# 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. # # 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 from test import support from test.support import findfile from xml.etree import ElementTree as ET SIMPLE_XMLFILE = findfile("simple.xml", subdir="xmltestdata") SIMPLE_NS_XMLFILE = findfile("simple-ns.xml", subdir="xmltestdata") SAMPLE_XML = """\ text
subtext
""" SAMPLE_SECTION = """\
subtext
""" SAMPLE_XML_NS = """ text
subtext
""" def sanity(): """ 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, **options): import io if options.get("encoding"): file = io.BytesIO() else: file = io.StringIO() tree = ET.ElementTree(elem) tree.write(file, **options) if to_string: return file.getvalue() else: file.seek(0) return file def summarize(elem): if elem.tag == ET.Comment: return "" 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) # -------------------------------------------------------------------- # 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 = 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("") >>> serialize(elem) '' >>> e = ET.Element("tag2") >>> elem.append(e) >>> serialize(elem) '' >>> elem.remove(e) >>> serialize(elem) '' >>> elem.insert(0, e) >>> serialize(elem) '' >>> elem.remove(e) >>> elem.extend([e]) >>> serialize(elem) '' >>> elem.remove(e) >>> element = ET.Element("tag", key="value") >>> serialize(element) # 1 '' >>> subelement = ET.Element("subtag") >>> element.append(subelement) >>> serialize(element) # 2 '' >>> element.insert(0, subelement) >>> serialize(element) # 3 '' >>> element.remove(subelement) >>> serialize(element) # 4 '' >>> element.remove(subelement) >>> serialize(element) # 5 '' >>> element.remove(subelement) Traceback (most recent call last): ValueError: list.remove(x): x not in list >>> serialize(element) # 6 '' >>> element[0:0] = [subelement, subelement, subelement] >>> serialize(element[1]) '' >>> element[1:9] == [element[1], element[2]] True >>> element[:9:2] == [element[0], element[2]] True >>> del element[1:2] >>> serialize(element) '' """ def cdata(): """ Test CDATA handling (etc). >>> serialize(ET.XML("hello")) 'hello' >>> serialize(ET.XML("hello")) 'hello' >>> serialize(ET.XML("")) 'hello' """ # 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' >>> 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("hello") >>> e2 = copy.copy(e1) >>> e3 = copy.deepcopy(e1) >>> e1.find("foo").tag = "bar" >>> serialize(e1) 'hello' >>> serialize(e2) 'hello' >>> serialize(e3) 'hello' """ 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) 'aa' >>> 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) 'aa' """ 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) '' >>> elem.clear() >>> serialize(elem) '' >>> elem.append(subelem) >>> serialize(elem) '' >>> elem.extend([subelem, subelem]) >>> serialize(elem) '' >>> elem[:] = [subelem] >>> serialize(elem) '' >>> elem[:] = tuple([subelem]) >>> serialize(elem) '' """ def parsefile(): """ Test parsing from file. >>> tree = ET.parse(SIMPLE_XMLFILE) >>> normalize_crlf(tree) >>> tree.write(sys.stdout) text texttail >>> tree = ET.parse(SIMPLE_NS_XMLFILE) >>> normalize_crlf(tree) >>> tree.write(sys.stdout) text texttail >>> parser = ET.XMLParser() >>> parser.version # XXX: Upgrade to 2.0.1? 'Expat 2.0.0' >>> parser.feed(open(SIMPLE_XMLFILE).read()) >>> print(serialize(parser.close())) text texttail >>> parser = ET.XMLTreeBuilder() # 1.2 compatibility >>> parser.feed(open(SIMPLE_XMLFILE).read()) >>> print(serialize(parser.close())) text texttail >>> target = ET.TreeBuilder() >>> parser = ET.XMLParser(target=target) >>> parser.feed(open(SIMPLE_XMLFILE).read()) >>> print(serialize(parser.close())) text texttail """ def parseliteral(): """ >>> element = ET.XML("text") >>> ET.ElementTree(element).write(sys.stdout) text >>> element = ET.fromstring("text") >>> ET.ElementTree(element).write(sys.stdout) text >>> sequence = ["", "text"] >>> element = ET.fromstringlist(sequence) >>> print(ET.tostring(element)) text >>> print("".join(ET.tostringlist(element))) text >>> ET.tostring(element, "ascii") b"\\ntext" >>> _, ids = ET.XMLID("text") >>> len(ids) 0 >>> _, ids = ET.XMLID("text") >>> 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") >>> context = iterparse(SIMPLE_XMLFILE, events) Traceback (most recent call last): ValueError: unknown event 'bogus' >>> import io >>> source = io.BytesIO( ... b"\\n" ... b"text\\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("junk") >>> try: ... for action, elem in iterparse(source): ... print(action, elem.tag) ... except ET.ParseError as v: ... print(v) junk after document element: line 1, column 12 """ def writefile(): """ >>> elem = ET.Element("tag") >>> elem.text = "text" >>> serialize(elem) 'text' >>> ET.SubElement(elem, "subtag").text = "subtext" >>> serialize(elem) 'textsubtext' Test tag suppression >>> elem.tag = None >>> serialize(elem) 'textsubtext' >>> elem.insert(0, ET.Comment("comment")) >>> serialize(elem) # assumes 1.3 'textsubtext' >>> elem[0] = ET.PI("key", "value") >>> serialize(elem) 'textsubtext' """ def custom_builder(): """ Test parser w. custom builder. >>> 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(open(SIMPLE_XMLFILE, "r").read()) start root start element end element start element end element start empty-element end empty-element end root >>> 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(open(SIMPLE_NS_XMLFILE, "r").read()) 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() >>> tree = ET.parse(open(SIMPLE_XMLFILE, "rb")) >>> 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("text") >>> ET.tostring(elem) 'text' >>> elem = ET.fromstring("text") >>> ET.tostring(elem) 'text' """ 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("" % encoding) def encoding(): r""" Test encoding issues. >>> elem = ET.Element("tag") >>> elem.text = "abc" >>> serialize(elem) 'abc' >>> serialize(elem, encoding="utf-8") b'abc' >>> serialize(elem, encoding="us-ascii") b'abc' >>> serialize(elem, encoding="iso-8859-1") b"\nabc" >>> elem.text = "<&\"\'>" >>> serialize(elem) '<&"\'>' >>> serialize(elem, encoding="utf-8") b'<&"\'>' >>> serialize(elem, encoding="us-ascii") # cdata characters b'<&"\'>' >>> serialize(elem, encoding="iso-8859-1") b'\n<&"\'>' >>> elem.attrib["key"] = "<&\"\'>" >>> elem.text = None >>> serialize(elem) '' >>> serialize(elem, encoding="utf-8") b'' >>> serialize(elem, encoding="us-ascii") b'' >>> serialize(elem, encoding="iso-8859-1") b'\n' >>> elem.text = '\xe5\xf6\xf6<>' >>> elem.attrib.clear() >>> serialize(elem) '\xe5\xf6\xf6<>' >>> serialize(elem, encoding="utf-8") b'\xc3\xa5\xc3\xb6\xc3\xb6<>' >>> serialize(elem, encoding="us-ascii") b'åöö<>' >>> serialize(elem, encoding="iso-8859-1") b"\n\xe5\xf6\xf6<>" >>> elem.attrib["key"] = '\xe5\xf6\xf6<>' >>> elem.text = None >>> serialize(elem) '' >>> serialize(elem, encoding="utf-8") b'' >>> serialize(elem, encoding="us-ascii") b'' >>> serialize(elem, encoding="iso-8859-1") b'\n' """ def methods(): r""" Test serialization methods. >>> e = ET.XML("") >>> e.tail = "\n" >>> serialize(e) '\n' >>> serialize(e, method=None) '\n' >>> serialize(e, method="xml") '\n' >>> serialize(e, method="html") '\n' >>> serialize(e, method="text") '1 < 2\n' """ def iterators(): """ Test iterators. >>> e = ET.XML("this is a paragraph...") >>> 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', ) >>> tree = ET.ElementTree(None) >>> tree.iter() Traceback (most recent call last): AttributeError: 'NoneType' object has no attribute 'iter' """ ENTITY_XML = """\ %user-entities; ]> &entity; """ def entity(): """ Test entity handling. 1) good entities >>> e = ET.XML("test") >>> serialize(e, encoding="us-ascii") b'test' >>> serialize(e) 'test' 2) bad entities >>> normalize_exception(ET.XML, "&entity;") 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) 'text' """ def error(xml): """ Test error handling. >>> issubclass(ET.ParseError, SyntaxError) True >>> error("foo").position (1, 0) >>> error("&foo;").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("") >>> serialize(elem) # 1.1 '' 2) other "well-known" namespaces >>> elem = ET.XML("") >>> serialize(elem) # 2.1 '' >>> elem = ET.XML("") >>> serialize(elem) # 2.2 '' >>> elem = ET.XML("") >>> serialize(elem) # 2.3 '' 3) unknown namespaces >>> elem = ET.XML(SAMPLE_XML_NS) >>> print(serialize(elem)) text subtext """ def qname(): """ Test QName handling. 1) decorated tags >>> elem = ET.Element("{uri}tag") >>> serialize(elem) # 1.1 '' >>> elem = ET.Element(ET.QName("{uri}tag")) >>> serialize(elem) # 1.2 '' >>> elem = ET.Element(ET.QName("uri", "tag")) >>> serialize(elem) # 1.3 '' 2) decorated attributes >>> elem.clear() >>> elem.attrib["{uri}key"] = "value" >>> serialize(elem) # 2.1 '' >>> elem.clear() >>> elem.attrib[ET.QName("{uri}key")] = "value" >>> serialize(elem) # 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" >>> serialize(elem) # 3.1 '' >>> elem.clear() >>> elem.attrib["{uri}key"] = ET.QName("{uri}value") >>> serialize(elem) # 3.2 '' >>> elem.clear() >>> subelem = ET.Element("tag") >>> subelem.attrib["{uri1}key"] = ET.QName("{uri2}value") >>> elem.append(subelem) >>> elem.append(subelem) >>> serialize(elem) # 3.3 '' 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('' ... 'text') """ 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')) '' >>> ET.tostring(ET.PI('test', 'instruction')) '' Issue #2746 >>> ET.tostring(ET.PI('test', '')) '?>' >>> ET.tostring(ET.PI('test', '\xe3'), 'latin1') b"\\n\\xe3?>" """ # # xinclude tests (samples from appendix C of the xinclude specification) XINCLUDE = {} XINCLUDE["C1.xml"] = """\

120 Mz is adequate for an average home user.

""" XINCLUDE["disclaimer.xml"] = """\

The opinions represented herein represent those of the individual and should not be interpreted as official policy endorsed by this organization.

""" XINCLUDE["C2.xml"] = """\

This document has been accessed times.

""" XINCLUDE["count.txt"] = "324387" XINCLUDE["C3.xml"] = """\

The following is the source of the "data.xml" resource:

""" XINCLUDE["data.xml"] = """\ """ XINCLUDE["C5.xml"] = """\ """ XINCLUDE["default.xml"] = """\

Example.

""".format(SIMPLE_XMLFILE) 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

120 Mz is adequate for an average home user.

The opinions represented herein represent those of the individual and should not be interpreted as official policy endorsed by this organization.

Textual inclusion example (XInclude C.2) >>> document = xinclude_loader("C2.xml") >>> ElementInclude.include(document, xinclude_loader) >>> print(serialize(document)) # C2

This document has been accessed 324387 times.

Textual inclusion of XML example (XInclude C.3) >>> document = xinclude_loader("C3.xml") >>> ElementInclude.include(document, xinclude_loader) >>> print(serialize(document)) # C3

The following is the source of the "data.xml" resource:

<?xml version='1.0'?> <data> <item><![CDATA[Brooks & Shields]]></item> </data>
Fallback example (XInclude C.5) Note! Fallback support is not yet implemented >>> document = xinclude_loader("C5.xml") >>> ElementInclude.include(document, xinclude_loader) Traceback (most recent call last): 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

Example.

text texttail
""" # # badly formatted xi:include tags XINCLUDE_BAD = {} XINCLUDE_BAD["B1.xml"] = """\

120 Mz is adequate for an average home user.

""" XINCLUDE_BAD["B2.xml"] = """\
""" def xinclude_failures(): r""" Test failure to locate included XML file. >>> from xml.etree import ElementInclude >>> def none_loader(href, parser, encoding=None): ... return None >>> document = ET.XML(XINCLUDE["C1.xml"]) >>> ElementInclude.include(document, loader=none_loader) Traceback (most recent call last): xml.etree.ElementInclude.FatalIncludeError: cannot load 'disclaimer.xml' as 'xml' Test failure to locate included text file. >>> document = ET.XML(XINCLUDE["C2.xml"]) >>> ElementInclude.include(document, loader=none_loader) Traceback (most recent call last): xml.etree.ElementInclude.FatalIncludeError: cannot load 'count.txt' as 'text' Test bad parse type. >>> document = ET.XML(XINCLUDE_BAD["B1.xml"]) >>> ElementInclude.include(document, loader=none_loader) Traceback (most recent call last): xml.etree.ElementInclude.FatalIncludeError: unknown parse type in xi:include tag ('BAD_TYPE') Test xi:fallback outside xi:include. >>> document = ET.XML(XINCLUDE_BAD["B2.xml"]) >>> ElementInclude.include(document, loader=none_loader) Traceback (most recent call last): xml.etree.ElementInclude.FatalIncludeError: xi:fallback tag must be child of xi:include ('{http://www.w3.org/2001/XInclude}fallback') """ # -------------------------------------------------------------------- # reported bugs def bug_xmltoolkit21(): """ marshaller gives obscure errors for non-string values >>> elem = ET.Element(123) >>> serialize(elem) # tag Traceback (most recent call last): TypeError: cannot serialize 123 (type int) >>> elem = ET.Element("elem") >>> elem.text = 123 >>> serialize(elem) # text Traceback (most recent call last): TypeError: cannot serialize 123 (type int) >>> elem = ET.Element("elem") >>> elem.tail = 123 >>> serialize(elem) # tail Traceback (most recent call last): TypeError: cannot serialize 123 (type int) >>> elem = ET.Element("elem") >>> elem.set(123, "123") >>> serialize(elem) # attribute key Traceback (most recent call last): TypeError: cannot serialize 123 (type int) >>> elem = ET.Element("elem") >>> elem.set("123", 123) >>> serialize(elem) # attribute value Traceback (most recent call last): TypeError: cannot serialize 123 (type int) """ def bug_xmltoolkit25(): """ typo in ElementTree.findtext >>> elem = ET.XML(SAMPLE_XML) >>> tree = ET.ElementTree(elem) >>> tree.findtext("tag") 'text' >>> tree.findtext("section/tag") 'subtext' """ def bug_xmltoolkit28(): """ .//tag causes exceptions >>> tree = ET.XML("
") >>> summarize_list(tree.findall(".//thead")) [] >>> summarize_list(tree.findall(".//tbody")) ['tbody'] """ def bug_xmltoolkitX1(): """ dump() doesn't flush the output buffer >>> tree = ET.XML("
") >>> ET.dump(tree); print("tail")
tail """ def bug_xmltoolkit39(): """ non-ascii element and attribute names doesn't work >>> tree = ET.XML(b"") >>> ET.tostring(tree, "utf-8") b'' >>> tree = ET.XML(b"") >>> tree.attrib {'\\xe4ttr': 'v\\xe4lue'} >>> ET.tostring(tree, "utf-8") b'' >>> tree = ET.XML(b"text") >>> ET.tostring(tree, "utf-8") b'text' >>> tree = ET.Element("t\u00e4g") >>> ET.tostring(tree, "utf-8") b'' >>> tree = ET.Element("tag") >>> tree.set("\u00e4ttr", "v\u00e4lue") >>> ET.tostring(tree, "utf-8") b'' """ def bug_xmltoolkit54(): """ problems handling internally defined entities >>> e = ET.XML("]>&ldots;") >>> serialize(e, encoding="us-ascii") b'' >>> serialize(e) '\u8230' """ def bug_xmltoolkit55(): """ make sure we're reporting the first error, not the last >>> normalize_exception(ET.XML, b"&ldots;&ndots;&rdots;") Traceback (most recent call last): ParseError: undefined entity &ldots;: line 1, column 36 """ class ExceptionFile: def read(self, x): raise IOError def xmltoolkit60(): """ Handle crash in stream source. >>> tree = ET.parse(ExceptionFile()) Traceback (most recent call last): IOError """ XMLTOOLKIT62_DOC = """ A new cultivar of Begonia plant named ‘BCT9801BEG’. """ def xmltoolkit62(): """ Don't crash when using custom entities. >>> xmltoolkit62() 'A new cultivar of Begonia plant named \u2018BCT9801BEG\u2019.' """ ENTITIES = {'rsquo': '\u2019', 'lsquo': '\u2018'} parser = ET.XMLTreeBuilder() parser.entity.update(ENTITIES) parser.feed(XMLTOOLKIT62_DOC) t = parser.close() return t.find('.//paragraph').text def xmltoolkit63(): """ Check reference leak. >>> xmltoolkit63() >>> count = sys.getrefcount(None) >>> for i in range(1000): ... xmltoolkit63() >>> sys.getrefcount(None) - count 0 """ tree = ET.TreeBuilder() tree.start("tag", {}) tree.data("text") tree.end("tag") # -------------------------------------------------------------------- def bug_200708_newline(): r""" Preserve newlines in attributes. >>> e = ET.Element('SomeTag', text="def _f():\n return 3\n") >>> ET.tostring(e) '' >>> ET.XML(ET.tostring(e)).get("text") 'def _f():\n return 3\n' >>> ET.tostring(ET.XML(ET.tostring(e))) '' """ def bug_200708_close(): """ Test default builder. >>> parser = ET.XMLParser() # default >>> parser.feed("some text") >>> summarize(parser.close()) 'element' Test custom builder. >>> class EchoTarget: ... def close(self): ... return ET.Element("element") # simulate root >>> parser = ET.XMLParser(EchoTarget()) >>> parser.feed("some text") >>> summarize(parser.close()) 'element' """ def bug_200709_default_namespace(): """ >>> e = ET.Element("{default}elem") >>> s = ET.SubElement(e, "{default}elem") >>> serialize(e, default_namespace="default") # 1 '' >>> e = ET.Element("{default}elem") >>> s = ET.SubElement(e, "{default}elem") >>> s = ET.SubElement(e, "{not-default}elem") >>> serialize(e, default_namespace="default") # 2 '' >>> e = ET.Element("{default}elem") >>> s = ET.SubElement(e, "{default}elem") >>> s = ET.SubElement(e, "elem") # unprefixed name >>> serialize(e, default_namespace="default") # 3 Traceback (most recent call last): ValueError: cannot use non-qualified names with default_namespace option """ def bug_200709_register_namespace(): """ >>> ET.tostring(ET.Element("{http://namespace.invalid/does/not/exist/}title")) '' >>> ET.register_namespace("foo", "http://namespace.invalid/does/not/exist/") >>> ET.tostring(ET.Element("{http://namespace.invalid/does/not/exist/}title")) '' And the Dublin Core namespace is in the default list: >>> ET.tostring(ET.Element("{http://purl.org/dc/elements/1.1/}title")) '' """ def bug_200709_element_comment(): """ Not sure if this can be fixed, really (since the serializer needs ET.Comment, not cET.comment). >>> a = ET.Element('a') >>> a.append(ET.Comment('foo')) >>> a[0].tag == ET.Comment True >>> a = ET.Element('a') >>> a.append(ET.PI('foo')) >>> a[0].tag == ET.PI True """ def bug_200709_element_insert(): """ >>> a = ET.Element('a') >>> b = ET.SubElement(a, 'b') >>> c = ET.SubElement(a, 'c') >>> d = ET.Element('d') >>> a.insert(0, d) >>> summarize_list(a) ['d', 'b', 'c'] >>> a.insert(-1, d) >>> summarize_list(a) ['d', 'b', 'd', 'c'] """ def bug_200709_iter_comment(): """ >>> a = ET.Element('a') >>> b = ET.SubElement(a, 'b') >>> comment_b = ET.Comment("TEST-b") >>> b.append(comment_b) >>> summarize_list(a.iter(ET.Comment)) [''] """ # -------------------------------------------------------------------- # reported on bugs.python.org def bug_1534630(): """ >>> bob = ET.TreeBuilder() >>> e = bob.data("data") >>> e = bob.start("tag", {}) >>> e = bob.end("tag") >>> e = bob.close() >>> serialize(e) '' """ def check_issue6233(): """ >>> e = ET.XML(b"t\\xc3\\xa3g") >>> ET.tostring(e, 'ascii') b"\\ntãg" >>> e = ET.XML(b"t\\xe3g") >>> ET.tostring(e, 'ascii') b"\\ntãg" """ def check_issue3151(): """ >>> e = ET.XML('') >>> e.tag '{${stuff}}localname' >>> t = ET.ElementTree(e) >>> ET.tostring(e) '' """ def check_issue6565(): """ >>> elem = ET.XML("") >>> summarize_list(elem) ['tag'] >>> newelem = ET.XML(SAMPLE_XML) >>> elem[:] = newelem[:] >>> summarize_list(elem) ['tag', 'tag', 'section'] """ # -------------------------------------------------------------------- class CleanContext(object): """Provide default namespace mapping and path cache.""" def __enter__(self): from xml.etree import ElementTree self._nsmap = ElementTree._namespace_map self._path_cache = ElementTree.ElementPath._cache # Copy the default namespace mapping ElementTree._namespace_map = self._nsmap.copy() # Copy the path cache (should be empty) ElementTree.ElementPath._cache = self._path_cache.copy() def __exit__(self, *args): from xml.etree import ElementTree # Restore mapping and path cache ElementTree._namespace_map = self._nsmap ElementTree.ElementPath._cache = self._path_cache def test_main(module_name='xml.etree.ElementTree'): import warnings from test import test_xml_etree def ignore(message, category=DeprecationWarning): warnings.filterwarnings("ignore", message, category) # The same doctests are used for both the Python and the C implementations assert test_xml_etree.ET.__name__ == module_name with warnings.catch_warnings(), CleanContext(): # Search behaviour is broken if search path starts with "/". ignore("This search is broken in 1.3 and earlier, and will be fixed " "in a future version. If you rely on the current behaviour, " "change it to '.+'", FutureWarning) # Element.getchildren() and Element.getiterator() are deprecated. ignore("This method will be removed in future versions. " "Use .+ instead.") # XMLParser.doctype() is deprecated. ignore("This method of XMLParser is deprecated. " "Define doctype.. method on the TreeBuilder target.") support.run_doctest(test_xml_etree, verbosity=True) # The module should not be changed by the tests assert test_xml_etree.ET.__name__ == module_name if __name__ == '__main__': test_main()