diff options
Diffstat (limited to 'Lib/test/test_htmlparser.py')
| -rw-r--r--[-rwxr-xr-x] | Lib/test/test_htmlparser.py | 575 |
1 files changed, 492 insertions, 83 deletions
diff --git a/Lib/test/test_htmlparser.py b/Lib/test/test_htmlparser.py index 661d41d..c4f80cc 100755..100644 --- a/Lib/test/test_htmlparser.py +++ b/Lib/test/test_htmlparser.py @@ -8,10 +8,10 @@ from test import support class EventCollector(html.parser.HTMLParser): - def __init__(self): + def __init__(self, *args, **kw): self.events = [] self.append = self.events.append - html.parser.HTMLParser.__init__(self) + html.parser.HTMLParser.__init__(self, *args, **kw) def get_events(self): # Normalize the list of events so that buffer artefacts don't @@ -72,8 +72,13 @@ class EventCollectorExtra(EventCollector): class TestCaseBase(unittest.TestCase): - def _run_check(self, source, expected_events, collector=EventCollector): - parser = collector() + def get_collector(self): + raise NotImplementedError + + def _run_check(self, source, expected_events, collector=None): + if collector is None: + collector = self.get_collector() + parser = collector for s in source: parser.feed(s) parser.close() @@ -84,17 +89,20 @@ class TestCaseBase(unittest.TestCase): "\nReceived:\n" + pprint.pformat(events)) def _run_check_extra(self, source, events): - self._run_check(source, events, EventCollectorExtra) + self._run_check(source, events, EventCollectorExtra()) def _parse_error(self, source): def parse(source=source): - parser = html.parser.HTMLParser() + parser = self.get_collector() parser.feed(source) parser.close() self.assertRaises(html.parser.HTMLParseError, parse) -class HTMLParserTestCase(TestCaseBase): +class HTMLParserStrictTestCase(TestCaseBase): + + def get_collector(self): + return EventCollector(strict=True) def test_processing_instruction_only(self): self._run_check("<?processing instruction>", [ @@ -114,7 +122,7 @@ comment1b--> <Img sRc='Bar' isMAP>sample text “ -<!--comment2a-- --comment2b--><!> +<!--comment2a-- --comment2b--> </Html> """, [ ("data", "\n"), @@ -149,24 +157,6 @@ text ("data", " foo"), ]) - def test_doctype_decl(self): - inside = """\ -DOCTYPE html [ - <!ELEMENT html - O EMPTY> - <!ATTLIST html - version CDATA #IMPLIED - profile CDATA 'DublinCore'> - <!NOTATION datatype SYSTEM 'http://xml.python.org/notations/python-module'> - <!ENTITY myEntity 'internal parsed entity'> - <!ENTITY anEntity SYSTEM 'http://xml.python.org/entities/something.xml'> - <!ENTITY % paramEntity 'name|name|name'> - %paramEntity; - <!-- comment --> -]""" - self._run_check("<!%s>" % inside, [ - ("decl", inside), - ]) - def test_bad_nesting(self): # Strangely, this *is* supposed to test that overlapping # elements are allowed. HTMLParser is more geared toward @@ -188,43 +178,6 @@ DOCTYPE html [ ("data", "this < text > contains < bare>pointy< brackets"), ]) - def test_attr_syntax(self): - output = [ - ("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", None)]) - ] - self._run_check("""<a b='v' c="v" d=v e>""", output) - self._run_check("""<a b = 'v' c = "v" d = v e>""", output) - self._run_check("""<a\nb\n=\n'v'\nc\n=\n"v"\nd\n=\nv\ne>""", output) - self._run_check("""<a\tb\t=\t'v'\tc\t=\t"v"\td\t=\tv\te>""", output) - - def test_attr_values(self): - self._run_check("""<a b='xxx\n\txxx' c="yyy\t\nyyy" d='\txyz\n'>""", - [("starttag", "a", [("b", "xxx\n\txxx"), - ("c", "yyy\t\nyyy"), - ("d", "\txyz\n")]) - ]) - self._run_check("""<a b='' c="">""", [ - ("starttag", "a", [("b", ""), ("c", "")]), - ]) - # Regression test for SF patch #669683. - self._run_check("<e a=rgb(1,2,3)>", [ - ("starttag", "e", [("a", "rgb(1,2,3)")]), - ]) - # Regression test for SF bug #921657. - self._run_check("<a href=mailto:xyz@example.com>", [ - ("starttag", "a", [("href", "mailto:xyz@example.com")]), - ]) - - def test_attr_entity_replacement(self): - self._run_check("""<a b='&><"''>""", [ - ("starttag", "a", [("b", "&><\"'")]), - ]) - - def test_attr_funky_names(self): - self._run_check("""<a a.b='v' c:d=v e-f=v>""", [ - ("starttag", "a", [("a.b", "v"), ("c:d", "v"), ("e-f", "v")]), - ]) - def test_illegal_declarations(self): self._parse_error('<!spacer type="block" height="25">') @@ -270,13 +223,35 @@ DOCTYPE html [ self._parse_error("<a<a>") self._parse_error("</a<a>") self._parse_error("<!") - self._parse_error("<a $>") self._parse_error("<a") self._parse_error("<a foo='bar'") self._parse_error("<a foo='bar") self._parse_error("<a foo='>'") self._parse_error("<a foo='>") - self._parse_error("<a foo=>") + + def test_valid_doctypes(self): + # from http://www.w3.org/QA/2002/04/valid-dtd-list.html + dtds = ['HTML', # HTML5 doctype + ('HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" ' + '"http://www.w3.org/TR/html4/strict.dtd"'), + ('HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" ' + '"http://www.w3.org/TR/html4/loose.dtd"'), + ('html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" ' + '"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"'), + ('html PUBLIC "-//W3C//DTD XHTML 1.0 Frameset//EN" ' + '"http://www.w3.org/TR/xhtml1/DTD/xhtml1-frameset.dtd"'), + ('math PUBLIC "-//W3C//DTD MathML 2.0//EN" ' + '"http://www.w3.org/Math/DTD/mathml2/mathml2.dtd"'), + ('html PUBLIC "-//W3C//DTD ' + 'XHTML 1.1 plus MathML 2.0 plus SVG 1.1//EN" ' + '"http://www.w3.org/2002/04/xhtml-math-svg/xhtml-math-svg.dtd"'), + ('svg PUBLIC "-//W3C//DTD SVG 1.1//EN" ' + '"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd"'), + 'html PUBLIC "-//IETF//DTD HTML 2.0//EN"', + 'html PUBLIC "-//W3C//DTD HTML 3.2 Final//EN"'] + for dtd in dtds: + self._run_check("<!DOCTYPE %s>" % dtd, + [('decl', 'DOCTYPE ' + dtd)]) def test_declaration_junk_chars(self): self._parse_error("<!DOCTYPE foo $ >") @@ -302,31 +277,465 @@ DOCTYPE html [ ("starttag_text", s)]) def test_cdata_content(self): - s = """<script> <!-- not a comment --> ¬-an-entity-ref; </script>""" - self._run_check(s, [ - ("starttag", "script", []), - ("data", " <!-- not a comment --> ¬-an-entity-ref; "), - ("endtag", "script"), - ]) - s = """<script> <not a='start tag'> </script>""" - self._run_check(s, [ - ("starttag", "script", []), - ("data", " <not a='start tag'> "), - ("endtag", "script"), - ]) + contents = [ + '<!-- not a comment --> ¬-an-entity-ref;', + "<not a='start tag'>", + '<a href="" /> <p> <span></span>', + 'foo = "</scr" + "ipt>";', + 'foo = "</SCRIPT" + ">";', + 'foo = <\n/script> ', + '<!-- document.write("</scr" + "ipt>"); -->', + ('\n//<![CDATA[\n' + 'document.write(\'<s\'+\'cript type="text/javascript" ' + 'src="http://www.example.org/r=\'+new ' + 'Date().getTime()+\'"><\\/s\'+\'cript>\');\n//]]>'), + '\n<!-- //\nvar foo = 3.14;\n// -->\n', + 'foo = "</sty" + "le>";', + '<!-- \u2603 -->', + # these two should be invalid according to the HTML 5 spec, + # section 8.1.2.2 + #'foo = </\nscript>', + #'foo = </ script>', + ] + elements = ['script', 'style', 'SCRIPT', 'STYLE', 'Script', 'Style'] + for content in contents: + for element in elements: + element_lower = element.lower() + s = '<{element}>{content}</{element}>'.format(element=element, + content=content) + self._run_check(s, [("starttag", element_lower, []), + ("data", content), + ("endtag", element_lower)]) + + def test_cdata_with_closing_tags(self): + # see issue #13358 + # make sure that HTMLParser calls handle_data only once for each CDATA. + # The normal event collector normalizes the events in get_events, + # so we override it to return the original list of events. + class Collector(EventCollector): + def get_events(self): + return self.events + + content = """<!-- not a comment --> ¬-an-entity-ref; + <a href="" /> </p><p> <span></span></style> + '</script' + '>'""" + for element in [' script', 'script ', ' script ', + '\nscript', 'script\n', '\nscript\n']: + element_lower = element.lower().strip() + s = '<script>{content}</{element}>'.format(element=element, + content=content) + self._run_check(s, [("starttag", element_lower, []), + ("data", content), + ("endtag", element_lower)], + collector=Collector()) + + def test_comments(self): + html = ("<!-- I'm a valid comment -->" + '<!--me too!-->' + '<!------>' + '<!---->' + '<!----I have many hyphens---->' + '<!-- I have a > in the middle -->' + '<!-- and I have -- in the middle! -->') + expected = [('comment', " I'm a valid comment "), + ('comment', 'me too!'), + ('comment', '--'), + ('comment', ''), + ('comment', '--I have many hyphens--'), + ('comment', ' I have a > in the middle '), + ('comment', ' and I have -- in the middle! ')] + self._run_check(html, expected) + + def test_condcoms(self): + html = ('<!--[if IE & !(lte IE 8)]>aren\'t<![endif]-->' + '<!--[if IE 8]>condcoms<![endif]-->' + '<!--[if lte IE 7]>pretty?<![endif]-->') + expected = [('comment', "[if IE & !(lte IE 8)]>aren't<![endif]"), + ('comment', '[if IE 8]>condcoms<![endif]'), + ('comment', '[if lte IE 7]>pretty?<![endif]')] + self._run_check(html, expected) + + +class HTMLParserTolerantTestCase(HTMLParserStrictTestCase): + + def get_collector(self): + return EventCollector(strict=False) + + def test_tolerant_parsing(self): + self._run_check('<html <html>te>>xt&a<<bc</a></html>\n' + '<img src="URL><//img></html</html>', [ + ('starttag', 'html', [('<html', None)]), + ('data', 'te>>xt'), + ('entityref', 'a'), + ('data', '<<bc'), + ('endtag', 'a'), + ('endtag', 'html'), + ('data', '\n<img src="URL>'), + ('comment', '/img'), + ('endtag', 'html<')]) + + def test_starttag_junk_chars(self): + self._run_check("</>", []) + self._run_check("</$>", [('comment', '$')]) + self._run_check("</", [('data', '</')]) + self._run_check("</a", [('data', '</a')]) + # XXX this might be wrong + self._run_check("<a<a>", [('data', '<a'), ('starttag', 'a', [])]) + self._run_check("</a<a>", [('endtag', 'a<a')]) + self._run_check("<!", [('data', '<!')]) + self._run_check("<a", [('data', '<a')]) + self._run_check("<a foo='bar'", [('data', "<a foo='bar'")]) + self._run_check("<a foo='bar", [('data', "<a foo='bar")]) + self._run_check("<a foo='>'", [('data', "<a foo='>'")]) + self._run_check("<a foo='>", [('data', "<a foo='>")]) + + def test_slashes_in_starttag(self): + self._run_check('<a foo="var"/>', [('startendtag', 'a', [('foo', 'var')])]) + html = ('<img width=902 height=250px ' + 'src="/sites/default/files/images/homepage/foo.jpg" ' + '/*what am I doing here*/ />') + expected = [( + 'startendtag', 'img', + [('width', '902'), ('height', '250px'), + ('src', '/sites/default/files/images/homepage/foo.jpg'), + ('*what', None), ('am', None), ('i', None), + ('doing', None), ('here*', None)] + )] + self._run_check(html, expected) + html = ('<a / /foo/ / /=/ / /bar/ / />' + '<a / /foo/ / /=/ / /bar/ / >') + expected = [ + ('startendtag', 'a', [('foo', None), ('=', None), ('bar', None)]), + ('starttag', 'a', [('foo', None), ('=', None), ('bar', None)]) + ] + self._run_check(html, expected) + #see issue #14538 + html = ('<meta><meta / ><meta // ><meta / / >' + '<meta/><meta /><meta //><meta//>') + expected = [ + ('starttag', 'meta', []), ('starttag', 'meta', []), + ('starttag', 'meta', []), ('starttag', 'meta', []), + ('startendtag', 'meta', []), ('startendtag', 'meta', []), + ('startendtag', 'meta', []), ('startendtag', 'meta', []), + ] + self._run_check(html, expected) + + def test_declaration_junk_chars(self): + self._run_check("<!DOCTYPE foo $ >", [('decl', 'DOCTYPE foo $ ')]) + + def test_illegal_declarations(self): + self._run_check('<!spacer type="block" height="25">', + [('comment', 'spacer type="block" height="25"')]) + + def test_with_unquoted_attributes(self): + # see #12008 + html = ("<html><body bgcolor=d0ca90 text='181008'>" + "<table cellspacing=0 cellpadding=1 width=100% ><tr>" + "<td align=left><font size=-1>" + "- <a href=/rabota/><span class=en> software-and-i</span></a>" + "- <a href='/1/'><span class=en> library</span></a></table>") + expected = [ + ('starttag', 'html', []), + ('starttag', 'body', [('bgcolor', 'd0ca90'), ('text', '181008')]), + ('starttag', 'table', + [('cellspacing', '0'), ('cellpadding', '1'), ('width', '100%')]), + ('starttag', 'tr', []), + ('starttag', 'td', [('align', 'left')]), + ('starttag', 'font', [('size', '-1')]), + ('data', '- '), ('starttag', 'a', [('href', '/rabota/')]), + ('starttag', 'span', [('class', 'en')]), ('data', ' software-and-i'), + ('endtag', 'span'), ('endtag', 'a'), + ('data', '- '), ('starttag', 'a', [('href', '/1/')]), + ('starttag', 'span', [('class', 'en')]), ('data', ' library'), + ('endtag', 'span'), ('endtag', 'a'), ('endtag', 'table') + ] + self._run_check(html, expected) + + def test_comma_between_attributes(self): + self._run_check('<form action="/xxx.php?a=1&b=2&", ' + 'method="post">', [ + ('starttag', 'form', + [('action', '/xxx.php?a=1&b=2&'), + (',', None), ('method', 'post')])]) + + def test_weird_chars_in_unquoted_attribute_values(self): + self._run_check('<form action=bogus|&#()value>', [ + ('starttag', 'form', + [('action', 'bogus|&#()value')])]) + + def test_invalid_end_tags(self): + # A collection of broken end tags. <br> is used as separator. + # see http://www.w3.org/TR/html5/tokenization.html#end-tag-open-state + # and #13993 + html = ('<br></label</p><br></div end tmAd-leaderBoard><br></<h4><br>' + '</li class="unit"><br></li\r\n\t\t\t\t\t\t</ul><br></><br>') + expected = [('starttag', 'br', []), + # < is part of the name, / is discarded, p is an attribute + ('endtag', 'label<'), + ('starttag', 'br', []), + # text and attributes are discarded + ('endtag', 'div'), + ('starttag', 'br', []), + # comment because the first char after </ is not a-zA-Z + ('comment', '<h4'), + ('starttag', 'br', []), + # attributes are discarded + ('endtag', 'li'), + ('starttag', 'br', []), + # everything till ul (included) is discarded + ('endtag', 'li'), + ('starttag', 'br', []), + # </> is ignored + ('starttag', 'br', [])] + self._run_check(html, expected) + + def test_broken_invalid_end_tag(self): + # This is technically wrong (the "> shouldn't be included in the 'data') + # but is probably not worth fixing it (in addition to all the cases of + # the previous test, it would require a full attribute parsing). + # see #13993 + html = '<b>This</b attr=">"> confuses the parser' + expected = [('starttag', 'b', []), + ('data', 'This'), + ('endtag', 'b'), + ('data', '"> confuses the parser')] + self._run_check(html, expected) + + def test_correct_detection_of_start_tags(self): + # see #13273 + html = ('<div style="" ><b>The <a href="some_url">rain</a> ' + '<br /> in <span>Spain</span></b></div>') + expected = [ + ('starttag', 'div', [('style', '')]), + ('starttag', 'b', []), + ('data', 'The '), + ('starttag', 'a', [('href', 'some_url')]), + ('data', 'rain'), + ('endtag', 'a'), + ('data', ' '), + ('startendtag', 'br', []), + ('data', ' in '), + ('starttag', 'span', []), + ('data', 'Spain'), + ('endtag', 'span'), + ('endtag', 'b'), + ('endtag', 'div') + ] + self._run_check(html, expected) + + html = '<div style="", foo = "bar" ><b>The <a href="some_url">rain</a>' + expected = [ + ('starttag', 'div', [('style', ''), (',', None), ('foo', 'bar')]), + ('starttag', 'b', []), + ('data', 'The '), + ('starttag', 'a', [('href', 'some_url')]), + ('data', 'rain'), + ('endtag', 'a'), + ] + self._run_check(html, expected) - def test_entityrefs_in_attributes(self): - self._run_check("<html foo='€&aa&unsupported;'>", [ - ("starttag", "html", [("foo", "\u20AC&aa&unsupported;")]) - ]) def test_unescape_function(self): - p = html.parser.HTMLParser() + p = self.get_collector() self.assertEqual(p.unescape('&#bad;'),'&#bad;') self.assertEqual(p.unescape('&'),'&') + # see #12888 + self.assertEqual(p.unescape('{ ' * 1050), '{ ' * 1050) + + def test_broken_comments(self): + html = ('<! not really a comment >' + '<! not a comment either -->' + '<! -- close enough -->' + '<!><!<-- this was an empty comment>' + '<!!! another bogus comment !!!>') + expected = [ + ('comment', ' not really a comment '), + ('comment', ' not a comment either --'), + ('comment', ' -- close enough --'), + ('comment', ''), + ('comment', '<-- this was an empty comment'), + ('comment', '!! another bogus comment !!!'), + ] + self._run_check(html, expected) + + def test_broken_condcoms(self): + # these condcoms are missing the '--' after '<!' and before the '>' + html = ('<![if !(IE)]>broken condcom<![endif]>' + '<![if ! IE]><link href="favicon.tiff"/><![endif]>' + '<![if !IE 6]><img src="firefox.png" /><![endif]>' + '<![if !ie 6]><b>foo</b><![endif]>' + '<![if (!IE)|(lt IE 9)]><img src="mammoth.bmp" /><![endif]>') + # According to the HTML5 specs sections "8.2.4.44 Bogus comment state" + # and "8.2.4.45 Markup declaration open state", comment tokens should + # be emitted instead of 'unknown decl', but calling unknown_decl + # provides more flexibility. + # See also Lib/_markupbase.py:parse_declaration + expected = [ + ('unknown decl', 'if !(IE)'), + ('data', 'broken condcom'), + ('unknown decl', 'endif'), + ('unknown decl', 'if ! IE'), + ('startendtag', 'link', [('href', 'favicon.tiff')]), + ('unknown decl', 'endif'), + ('unknown decl', 'if !IE 6'), + ('startendtag', 'img', [('src', 'firefox.png')]), + ('unknown decl', 'endif'), + ('unknown decl', 'if !ie 6'), + ('starttag', 'b', []), + ('data', 'foo'), + ('endtag', 'b'), + ('unknown decl', 'endif'), + ('unknown decl', 'if (!IE)|(lt IE 9)'), + ('startendtag', 'img', [('src', 'mammoth.bmp')]), + ('unknown decl', 'endif') + ] + self._run_check(html, expected) + + +class AttributesStrictTestCase(TestCaseBase): + + def get_collector(self): + return EventCollector(strict=True) + + def test_attr_syntax(self): + output = [ + ("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", None)]) + ] + self._run_check("""<a b='v' c="v" d=v e>""", output) + self._run_check("""<a b = 'v' c = "v" d = v e>""", output) + self._run_check("""<a\nb\n=\n'v'\nc\n=\n"v"\nd\n=\nv\ne>""", output) + self._run_check("""<a\tb\t=\t'v'\tc\t=\t"v"\td\t=\tv\te>""", output) + + def test_attr_values(self): + self._run_check("""<a b='xxx\n\txxx' c="yyy\t\nyyy" d='\txyz\n'>""", + [("starttag", "a", [("b", "xxx\n\txxx"), + ("c", "yyy\t\nyyy"), + ("d", "\txyz\n")])]) + self._run_check("""<a b='' c="">""", + [("starttag", "a", [("b", ""), ("c", "")])]) + # Regression test for SF patch #669683. + self._run_check("<e a=rgb(1,2,3)>", + [("starttag", "e", [("a", "rgb(1,2,3)")])]) + # Regression test for SF bug #921657. + self._run_check( + "<a href=mailto:xyz@example.com>", + [("starttag", "a", [("href", "mailto:xyz@example.com")])]) + + def test_attr_nonascii(self): + # see issue 7311 + self._run_check( + "<img src=/foo/bar.png alt=\u4e2d\u6587>", + [("starttag", "img", [("src", "/foo/bar.png"), + ("alt", "\u4e2d\u6587")])]) + self._run_check( + "<a title='\u30c6\u30b9\u30c8' href='\u30c6\u30b9\u30c8.html'>", + [("starttag", "a", [("title", "\u30c6\u30b9\u30c8"), + ("href", "\u30c6\u30b9\u30c8.html")])]) + self._run_check( + '<a title="\u30c6\u30b9\u30c8" href="\u30c6\u30b9\u30c8.html">', + [("starttag", "a", [("title", "\u30c6\u30b9\u30c8"), + ("href", "\u30c6\u30b9\u30c8.html")])]) + + def test_attr_entity_replacement(self): + self._run_check( + "<a b='&><"''>", + [("starttag", "a", [("b", "&><\"'")])]) + + def test_attr_funky_names(self): + self._run_check( + "<a a.b='v' c:d=v e-f=v>", + [("starttag", "a", [("a.b", "v"), ("c:d", "v"), ("e-f", "v")])]) + + def test_entityrefs_in_attributes(self): + self._run_check( + "<html foo='€&aa&unsupported;'>", + [("starttag", "html", [("foo", "\u20AC&aa&unsupported;")])]) + + + +class AttributesTolerantTestCase(AttributesStrictTestCase): + + def get_collector(self): + return EventCollector(strict=False) + + def test_attr_funky_names2(self): + self._run_check( + "<a $><b $=%><c \=/>", + [("starttag", "a", [("$", None)]), + ("starttag", "b", [("$", "%")]), + ("starttag", "c", [("\\", "/")])]) + + def test_entities_in_attribute_value(self): + # see #1200313 + for entity in ['&', '&', '&', '&']: + self._run_check('<a href="%s">' % entity, + [("starttag", "a", [("href", "&")])]) + self._run_check("<a href='%s'>" % entity, + [("starttag", "a", [("href", "&")])]) + self._run_check("<a href=%s>" % entity, + [("starttag", "a", [("href", "&")])]) + + def test_malformed_attributes(self): + # see #13357 + html = ( + "<a href=test'style='color:red;bad1'>test - bad1</a>" + "<a href=test'+style='color:red;ba2'>test - bad2</a>" + "<a href=test' style='color:red;bad3'>test - bad3</a>" + "<a href = test' style='color:red;bad4' >test - bad4</a>" + ) + expected = [ + ('starttag', 'a', [('href', "test'style='color:red;bad1'")]), + ('data', 'test - bad1'), ('endtag', 'a'), + ('starttag', 'a', [('href', "test'+style='color:red;ba2'")]), + ('data', 'test - bad2'), ('endtag', 'a'), + ('starttag', 'a', [('href', "test'\xa0style='color:red;bad3'")]), + ('data', 'test - bad3'), ('endtag', 'a'), + ('starttag', 'a', [('href', "test'\xa0style='color:red;bad4'")]), + ('data', 'test - bad4'), ('endtag', 'a') + ] + self._run_check(html, expected) + + def test_malformed_adjacent_attributes(self): + # see #12629 + self._run_check('<x><y z=""o"" /></x>', + [('starttag', 'x', []), + ('startendtag', 'y', [('z', ''), ('o""', None)]), + ('endtag', 'x')]) + self._run_check('<x><y z="""" /></x>', + [('starttag', 'x', []), + ('startendtag', 'y', [('z', ''), ('""', None)]), + ('endtag', 'x')]) + + # see #755670 for the following 3 tests + def test_adjacent_attributes(self): + self._run_check('<a width="100%"cellspacing=0>', + [("starttag", "a", + [("width", "100%"), ("cellspacing","0")])]) + + self._run_check('<a id="foo"class="bar">', + [("starttag", "a", + [("id", "foo"), ("class","bar")])]) + + def test_missing_attribute_value(self): + self._run_check('<a v=>', + [("starttag", "a", [("v", "")])]) + + def test_javascript_attribute_value(self): + self._run_check("<a href=javascript:popup('/popup/help.html')>", + [("starttag", "a", + [("href", "javascript:popup('/popup/help.html')")])]) + + def test_end_tag_in_attribute_value(self): + # see #1745761 + self._run_check("<a href='http://www.example.org/\">;'>spam</a>", + [("starttag", "a", + [("href", "http://www.example.org/\">;")]), + ("data", "spam"), ("endtag", "a")]) + def test_main(): - support.run_unittest(HTMLParserTestCase) + support.run_unittest(HTMLParserStrictTestCase, HTMLParserTolerantTestCase, + AttributesStrictTestCase, AttributesTolerantTestCase) if __name__ == "__main__": |