0:section>\n' + '</ns0:body>') + + def test_qname(self): + # Test QName handling. + + # 1) decorated tags + + elem = ET.Element("{uri}tag") + self.serialize_check(elem, '<ns0:tag xmlns:ns0="uri" />') # 1.1 + elem = ET.Element(ET.QName("{uri}tag")) + self.serialize_check(elem, '<ns0:tag xmlns:ns0="uri" />') # 1.2 + elem = ET.Element(ET.QName("uri", "tag")) + self.serialize_check(elem, '<ns0:tag xmlns:ns0="uri" />') # 1.3 + elem = ET.Element(ET.QName("uri", "tag")) + subelem = ET.SubElement(elem, ET.QName("uri", "tag1")) + subelem = ET.SubElement(elem, ET.QName("uri", "tag2")) + self.serialize_check(elem, + '<ns0:tag xmlns:ns0="uri"><ns0:tag1 /><ns0:tag2 /></ns0:tag>') # 1.4 + + # 2) decorated attributes + + elem.clear() + elem.attrib["{uri}key"] = "value" + self.serialize_check(elem, + '<ns0:tag xmlns:ns0="uri" ns0:key="value" />') # 2.1 + + elem.clear() + elem.attrib[ET.QName("{uri}key")] = "value" + self.serialize_check(elem, + '<ns0:tag xmlns:ns0="uri" ns0:key="value" />') # 2.2 + + # 3) decorated values are not converted by default, but the + # QName wrapper can be used for values + + elem.clear() + elem.attrib["{uri}key"] = "{uri}value" + self.serialize_check(elem, + '<ns0:tag xmlns:ns0="uri" ns0:key="{uri}value" />') # 3.1 + + elem.clear() + elem.attrib["{uri}key"] = ET.QName("{uri}value") + self.serialize_check(elem, + '<ns0:tag xmlns:ns0="uri" ns0:key="ns0:value" />') # 3.2 + + elem.clear() + subelem = ET.Element("tag") + subelem.attrib["{uri1}key"] = ET.QName("{uri2}value") + elem.append(subelem) + elem.append(subelem) + self.serialize_check(elem, + '<ns0:tag xmlns:ns0="uri" xmlns:ns1="uri1" xmlns:ns2="uri2">' + '<tag ns1:key="ns2:value" />' + '<tag ns1:key="ns2:value" />' + '</ns0:tag>') # 3.3 + + # 4) Direct QName tests + + self.assertEqual(str(ET.QName('ns', 'tag')), '{ns}tag') + self.assertEqual(str(ET.QName('{ns}tag')), '{ns}tag') + q1 = ET.QName('ns', 'tag') + q2 = ET.QName('ns', 'tag') + self.assertEqual(q1, q2) + q2 = ET.QName('ns', 'other-tag') + self.assertNotEqual(q1, q2) + self.assertNotEqual(q1, 'ns:tag') + self.assertEqual(q1, '{ns}tag') + + def test_doctype_public(self): + # Test PUBLIC doctype. + + elem = ET.XML('<!DOCTYPE html PUBLIC' + ' "-//W3C//DTD XHTML 1.0 Transitional//EN"' + ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">' + '<html>text</html>') + + def test_xpath_tokenizer(self): + # Test the XPath tokenizer. + from xml.etree import ElementPath + def check(p, expected): + self.assertEqual([op or tag + for op, tag in ElementPath.xpath_tokenizer(p)], + expected) + + # tests from the xml specification + check("*", ['*']) + check("text()", ['text', '()']) + check("@name", ['@', 'name']) + check("@*", ['@', '*']) + check("para[1]", ['para', '[', '1', ']']) + check("para[last()]", ['para', '[', 'last', '()', ']']) + check("*/para", ['*', '/', 'para']) + check("/doc/chapter[5]/section[2]", + ['/', 'doc', '/', 'chapter', '[', '5', ']', + '/', 'section', '[', '2', ']']) + check("chapter//para", ['chapter', '//', 'para']) + check("//para", ['//', 'para']) + check("//olist/item", ['//', 'olist', '/', 'item']) + check(".", ['.']) + check(".//para", ['.', '//', 'para']) + check("..", ['..']) + check("../@lang", ['..', '/', '@', 'lang']) + check("chapter[title]", ['chapter', '[', 'title', ']']) + check("employee[@secretary and @assistant]", ['employee', + '[', '@', 'secretary', '', 'and', '', '@', 'assistant', ']']) + + # additional tests + check("{http://spam}egg", ['{http://spam}egg']) + check("./spam.egg", ['.', '/', 'spam.egg']) + check(".//{http://spam}egg", ['.', '//', '{http://spam}egg']) + + def test_processinginstruction(self): + # Test ProcessingInstruction directly + + self.assertEqual(ET.tostring(ET.ProcessingInstruction('test', 'instruction')), + b'<?test instruction?>') + self.assertEqual(ET.tostring(ET.PI('test', 'instruction')), + b'<?test instruction?>') + + # Issue #2746 + + self.assertEqual(ET.tostring(ET.PI('test', '<testing&>')), + b'<?test <testing&>?>') + self.assertEqual(ET.tostring(ET.PI('test', '<testing&>\xe3'), 'latin-1'), + b"<?xml version='1.0' encoding='latin-1'?>\n" + b"<?test <testing&>\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></%s>' % (elem,elem)), + method='html') + self.assertEqual(serialized, expected) -ENTITY_XML = """\ -<!DOCTYPE points [ -<!ENTITY % user-entities SYSTEM 'user-entities.xml'> -%user-entities; -]> -<document>&entity;</document> -""" - -def entity(): - """ - Test entity handling. - - 1) good entities - - >>> e = ET.XML("<document title='舰'>test</document>") - >>> serialize(e, encoding="us-ascii") - b'<document title="舰">test</document>' - >>> serialize(e) - '<document title="\u8230">test</document>' - - 2) bad entities - - >>> normalize_exception(ET.XML, "<document>&entity;</document>") - Traceback (most recent call last): - ParseError: undefined entity: line 1, column 10 - - >>> normalize_exception(ET.XML, ENTITY_XML) - Traceback (most recent call last): - ParseError: undefined entity &entity;: line 5, column 10 - - 3) custom entity - - >>> parser = ET.XMLParser() - >>> parser.entity["entity"] = "text" - >>> parser.feed(ENTITY_XML) - >>> root = parser.close() - >>> serialize(root) - '<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. - - 1) xml namespace - - >>> elem = ET.XML("<tag xml:lang='en' />") - >>> serialize(elem) # 1.1 - '<tag xml:lang="en" />' - - 2) other "well-known" namespaces - - >>> elem = ET.XML("<rdf:RDF xmlns:rdf='http://www.w3.org/1999/02/22-rdf-syntax-ns#' />") - >>> serialize(elem) # 2.1 - '<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" />' - - >>> elem = ET.XML("<html:html xmlns:html='http://www.w3.org/1999/xhtml' />") - >>> serialize(elem) # 2.2 - '<html:html xmlns:html="http://www.w3.org/1999/xhtml" />' - - >>> elem = ET.XML("<soap:Envelope xmlns:soap='http://schemas.xmlsoap.org/soap/envelope' />") - >>> serialize(elem) # 2.3 - '<ns0:Envelope xmlns:ns0="http://schemas.xmlsoap.org/soap/envelope" />' - - 3) unknown namespaces - >>> elem = ET.XML(SAMPLE_XML_NS) - >>> print(serialize(elem)) - <ns0:body xmlns:ns0="http://effbot.org/ns"> - <ns0:tag>text</ns0:tag> - <ns0:tag /> - <ns0:section> - <ns0:tag>subtext</ns0:tag> - </ns0:section> - </ns0:body> - """ - -def qname(): - """ - Test QName handling. - - 1) decorated tags - - >>> elem = ET.Element("{uri}tag") - >>> serialize(elem) # 1.1 - '<ns0:tag xmlns:ns0="uri" />' - >>> elem = ET.Element(ET.QName("{uri}tag")) - >>> serialize(elem) # 1.2 - '<ns0:tag xmlns:ns0="uri" />' - >>> elem = ET.Element(ET.QName("uri", "tag")) - >>> serialize(elem) # 1.3 - '<ns0:tag xmlns:ns0="uri" />' - >>> elem = ET.Element(ET.QName("uri", "tag")) - >>> subelem = ET.SubElement(elem, ET.QName("uri", "tag1")) - >>> subelem = ET.SubElement(elem, ET.QName("uri", "tag2")) - >>> serialize(elem) # 1.4 - '<ns0:tag xmlns:ns0="uri"><ns0:tag1 /><ns0:tag2 /></ns0:tag>' - - 2) decorated attributes - - >>> elem.clear() - >>> elem.attrib["{uri}key"] = "value" - >>> serialize(elem) # 2.1 - '<ns0:tag xmlns:ns0="uri" ns0:key="value" />' - - >>> elem.clear() - >>> elem.attrib[ET.QName("{uri}key")] = "value" - >>> serialize(elem) # 2.2 - '<ns0:tag xmlns:ns0="uri" ns0:key="value" />' - - 3) decorated values are not converted by default, but the - QName wrapper can be used for values - - >>> elem.clear() - >>> elem.attrib["{uri}key"] = "{uri}value" - >>> serialize(elem) # 3.1 - '<ns0:tag xmlns:ns0="uri" ns0:key="{uri}value" />' - - >>> elem.clear() - >>> elem.attrib["{uri}key"] = ET.QName("{uri}value") - >>> serialize(elem) # 3.2 - '<ns0:tag xmlns:ns0="uri" ns0:key="ns0:value" />' - - >>> elem.clear() - >>> subelem = ET.Element("tag") - >>> subelem.attrib["{uri1}key"] = ET.QName("{uri2}value") - >>> elem.append(subelem) - >>> elem.append(subelem) - >>> serialize(elem) # 3.3 - '<ns0:tag xmlns:ns0="uri" xmlns:ns1="uri1" xmlns:ns2="uri2"><tag ns1:key="ns2:value" /><tag ns1:key="ns2:value" /></ns0:tag>' - - 4) Direct QName tests - - >>> str(ET.QName('ns', 'tag')) - '{ns}tag' - >>> str(ET.QName('{ns}tag')) - '{ns}tag' - >>> q1 = ET.QName('ns', 'tag') - >>> q2 = ET.QName('ns', 'tag') - >>> q1 == q2 - True - >>> q2 = ET.QName('ns', 'other-tag') - >>> q1 == q2 - False - >>> q1 == 'ns:tag' - False - >>> q1 == '{ns}tag' - True - """ - -def doctype_public(): - """ - Test PUBLIC doctype. - - >>> elem = ET.XML('<!DOCTYPE html PUBLIC' - ... ' "-//W3C//DTD XHTML 1.0 Transitional//EN"' - ... ' "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">' - ... '<html>text</html>') - - """ - -def xpath_tokenizer(p): - """ - Test the XPath tokenizer. - - >>> # tests from the xml specification - >>> xpath_tokenizer("*") - ['*'] - >>> xpath_tokenizer("text()") - ['text', '()'] - >>> xpath_tokenizer("@name") - ['@', 'name'] - >>> xpath_tokenizer("@*") - ['@', '*'] - >>> xpath_tokenizer("para[1]") - ['para', '[', '1', ']'] - >>> xpath_tokenizer("para[last()]") - ['para', '[', 'last', '()', ']'] - >>> xpath_tokenizer("*/para") - ['*', '/', 'para'] - >>> xpath_tokenizer("/doc/chapter[5]/section[2]") - ['/', 'doc', '/', 'chapter', '[', '5', ']', '/', 'section', '[', '2', ']'] - >>> xpath_tokenizer("chapter//para") - ['chapter', '//', 'para'] - >>> xpath_tokenizer("//para") - ['//', 'para'] - >>> xpath_tokenizer("//olist/item") - ['//', 'olist', '/', 'item'] - >>> xpath_tokenizer(".") - ['.'] - >>> xpath_tokenizer(".//para") - ['.', '//', 'para'] - >>> xpath_tokenizer("..") - ['..'] - >>> xpath_tokenizer("../@lang") - ['..', '/', '@', 'lang'] - >>> xpath_tokenizer("chapter[title]") - ['chapter', '[', 'title', ']'] - >>> xpath_tokenizer("employee[@secretary and @assistant]") - ['employee', '[', '@', 'secretary', '', 'and', '', '@', 'assistant', ']'] - - >>> # additional tests - >>> xpath_tokenizer("{http://spam}egg") - ['{http://spam}egg'] - >>> xpath_tokenizer("./spam.egg") - ['.', '/', 'spam.egg'] - >>> xpath_tokenizer(".//{http://spam}egg") - ['.', '//', '{http://spam}egg'] - """ - from xml.etree import ElementPath - out = [] - for op, tag in ElementPath.xpath_tokenizer(p): - out.append(op or tag) - return out - -def processinginstruction(): - """ - Test ProcessingInstruction directly - - >>> ET.tostring(ET.ProcessingInstruction('test', 'instruction')) - b'<?test instruction?>' - >>> ET.tostring(ET.PI('test', 'instruction')) - b'<?test instruction?>' - - Issue #2746 - - >>> 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?>" - """ # # xinclude tests (samples from appendix C of the xinclude specification) @@ -1338,96 +979,6 @@ 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") - if parse == "xml": - from xml.etree.ElementTree import XML - return XML(data) - return data - -def xinclude(): - r""" - Basic inclusion example (XInclude C.1) - - >>> from xml.etree import ElementTree as ET - >>> from xml.etree import ElementInclude - - >>> document = xinclude_loader("C1.xml") - >>> ElementInclude.include(document, xinclude_loader) - >>> print(serialize(document)) # C1 - <document> - <p>120 Mz is adequate for an average home user.</p> - <disclaimer> - <p>The opinions represented herein represent those of the individual - and should not be interpreted as official policy endorsed by this - organization.</p> - </disclaimer> - </document> - - Textual inclusion example (XInclude C.2) - - >>> document = xinclude_loader("C2.xml") - >>> ElementInclude.include(document, xinclude_loader) - >>> print(serialize(document)) # C2 - <document> - <p>This document has been accessed - 324387 times.</p> - </document> - - Textual inclusion after sibling element (based on modified XInclude C.2) - - >>> document = xinclude_loader("C2b.xml") - >>> ElementInclude.include(document, xinclude_loader) - >>> print(serialize(document)) # C2b - <document> - <p>This document has been <em>accessed</em> - 324387 times.</p> - </document> - - Textual inclusion of XML example (XInclude C.3) - - >>> document = xinclude_loader("C3.xml") - >>> ElementInclude.include(document, xinclude_loader) - >>> print(serialize(document)) # C3 - <document> - <p>The following is the source of the "data.xml" resource:</p> - <example><?xml version='1.0'?> - <data> - <item><![CDATA[Brooks & Shields]]></item> - </data> - </example> - </document> - - Fallback example (XInclude C.5) - Note! Fallback support is not yet implemented - - >>> document = xinclude_loader("C5.xml") - >>> ElementInclude.include(document, xinclude_loader) - Traceback (most recent call last): - IOError: 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 @@ -1448,430 +999,1230 @@ XINCLUDE_BAD["B2.xml"] = """\ </div> """ -def xinclude_failures(): - r""" - Test failure to locate included XML file. - - >>> from xml.etree import ElementInclude - - >>> def none_loader(href, parser, encoding=None): - ... return None - - >>> document = ET.XML(XINCLUDE["C1.xml"]) - >>> ElementInclude.include(document, loader=none_loader) - Traceback (most recent call last): - xml.etree.ElementInclude.FatalIncludeError: cannot load 'disclaimer.xml' as 'xml' - - Test failure to locate included text file. - - >>> document = ET.XML(XINCLUDE["C2.xml"]) - >>> ElementInclude.include(document, loader=none_loader) - Traceback (most recent call last): - xml.etree.ElementInclude.FatalIncludeError: cannot load 'count.txt' as 'text' - - Test bad parse type. - - >>> document = ET.XML(XINCLUDE_BAD["B1.xml"]) - >>> ElementInclude.include(document, loader=none_loader) - Traceback (most recent call last): - xml.etree.ElementInclude.FatalIncludeError: unknown parse type in xi:include tag ('BAD_TYPE') - - Test xi:fallback outside xi:include. - - >>> document = ET.XML(XINCLUDE_BAD["B2.xml"]) - >>> ElementInclude.include(document, loader=none_loader) - Traceback (most recent call last): - xml.etree.ElementInclude.FatalIncludeError: xi:fallback tag must be child of xi:include ('{http://www.w3.org/2001/XInclude}fallback') - """ +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), + '<document>\n' + ' <p>Example.</p>\n' + ' <root>\n' + ' <element key="value">text</element>\n' + ' <element>text</element>tail\n' + ' <empty-element />\n' + '</root>\n' + '</document>') + + 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), + '<document>\n' + ' <p>120 Mz is adequate for an average home user.</p>\n' + ' <disclaimer>\n' + ' <p>The opinions represented herein represent those of the individual\n' + ' and should not be interpreted as official policy endorsed by this\n' + ' organization.</p>\n' + '</disclaimer>\n' + '</document>') # C1 + + # Textual inclusion example (XInclude C.2) + document = self.xinclude_loader("C2.xml") + ElementInclude.include(document, self.xinclude_loader) + self.assertEqual(serialize(document), + '<document>\n' + ' <p>This document has been accessed\n' + ' 324387 times.</p>\n' + '</document>') # 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), + '<document>\n' + ' <p>This document has been <em>accessed</em>\n' + ' 324387 times.</p>\n' + '</document>') # 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), + '<document>\n' + ' <p>The following is the source of the "data.xml" resource:</p>\n' + " <example><?xml version='1.0'?>\n" + '<data>\n' + ' <item><![CDATA[Brooks & Shields]]></item>\n' + '</data>\n' + '</example>\n' + '</document>') # 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), + '<div xmlns:ns0="http://www.w3.org/2001/XInclude">\n' + ' <ns0:include href="example.txt" parse="text">\n' + ' <ns0:fallback>\n' + ' <ns0:include href="fallback-example.txt" parse="text">\n' + ' <ns0:fallback><a href="mailto:bob@example.org">Report error</a></ns0:fallback>\n' + ' </ns0:include>\n' + ' </ns0:fallback>\n' + ' </ns0:include>\n' + '</div>') # C5 + + 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')") # -------------------------------------------------------------------- # reported bugs -def bug_xmltoolkit21(): - """ - - marshaller gives obscure errors for non-string values - - >>> elem = ET.Element(123) - >>> serialize(elem) # tag - Traceback (most recent call last): - TypeError: cannot serialize 123 (type int) - >>> elem = ET.Element("elem") - >>> elem.text = 123 - >>> serialize(elem) # text - Traceback (most recent call last): - TypeError: cannot serialize 123 (type int) - >>> elem = ET.Element("elem") - >>> elem.tail = 123 - >>> serialize(elem) # tail - Traceback (most recent call last): - TypeError: cannot serialize 123 (type int) - >>> elem = ET.Element("elem") - >>> elem.set(123, "123") - >>> serialize(elem) # attribute key - Traceback (most recent call last): - TypeError: cannot serialize 123 (type int) - >>> elem = ET.Element("elem") - >>> elem.set("123", 123) - >>> serialize(elem) # attribute value - Traceback (most recent call last): - TypeError: cannot serialize 123 (type int) - - """ - -def bug_xmltoolkit25(): - """ - - typo in ElementTree.findtext +class BugsTest(unittest.TestCase): - >>> elem = ET.XML(SAMPLE_XML) - >>> tree = ET.ElementTree(elem) - >>> tree.findtext("tag") - 'text' - >>> tree.findtext("section/tag") - 'subtext' + 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)') -def bug_xmltoolkit28(): - """ + elem = ET.Element(123) + check(elem) # tag - .//tag causes exceptions + elem = ET.Element("elem") + elem.text = 123 + check(elem) # text - >>> tree = ET.XML("<doc><table><tbody/></table></doc>") - >>> summarize_list(tree.findall(".//thead")) - [] - >>> summarize_list(tree.findall(".//tbody")) - ['tbody'] + elem = ET.Element("elem") + elem.tail = 123 + check(elem) # tail - """ + elem = ET.Element("elem") + elem.set(123, "123") + check(elem) # attribute key -def bug_xmltoolkitX1(): - """ + elem = ET.Element("elem") + elem.set("123", 123) + check(elem) # attribute value - dump() doesn't flush the output buffer + def test_bug_xmltoolkit25(self): + # typo in ElementTree.findtext - >>> tree = ET.XML("<doc><table><tbody/></table></doc>") - >>> ET.dump(tree); print("tail") - <doc><table><tbody /></table></doc> - tail + 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 -def bug_xmltoolkit39(): - """ + tree = ET.XML("<doc><table><tbody/></table></doc>") + self.assertEqual(summarize_list(tree.findall(".//thead")), []) + self.assertEqual(summarize_list(tree.findall(".//tbody")), ['tbody']) - non-ascii element and attribute names doesn't work + def test_bug_xmltoolkitX1(self): + # dump() doesn't flush the output buffer - >>> tree = ET.XML(b"<?xml version='1.0' encoding='iso-8859-1'?><t\\xe4g />") - >>> ET.tostring(tree, "utf-8") - b'<t\\xc3\\xa4g />' + tree = ET.XML("<doc><table><tbody/></table></doc>") + with support.captured_stdout() as stdout: + ET.dump(tree) + self.assertEqual(stdout.getvalue(), '<doc><table><tbody /></table></doc>\n') - >>> tree = ET.XML(b"<?xml version='1.0' encoding='iso-8859-1'?><tag \\xe4ttr='välue' />") - >>> tree.attrib - {'\\xe4ttr': 'v\\xe4lue'} - >>> ET.tostring(tree, "utf-8") - b'<tag \\xc3\\xa4ttr="v\\xc3\\xa4lue" />' + def test_bug_xmltoolkit39(self): + # non-ascii element and attribute names doesn't work - >>> tree = ET.XML(b"<?xml version='1.0' encoding='iso-8859-1'?><t\\xe4g>text</t\\xe4g>") - >>> ET.tostring(tree, "utf-8") - b'<t\\xc3\\xa4g>text</t\\xc3\\xa4g>' + tree = ET.XML(b"<?xml version='1.0' encoding='iso-8859-1'?><t\xe4g />") + self.assertEqual(ET.tostring(tree, "utf-8"), b'<t\xc3\xa4g />') - >>> tree = ET.Element("t\u00e4g") - >>> ET.tostring(tree, "utf-8") - b'<t\\xc3\\xa4g />' + tree = ET.XML(b"<?xml version='1.0' encoding='iso-8859-1'?>" + b"<tag \xe4ttr='välue' />") + self.assertEqual(tree.attrib, {'\xe4ttr': 'v\xe4lue'}) + self.assertEqual(ET.tostring(tree, "utf-8"), + b'<tag \xc3\xa4ttr="v\xc3\xa4lue" />') - >>> tree = ET.Element("tag") - >>> tree.set("\u00e4ttr", "v\u00e4lue") - >>> ET.tostring(tree, "utf-8") - b'<tag \\xc3\\xa4ttr="v\\xc3\\xa4lue" />' + tree = ET.XML(b"<?xml version='1.0' encoding='iso-8859-1'?>" + b'<t\xe4g>text</t\xe4g>') + self.assertEqual(ET.tostring(tree, "utf-8"), + b'<t\xc3\xa4g>text</t\xc3\xa4g>') - """ + tree = ET.Element("t\u00e4g") + self.assertEqual(ET.tostring(tree, "utf-8"), b'<t\xc3\xa4g />') -def bug_xmltoolkit54(): - """ + tree = ET.Element("tag") + tree.set("\u00e4ttr", "v\u00e4lue") + self.assertEqual(ET.tostring(tree, "utf-8"), + b'<tag \xc3\xa4ttr="v\xc3\xa4lue" />') - problems handling internally defined entities + def test_bug_xmltoolkit54(self): + # problems handling internally defined entities - >>> e = ET.XML("<!DOCTYPE doc [<!ENTITY ldots '舰'>]><doc>&ldots;</doc>") - >>> serialize(e, encoding="us-ascii") - b'<doc>舰</doc>' - >>> serialize(e) - '<doc>\u8230</doc>' + e = ET.XML("<!DOCTYPE doc [<!ENTITY ldots '舰'>]>" + '<doc>&ldots;</doc>') + self.assertEqual(serialize(e, encoding="us-ascii"), + b'<doc>舰</doc>') + self.assertEqual(serialize(e), '<doc>\u8230</doc>') - """ + def test_bug_xmltoolkit55(self): + # make sure we're reporting the first error, not the last -def bug_xmltoolkit55(): - """ + with self.assertRaises(ET.ParseError) as cm: + ET.XML(b"<!DOCTYPE doc SYSTEM 'doc.dtd'>" + b'<doc>&ldots;&ndots;&rdots;</doc>') + self.assertEqual(str(cm.exception), + 'undefined entity &ldots;: line 1, column 36') - make sure we're reporting the first error, not the last + def test_bug_xmltoolkit60(self): + # Handle crash in stream source. - >>> normalize_exception(ET.XML, b"<!DOCTYPE doc SYSTEM 'doc.dtd'><doc>&ldots;&ndots;&rdots;</doc>") - Traceback (most recent call last): - ParseError: undefined entity &ldots;: line 1, column 36 + class ExceptionFile: + def read(self, x): + raise OSError - """ + self.assertRaises(OSError, ET.parse, ExceptionFile()) -class ExceptionFile: - def read(self, x): - raise IOError + def test_bug_xmltoolkit62(self): + # Don't crash when using custom entities. -def xmltoolkit60(): - """ - - Handle crash in stream source. - >>> tree = ET.parse(ExceptionFile()) - Traceback (most recent call last): - IOError - - """ - -XMLTOOLKIT62_DOC = """<?xml version="1.0" encoding="UTF-8"?> + ENTITIES = {'rsquo': '\u2019', 'lsquo': '\u2018'} + parser = ET.XMLTreeBuilder() + parser.entity.update(ENTITIES) + parser.feed("""<?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE patent-application-publication SYSTEM "pap-v15-2001-01-31.dtd" []> <patent-application-publication> <subdoc-abstract> <paragraph id="A-0001" lvl="0">A new cultivar of Begonia plant named ‘BCT9801BEG’.</paragraph> </subdoc-abstract> -</patent-application-publication>""" - - -def xmltoolkit62(): - """ - - Don't crash when using custom entities. - - >>> xmltoolkit62() - 'A new cultivar of Begonia plant named \u2018BCT9801BEG\u2019.' - - """ - ENTITIES = {'rsquo': '\u2019', 'lsquo': '\u2018'} - parser = ET.XMLTreeBuilder() - parser.entity.update(ENTITIES) - parser.feed(XMLTOOLKIT62_DOC) - t = parser.close() - return t.find('.//paragraph').text - -def xmltoolkit63(): - """ - - Check reference leak. - >>> xmltoolkit63() - >>> count = sys.getrefcount(None) - >>> for i in range(1000): - ... xmltoolkit63() - >>> sys.getrefcount(None) - count - 0 - - """ - tree = ET.TreeBuilder() - tree.start("tag", {}) - tree.data("text") - tree.end("tag") +</patent-application-publication>""") + t = parser.close() + self.assertEqual(t.find('.//paragraph').text, + 'A new cultivar of Begonia plant named \u2018BCT9801BEG\u2019.') + + 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'<SomeTag text="def _f(): return 3 " />') + 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'<SomeTag text="def _f(): return 3 " />') + + def test_bug_200708_close(self): + # Test default builder. + parser = ET.XMLParser() # default + parser.feed("<element>some text</element>") + self.assertEqual(parser.close().tag, 'element') + + # Test custom builder. + class EchoTarget: + def close(self): + return ET.Element("element") # simulate root + parser = ET.XMLParser(EchoTarget()) + parser.feed("<element>some text</element>") + 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 + '<elem xmlns="default"><elem /></elem>') + + 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 + '<elem xmlns="default" xmlns:ns1="not-default">' + '<elem />' + '<ns1:elem />' + '</elem>') + + 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'<ns0:title xmlns:ns0="http://namespace.invalid/does/not/exist/" />') + 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'<foo:title xmlns:foo="http://namespace.invalid/does/not/exist/" />') + + # 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'<dc:title xmlns:dc="http://purl.org/dc/elements/1.1/" />') + + 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), '<tag />') + + def test_issue6233(self): + e = ET.XML(b"<?xml version='1.0' encoding='utf-8'?>" + b'<body>t\xc3\xa3g</body>') + self.assertEqual(ET.tostring(e, 'ascii'), + b"<?xml version='1.0' encoding='ascii'?>\n" + b'<body>tãg</body>') + e = ET.XML(b"<?xml version='1.0' encoding='iso-8859-1'?>" + b'<body>t\xe3g</body>') + self.assertEqual(ET.tostring(e, 'ascii'), + b"<?xml version='1.0' encoding='ascii'?>\n" + b'<body>tãg</body>') + + def test_issue3151(self): + e = ET.XML('<prefix:localname xmlns:prefix="${stuff}"/>') + self.assertEqual(e.tag, '{${stuff}}localname') + t = ET.ElementTree(e) + self.assertEqual(ET.tostring(e), b'<ns0:localname xmlns:ns0="${stuff}" />') + + def test_issue6565(self): + elem = ET.XML("<body><tag/></body>") + 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 bug_200708_newline(): - r""" - - Preserve newlines in attributes. - - >>> e = ET.Element('SomeTag', text="def _f():\n return 3\n") - >>> ET.tostring(e) - b'<SomeTag text="def _f(): return 3 " />' - >>> ET.XML(ET.tostring(e)).get("text") - 'def _f():\n return 3\n' - >>> ET.tostring(ET.XML(ET.tostring(e))) - b'<SomeTag text="def _f(): return 3 " />' - - """ - -def bug_200708_close(): - """ - - Test default builder. - >>> parser = ET.XMLParser() # default - >>> parser.feed("<element>some text</element>") - >>> summarize(parser.close()) - 'element' - - Test custom builder. - >>> class EchoTarget: - ... def close(self): - ... return ET.Element("element") # simulate root - >>> parser = ET.XMLParser(EchoTarget()) - >>> parser.feed("<element>some text</element>") - >>> summarize(parser.close()) - 'element' - - """ +class BasicElementTest(ElementTestCase, unittest.TestCase): + def test_augmentation_type_errors(self): + e = ET.Element('joe') + self.assertRaises(TypeError, e.append, 'b') + self.assertRaises(TypeError, e.extend, [ET.Element('bar'), 'foo']) + self.assertRaises(TypeError, e.insert, 0, 'foo') + + def test_cyclic_gc(self): + class Dummy: + pass + + # Test the shortest cycle: d->element->d + d = Dummy() + d.dummyref = ET.Element('joe', attr=d) + wref = weakref.ref(d) + del d + gc_collect() + self.assertIsNone(wref()) + + # A longer cycle: d->e->e2->d + e = ET.Element('joe') + d = Dummy() + d.dummyref = e + wref = weakref.ref(d) + e2 = ET.SubElement(e, 'foo', attr=d) + del d, e, e2 + gc_collect() + self.assertIsNone(wref()) + + # A cycle between Element objects as children of one another + # e1->e2->e3->e1 + e1 = ET.Element('e1') + e2 = ET.Element('e2') + e3 = ET.Element('e3') + e1.append(e2) + e2.append(e2) + e3.append(e1) + wref = weakref.ref(e1) + del e1, e2, e3 + gc_collect() + self.assertIsNone(wref()) + + def test_weakref(self): + flag = False + def wref_cb(w): + nonlocal flag + flag = True + e = ET.Element('e') + wref = weakref.ref(e, wref_cb) + self.assertEqual(wref().tag, 'e') + del e + self.assertEqual(flag, True) + self.assertEqual(wref(), None) + + def test_get_keyword_args(self): + e1 = ET.Element('foo' , x=1, y=2, z=3) + self.assertEqual(e1.get('x', default=7), 1) + self.assertEqual(e1.get('w', default=7), 7) + + def test_pickle(self): + # issue #16076: the C implementation wasn't pickleable. + for dumper, loader in product(self.modules, repeat=2): + e = dumper.Element('foo', bar=42) + e.text = "text goes here" + e.tail = "opposite of head" + dumper.SubElement(e, 'child').append(dumper.Element('grandchild')) + e.append(dumper.Element('child')) + e.findall('.//grandchild')[0].set('attr', 'other value') + + e2 = self.pickleRoundTrip(e, 'xml.etree.ElementTree', + dumper, loader) + + self.assertEqual(e2.tag, 'foo') + self.assertEqual(e2.attrib['bar'], 42) + self.assertEqual(len(e2), 2) + self.assertEqualElements(e, e2) + + +class 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') + + +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('<tag><empty /></tag>').findtext('empty'), '') + + def test_find_xpath(self): + LINEAR_XML = ''' + <body> + <tag class='a'/> + <tag class='b'/> + <tag class='c'/> + <tag class='d'/> + </body>''' + e = ET.XML(LINEAR_XML) + + # Test for numeric indexing and last() + self.assertEqual(e.find('./tag[1]').attrib['class'], 'a') + self.assertEqual(e.find('./tag[2]').attrib['class'], 'b') + self.assertEqual(e.find('./tag[last()]').attrib['class'], 'd') + self.assertEqual(e.find('./tag[last()-1]').attrib['class'], 'c') + self.assertEqual(e.find('./tag[last()-2]').attrib['class'], 'b') + + def test_findall(self): + e = ET.XML(SAMPLE_XML) + e[2] = ET.XML(SAMPLE_SECTION) + self.assertEqual(summarize_list(e.findall('.')), ['body']) + self.assertEqual(summarize_list(e.findall('tag')), ['tag', 'tag']) + self.assertEqual(summarize_list(e.findall('tog')), []) + self.assertEqual(summarize_list(e.findall('tog/foo')), []) + self.assertEqual(summarize_list(e.findall('*')), + ['tag', 'tag', 'section']) + self.assertEqual(summarize_list(e.findall('.//tag')), + ['tag'] * 4) + self.assertEqual(summarize_list(e.findall('section/tag')), ['tag']) + self.assertEqual(summarize_list(e.findall('section//tag')), ['tag'] * 2) + self.assertEqual(summarize_list(e.findall('section/*')), + ['tag', 'nexttag', 'nextsection']) + self.assertEqual(summarize_list(e.findall('section//*')), + ['tag', 'nexttag', 'nextsection', 'tag']) + self.assertEqual(summarize_list(e.findall('section/.//*')), + ['tag', 'nexttag', 'nextsection', 'tag']) + self.assertEqual(summarize_list(e.findall('*/*')), + ['tag', 'nexttag', 'nextsection']) + self.assertEqual(summarize_list(e.findall('*//*')), + ['tag', 'nexttag', 'nextsection', 'tag']) + self.assertEqual(summarize_list(e.findall('*/tag')), ['tag']) + self.assertEqual(summarize_list(e.findall('*/./tag')), ['tag']) + self.assertEqual(summarize_list(e.findall('./tag')), ['tag'] * 2) + self.assertEqual(summarize_list(e.findall('././tag')), ['tag'] * 2) + + self.assertEqual(summarize_list(e.findall('.//tag[@class]')), + ['tag'] * 3) + self.assertEqual(summarize_list(e.findall('.//tag[@class="a"]')), + ['tag']) + self.assertEqual(summarize_list(e.findall('.//tag[@class="b"]')), + ['tag'] * 2) + self.assertEqual(summarize_list(e.findall('.//tag[@id]')), + ['tag']) + self.assertEqual(summarize_list(e.findall('.//section[tag]')), + ['section']) + self.assertEqual(summarize_list(e.findall('.//section[element]')), []) + self.assertEqual(summarize_list(e.findall('../tag')), []) + self.assertEqual(summarize_list(e.findall('section/../tag')), + ['tag'] * 2) + self.assertEqual(e.findall('section//'), e.findall('section//*')) + + def test_test_find_with_ns(self): + e = ET.XML(SAMPLE_XML_NS) + self.assertEqual(summarize_list(e.findall('tag')), []) + self.assertEqual( + summarize_list(e.findall("{http://effbot.org/ns}tag")), + ['{http://effbot.org/ns}tag'] * 2) + self.assertEqual( + summarize_list(e.findall(".//{http://effbot.org/ns}tag")), + ['{http://effbot.org/ns}tag'] * 3) + + def test_bad_find(self): + e = ET.XML(SAMPLE_XML) + with self.assertRaisesRegex(SyntaxError, 'cannot use absolute path'): + e.findall('/tag') + + def test_find_through_ElementTree(self): + e = ET.XML(SAMPLE_XML) + self.assertEqual(ET.ElementTree(e).find('tag').tag, 'tag') + self.assertEqual(ET.ElementTree(e).findtext('tag'), 'text') + self.assertEqual(summarize_list(ET.ElementTree(e).findall('tag')), + ['tag'] * 2) + # this produces a warning + self.assertEqual(summarize_list(ET.ElementTree(e).findall('//tag')), + ['tag'] * 3) + + +class ElementIterTest(unittest.TestCase): + def _ilist(self, elem, tag=None): + return summarize_list(elem.iter(tag)) + + def test_basic(self): + doc = ET.XML("<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') + + # With an explitit 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("<root>a&<sub>b&</sub>c&</root>") + 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(''' + <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) + + # test that iter also accepts 'tag' as a keyword arg + self.assertEqual( + summarize_list(doc.iter(tag='room')), + ['room'] * 3) + + # make sure both tag=None and tag='*' return all tags + all_tags = ['document', 'house', 'room', 'room', + 'shed', 'house', 'room'] + self.assertEqual(self._ilist(doc), all_tags) + self.assertEqual(self._ilist(doc, '*'), all_tags) + + +class TreeBuilderTest(unittest.TestCase): + sample1 = ('<!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 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): + support.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") -def bug_200709_default_namespace(): - """ - - >>> e = ET.Element("{default}elem") - >>> s = ET.SubElement(e, "{default}elem") - >>> serialize(e, default_namespace="default") # 1 - '<elem xmlns="default"><elem /></elem>' - - >>> e = ET.Element("{default}elem") - >>> s = ET.SubElement(e, "{default}elem") - >>> s = ET.SubElement(e, "{not-default}elem") - >>> serialize(e, default_namespace="default") # 2 - '<elem xmlns="default" xmlns:ns1="not-default"><elem /><ns1:elem /></elem>' - - >>> e = ET.Element("{default}elem") - >>> s = ET.SubElement(e, "{default}elem") - >>> s = ET.SubElement(e, "elem") # unprefixed name - >>> serialize(e, default_namespace="default") # 3 - Traceback (most recent call last): - ValueError: cannot use non-qualified names with default_namespace option - - """ - -def bug_200709_register_namespace(): - """ - - >>> ET.tostring(ET.Element("{http://namespace.invalid/does/not/exist/}title")) - b'<ns0:title xmlns:ns0="http://namespace.invalid/does/not/exist/" />' - >>> ET.register_namespace("foo", "http://namespace.invalid/does/not/exist/") - >>> ET.tostring(ET.Element("{http://namespace.invalid/does/not/exist/}title")) - b'<foo:title xmlns:foo="http://namespace.invalid/does/not/exist/" />' - - And the Dublin Core namespace is in the default list: - - >>> ET.tostring(ET.Element("{http://purl.org/dc/elements/1.1/}title")) - b'<dc:title xmlns:dc="http://purl.org/dc/elements/1.1/" />' - - """ - -def bug_200709_element_comment(): - """ - - Not sure if this can be fixed, really (since the serializer needs - ET.Comment, not cET.comment). - - >>> a = ET.Element('a') - >>> a.append(ET.Comment('foo')) - >>> a[0].tag == ET.Comment - True - - >>> a = ET.Element('a') - >>> a.append(ET.PI('foo')) - >>> a[0].tag == ET.PI - True - - """ - -def bug_200709_element_insert(): - """ - - >>> a = ET.Element('a') - >>> b = ET.SubElement(a, 'b') - >>> c = ET.SubElement(a, 'c') - >>> d = ET.Element('d') - >>> a.insert(0, d) - >>> summarize_list(a) - ['d', 'b', 'c'] - >>> a.insert(-1, d) - >>> summarize_list(a) - ['d', 'b', 'd', 'c'] - - """ - -def bug_200709_iter_comment(): - """ - - >>> a = ET.Element('a') - >>> b = ET.SubElement(a, 'b') - >>> comment_b = ET.Comment("TEST-b") - >>> b.append(comment_b) - >>> summarize_list(a.iter(ET.Comment)) - ['<Comment>'] +# -------------------------------------------------------------------- - """ +class NoAcceleratorTest(unittest.TestCase): + def setUp(self): + if not pyET: + raise unittest.SkipTest('only for the Python version') -# -------------------------------------------------------------------- -# reported on bugs.python.org - -def bug_1534630(): - """ - - >>> bob = ET.TreeBuilder() - >>> e = bob.data("data") - >>> e = bob.start("tag", {}) - >>> e = bob.end("tag") - >>> e = bob.close() - >>> serialize(e) - '<tag />' - - """ - -def check_issue6233(): - """ - - >>> e = ET.XML(b"<?xml version='1.0' encoding='utf-8'?><body>t\\xc3\\xa3g</body>") - >>> ET.tostring(e, 'ascii') - b"<?xml version='1.0' encoding='ascii'?>\\n<body>tãg</body>" - >>> e = ET.XML(b"<?xml version='1.0' encoding='iso-8859-1'?><body>t\\xe3g</body>") - >>> ET.tostring(e, 'ascii') - b"<?xml version='1.0' encoding='ascii'?>\\n<body>tãg</body>" - - """ - -def check_issue3151(): - """ - - >>> e = ET.XML('<prefix:localname xmlns:prefix="${stuff}"/>') - >>> e.tag - '{${stuff}}localname' - >>> t = ET.ElementTree(e) - >>> ET.tostring(e) - b'<ns0:localname xmlns:ns0="${stuff}" />' - - """ - -def check_issue6565(): - """ - - >>> elem = ET.XML("<body><tag/></body>") - >>> summarize_list(elem) - ['tag'] - >>> newelem = ET.XML(SAMPLE_XML) - >>> elem[:] = newelem[:] - >>> summarize_list(elem) - ['tag', 'tag', 'section'] - - """ - -def check_issue10777(): - """ - Registering a namespace twice caused a "dictionary changed size during - iteration" bug. - - >>> ET.register_namespace('test10777', 'http://myuri/') - >>> ET.register_namespace('test10777', 'http://myuri/') - """ - -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>' - - """ + # 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') # -------------------------------------------------------------------- @@ -1894,44 +2245,71 @@ 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. + 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, + ElementTreeTest, + IOTest, + ParseErrorTest, + XIncludeTest, + ElementTreeTypeTest, + ElementFindTest, + ElementIterTest, + TreeBuilderTest, + BugsTest, + ] + + # 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, + ]) - 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 - - # 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) + try: + # XXX the C module should give the same warnings as the Python module + with CleanContext(quiet=(pyET is not ET)): + support.run_unittest(*test_classes) + 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() |