blob: cde2bd23b74f9b39254a6557b8bdd30d40d26035 [file] [log] [blame]
Fred Drakebd3090d2001-05-18 15:32:59 +00001"""Tests for HTMLParser.py."""
2
Georg Brandlbcdafa42008-05-20 07:58:42 +00003import HTMLParser
Fred Drake029acfb2001-08-20 21:24:19 +00004import pprint
Fred Drakebd3090d2001-05-18 15:32:59 +00005import unittest
Barry Warsaw04f357c2002-07-23 19:04:11 +00006from test import test_support
Fred Drakebd3090d2001-05-18 15:32:59 +00007
8
Georg Brandlbcdafa42008-05-20 07:58:42 +00009class EventCollector(HTMLParser.HTMLParser):
Fred Drakebd3090d2001-05-18 15:32:59 +000010
11 def __init__(self):
12 self.events = []
13 self.append = self.events.append
Georg Brandlbcdafa42008-05-20 07:58:42 +000014 HTMLParser.HTMLParser.__init__(self)
Fred Drakebd3090d2001-05-18 15:32:59 +000015
16 def get_events(self):
17 # Normalize the list of events so that buffer artefacts don't
18 # separate runs of contiguous characters.
19 L = []
20 prevtype = None
21 for event in self.events:
22 type = event[0]
23 if type == prevtype == "data":
24 L[-1] = ("data", L[-1][1] + event[1])
25 else:
26 L.append(event)
27 prevtype = type
28 self.events = L
29 return L
30
31 # structure markup
32
33 def handle_starttag(self, tag, attrs):
34 self.append(("starttag", tag, attrs))
35
36 def handle_startendtag(self, tag, attrs):
37 self.append(("startendtag", tag, attrs))
38
39 def handle_endtag(self, tag):
40 self.append(("endtag", tag))
41
42 # all other markup
43
44 def handle_comment(self, data):
45 self.append(("comment", data))
46
47 def handle_charref(self, data):
48 self.append(("charref", data))
49
50 def handle_data(self, data):
51 self.append(("data", data))
52
53 def handle_decl(self, data):
54 self.append(("decl", data))
55
56 def handle_entityref(self, data):
57 self.append(("entityref", data))
58
59 def handle_pi(self, data):
60 self.append(("pi", data))
61
Fred Drakec20a6982001-09-04 15:13:04 +000062 def unknown_decl(self, decl):
63 self.append(("unknown decl", decl))
64
Fred Drakebd3090d2001-05-18 15:32:59 +000065
66class EventCollectorExtra(EventCollector):
67
68 def handle_starttag(self, tag, attrs):
69 EventCollector.handle_starttag(self, tag, attrs)
70 self.append(("starttag_text", self.get_starttag_text()))
71
72
73class TestCaseBase(unittest.TestCase):
74
Fred Drakec20a6982001-09-04 15:13:04 +000075 def _run_check(self, source, expected_events, collector=EventCollector):
Fred Drakebd3090d2001-05-18 15:32:59 +000076 parser = collector()
Fred Drakebd3090d2001-05-18 15:32:59 +000077 for s in source:
78 parser.feed(s)
Fred Drakebd3090d2001-05-18 15:32:59 +000079 parser.close()
Fred Drake029acfb2001-08-20 21:24:19 +000080 events = parser.get_events()
Fred Drakec20a6982001-09-04 15:13:04 +000081 if events != expected_events:
82 self.fail("received events did not match expected events\n"
83 "Expected:\n" + pprint.pformat(expected_events) +
84 "\nReceived:\n" + pprint.pformat(events))
Fred Drakebd3090d2001-05-18 15:32:59 +000085
86 def _run_check_extra(self, source, events):
87 self._run_check(source, events, EventCollectorExtra)
88
89 def _parse_error(self, source):
90 def parse(source=source):
Georg Brandlbcdafa42008-05-20 07:58:42 +000091 parser = HTMLParser.HTMLParser()
Fred Drakebd3090d2001-05-18 15:32:59 +000092 parser.feed(source)
93 parser.close()
Georg Brandlbcdafa42008-05-20 07:58:42 +000094 self.assertRaises(HTMLParser.HTMLParseError, parse)
Fred Drakebd3090d2001-05-18 15:32:59 +000095
96
97class HTMLParserTestCase(TestCaseBase):
98
Fred Drake84bb9d82001-08-03 19:53:01 +000099 def test_processing_instruction_only(self):
Fred Drakebd3090d2001-05-18 15:32:59 +0000100 self._run_check("<?processing instruction>", [
101 ("pi", "processing instruction"),
102 ])
Fred Drakefafd56f2003-04-17 22:19:26 +0000103 self._run_check("<?processing instruction ?>", [
104 ("pi", "processing instruction ?"),
105 ])
Fred Drakebd3090d2001-05-18 15:32:59 +0000106
Fred Drake84bb9d82001-08-03 19:53:01 +0000107 def test_simple_html(self):
Fred Drakebd3090d2001-05-18 15:32:59 +0000108 self._run_check("""
109<!DOCTYPE html PUBLIC 'foo'>
110<HTML>&entity;&#32;
111<!--comment1a
112-></foo><bar>&lt;<?pi?></foo<bar
113comment1b-->
114<Img sRc='Bar' isMAP>sample
115text
Fred Drake84bb9d82001-08-03 19:53:01 +0000116&#x201C;
Ezio Melotti4b92cc32012-02-13 16:10:44 +0200117<!--comment2a-- --comment2b-->
Fred Drakebd3090d2001-05-18 15:32:59 +0000118</Html>
119""", [
120 ("data", "\n"),
121 ("decl", "DOCTYPE html PUBLIC 'foo'"),
122 ("data", "\n"),
123 ("starttag", "html", []),
124 ("entityref", "entity"),
125 ("charref", "32"),
126 ("data", "\n"),
127 ("comment", "comment1a\n-></foo><bar>&lt;<?pi?></foo<bar\ncomment1b"),
128 ("data", "\n"),
129 ("starttag", "img", [("src", "Bar"), ("ismap", None)]),
130 ("data", "sample\ntext\n"),
Fred Drake84bb9d82001-08-03 19:53:01 +0000131 ("charref", "x201C"),
132 ("data", "\n"),
Fred Drakebd3090d2001-05-18 15:32:59 +0000133 ("comment", "comment2a-- --comment2b"),
134 ("data", "\n"),
135 ("endtag", "html"),
136 ("data", "\n"),
137 ])
138
Fred Drake073148c2001-12-03 16:44:09 +0000139 def test_unclosed_entityref(self):
140 self._run_check("&entityref foo", [
141 ("entityref", "entityref"),
142 ("data", " foo"),
143 ])
144
Fred Drake84bb9d82001-08-03 19:53:01 +0000145 def test_bad_nesting(self):
146 # Strangely, this *is* supposed to test that overlapping
147 # elements are allowed. HTMLParser is more geared toward
148 # lexing the input that parsing the structure.
Fred Drakebd3090d2001-05-18 15:32:59 +0000149 self._run_check("<a><b></a></b>", [
150 ("starttag", "a", []),
151 ("starttag", "b", []),
152 ("endtag", "a"),
153 ("endtag", "b"),
154 ])
155
Fred Drake029acfb2001-08-20 21:24:19 +0000156 def test_bare_ampersands(self):
157 self._run_check("this text & contains & ampersands &", [
158 ("data", "this text & contains & ampersands &"),
159 ])
160
161 def test_bare_pointy_brackets(self):
162 self._run_check("this < text > contains < bare>pointy< brackets", [
163 ("data", "this < text > contains < bare>pointy< brackets"),
164 ])
165
Fred Drakec20a6982001-09-04 15:13:04 +0000166 def test_illegal_declarations(self):
Ezio Melotti4b92cc32012-02-13 16:10:44 +0200167 self._run_check('<!spacer type="block" height="25">',
168 [('comment', 'spacer type="block" height="25"')])
Fred Drakec20a6982001-09-04 15:13:04 +0000169
Fred Drake84bb9d82001-08-03 19:53:01 +0000170 def test_starttag_end_boundary(self):
Fred Drakebd3090d2001-05-18 15:32:59 +0000171 self._run_check("""<a b='<'>""", [("starttag", "a", [("b", "<")])])
172 self._run_check("""<a b='>'>""", [("starttag", "a", [("b", ">")])])
173
Fred Drake84bb9d82001-08-03 19:53:01 +0000174 def test_buffer_artefacts(self):
Fred Drakebd3090d2001-05-18 15:32:59 +0000175 output = [("starttag", "a", [("b", "<")])]
176 self._run_check(["<a b='<'>"], output)
177 self._run_check(["<a ", "b='<'>"], output)
178 self._run_check(["<a b", "='<'>"], output)
179 self._run_check(["<a b=", "'<'>"], output)
180 self._run_check(["<a b='<", "'>"], output)
181 self._run_check(["<a b='<'", ">"], output)
182
183 output = [("starttag", "a", [("b", ">")])]
184 self._run_check(["<a b='>'>"], output)
185 self._run_check(["<a ", "b='>'>"], output)
186 self._run_check(["<a b", "='>'>"], output)
187 self._run_check(["<a b=", "'>'>"], output)
188 self._run_check(["<a b='>", "'>"], output)
189 self._run_check(["<a b='>'", ">"], output)
190
Fred Drake75d9a622004-09-08 22:57:01 +0000191 output = [("comment", "abc")]
192 self._run_check(["", "<!--abc-->"], output)
193 self._run_check(["<", "!--abc-->"], output)
194 self._run_check(["<!", "--abc-->"], output)
195 self._run_check(["<!-", "-abc-->"], output)
196 self._run_check(["<!--", "abc-->"], output)
197 self._run_check(["<!--a", "bc-->"], output)
198 self._run_check(["<!--ab", "c-->"], output)
199 self._run_check(["<!--abc", "-->"], output)
200 self._run_check(["<!--abc-", "->"], output)
201 self._run_check(["<!--abc--", ">"], output)
202 self._run_check(["<!--abc-->", ""], output)
203
Fred Drake84bb9d82001-08-03 19:53:01 +0000204 def test_starttag_junk_chars(self):
Ezio Melottif1174432012-02-13 16:28:54 +0200205 self._run_check("</>", [])
206 self._run_check("</$>", [('comment', '$')])
Ezio Melottid2307cb2012-02-15 12:44:23 +0200207 self._run_check("</", [('data', '</')])
208 self._run_check("</a", [('data', '</a')])
Ezio Melottib8147452013-11-07 18:31:36 +0200209 self._run_check("<a<a>", [('starttag', 'a<a', [])])
Ezio Melottif1174432012-02-13 16:28:54 +0200210 self._run_check("</a<a>", [('endtag', 'a<a')])
Ezio Melottid2307cb2012-02-15 12:44:23 +0200211 self._run_check("<!", [('data', '<!')])
212 self._run_check("<a", [('data', '<a')])
213 self._run_check("<a foo='bar'", [('data', "<a foo='bar'")])
214 self._run_check("<a foo='bar", [('data', "<a foo='bar")])
215 self._run_check("<a foo='>'", [('data', "<a foo='>'")])
216 self._run_check("<a foo='>", [('data', "<a foo='>")])
Ezio Melottib8147452013-11-07 18:31:36 +0200217 self._run_check("<a$>", [('starttag', 'a$', [])])
218 self._run_check("<a$b>", [('starttag', 'a$b', [])])
219 self._run_check("<a$b/>", [('startendtag', 'a$b', [])])
220 self._run_check("<a$b >", [('starttag', 'a$b', [])])
221 self._run_check("<a$b />", [('startendtag', 'a$b', [])])
Fred Drakebd3090d2001-05-18 15:32:59 +0000222
Ezio Melotti369cbd72012-02-13 20:36:55 +0200223 def test_valid_doctypes(self):
224 # from http://www.w3.org/QA/2002/04/valid-dtd-list.html
225 dtds = ['HTML', # HTML5 doctype
226 ('HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" '
227 '"http://www.w3.org/TR/html4/strict.dtd"'),
228 ('HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" '
229 '"http://www.w3.org/TR/html4/loose.dtd"'),
230 ('html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" '
231 '"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"'),
232 ('html PUBLIC "-//W3C//DTD XHTML 1.0 Frameset//EN" '
233 '"http://www.w3.org/TR/xhtml1/DTD/xhtml1-frameset.dtd"'),
234 ('math PUBLIC "-//W3C//DTD MathML 2.0//EN" '
235 '"http://www.w3.org/Math/DTD/mathml2/mathml2.dtd"'),
236 ('html PUBLIC "-//W3C//DTD '
237 'XHTML 1.1 plus MathML 2.0 plus SVG 1.1//EN" '
238 '"http://www.w3.org/2002/04/xhtml-math-svg/xhtml-math-svg.dtd"'),
239 ('svg PUBLIC "-//W3C//DTD SVG 1.1//EN" '
240 '"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd"'),
241 'html PUBLIC "-//IETF//DTD HTML 2.0//EN"',
242 'html PUBLIC "-//W3C//DTD HTML 3.2 Final//EN"']
243 for dtd in dtds:
244 self._run_check("<!DOCTYPE %s>" % dtd,
245 [('decl', 'DOCTYPE ' + dtd)])
246
Ezio Melotti36b73612012-02-21 09:22:16 +0200247 def test_slashes_in_starttag(self):
248 self._run_check('<a foo="var"/>', [('startendtag', 'a', [('foo', 'var')])])
249 html = ('<img width=902 height=250px '
250 'src="/sites/default/files/images/homepage/foo.jpg" '
251 '/*what am I doing here*/ />')
252 expected = [(
253 'startendtag', 'img',
254 [('width', '902'), ('height', '250px'),
255 ('src', '/sites/default/files/images/homepage/foo.jpg'),
256 ('*what', None), ('am', None), ('i', None),
257 ('doing', None), ('here*', None)]
258 )]
259 self._run_check(html, expected)
260 html = ('<a / /foo/ / /=/ / /bar/ / />'
261 '<a / /foo/ / /=/ / /bar/ / >')
262 expected = [
263 ('startendtag', 'a', [('foo', None), ('=', None), ('bar', None)]),
264 ('starttag', 'a', [('foo', None), ('=', None), ('bar', None)])
265 ]
266 self._run_check(html, expected)
Ezio Melottic45868e2012-04-18 19:08:41 -0600267 #see issue #14538
268 html = ('<meta><meta / ><meta // ><meta / / >'
269 '<meta/><meta /><meta //><meta//>')
270 expected = [
271 ('starttag', 'meta', []), ('starttag', 'meta', []),
272 ('starttag', 'meta', []), ('starttag', 'meta', []),
273 ('startendtag', 'meta', []), ('startendtag', 'meta', []),
274 ('startendtag', 'meta', []), ('startendtag', 'meta', []),
275 ]
276 self._run_check(html, expected)
Ezio Melotti36b73612012-02-21 09:22:16 +0200277
Fred Drake84bb9d82001-08-03 19:53:01 +0000278 def test_declaration_junk_chars(self):
Ezio Melotti4b92cc32012-02-13 16:10:44 +0200279 self._run_check("<!DOCTYPE foo $ >", [('decl', 'DOCTYPE foo $ ')])
Fred Drakebd3090d2001-05-18 15:32:59 +0000280
Fred Drake84bb9d82001-08-03 19:53:01 +0000281 def test_startendtag(self):
Fred Drakebd3090d2001-05-18 15:32:59 +0000282 self._run_check("<p/>", [
283 ("startendtag", "p", []),
284 ])
285 self._run_check("<p></p>", [
286 ("starttag", "p", []),
287 ("endtag", "p"),
288 ])
289 self._run_check("<p><img src='foo' /></p>", [
290 ("starttag", "p", []),
291 ("startendtag", "img", [("src", "foo")]),
292 ("endtag", "p"),
293 ])
294
Ezio Melottif1174432012-02-13 16:28:54 +0200295 def test_invalid_end_tags(self):
296 # A collection of broken end tags. <br> is used as separator.
297 # see http://www.w3.org/TR/html5/tokenization.html#end-tag-open-state
298 # and #13993
299 html = ('<br></label</p><br></div end tmAd-leaderBoard><br></<h4><br>'
300 '</li class="unit"><br></li\r\n\t\t\t\t\t\t</ul><br></><br>')
301 expected = [('starttag', 'br', []),
302 # < is part of the name, / is discarded, p is an attribute
303 ('endtag', 'label<'),
304 ('starttag', 'br', []),
305 # text and attributes are discarded
306 ('endtag', 'div'),
307 ('starttag', 'br', []),
308 # comment because the first char after </ is not a-zA-Z
309 ('comment', '<h4'),
310 ('starttag', 'br', []),
311 # attributes are discarded
312 ('endtag', 'li'),
313 ('starttag', 'br', []),
314 # everything till ul (included) is discarded
315 ('endtag', 'li'),
316 ('starttag', 'br', []),
317 # </> is ignored
318 ('starttag', 'br', [])]
319 self._run_check(html, expected)
320
321 def test_broken_invalid_end_tag(self):
322 # This is technically wrong (the "> shouldn't be included in the 'data')
323 # but is probably not worth fixing it (in addition to all the cases of
324 # the previous test, it would require a full attribute parsing).
325 # see #13993
326 html = '<b>This</b attr=">"> confuses the parser'
327 expected = [('starttag', 'b', []),
328 ('data', 'This'),
329 ('endtag', 'b'),
330 ('data', '"> confuses the parser')]
331 self._run_check(html, expected)
332
Fred Drake84bb9d82001-08-03 19:53:01 +0000333 def test_get_starttag_text(self):
Fred Drakebd3090d2001-05-18 15:32:59 +0000334 s = """<foo:bar \n one="1"\ttwo=2 >"""
335 self._run_check_extra(s, [
336 ("starttag", "foo:bar", [("one", "1"), ("two", "2")]),
337 ("starttag_text", s)])
338
Fred Drake84bb9d82001-08-03 19:53:01 +0000339 def test_cdata_content(self):
Ezio Melotti7e82b272011-11-01 14:09:56 +0200340 contents = [
341 '<!-- not a comment --> &not-an-entity-ref;',
342 "<not a='start tag'>",
343 '<a href="" /> <p> <span></span>',
344 'foo = "</scr" + "ipt>";',
345 'foo = "</SCRIPT" + ">";',
346 'foo = <\n/script> ',
347 '<!-- document.write("</scr" + "ipt>"); -->',
348 ('\n//<![CDATA[\n'
349 'document.write(\'<s\'+\'cript type="text/javascript" '
350 'src="http://www.example.org/r=\'+new '
351 'Date().getTime()+\'"><\\/s\'+\'cript>\');\n//]]>'),
352 '\n<!-- //\nvar foo = 3.14;\n// -->\n',
353 'foo = "</sty" + "le>";',
354 u'<!-- \u2603 -->',
355 # these two should be invalid according to the HTML 5 spec,
356 # section 8.1.2.2
357 #'foo = </\nscript>',
358 #'foo = </ script>',
359 ]
360 elements = ['script', 'style', 'SCRIPT', 'STYLE', 'Script', 'Style']
361 for content in contents:
362 for element in elements:
363 element_lower = element.lower()
364 s = u'<{element}>{content}</{element}>'.format(element=element,
365 content=content)
366 self._run_check(s, [("starttag", element_lower, []),
367 ("data", content),
368 ("endtag", element_lower)])
369
Ezio Melotti00dc60b2011-11-18 18:00:40 +0200370 def test_cdata_with_closing_tags(self):
371 # see issue #13358
372 # make sure that HTMLParser calls handle_data only once for each CDATA.
373 # The normal event collector normalizes the events in get_events,
374 # so we override it to return the original list of events.
375 class Collector(EventCollector):
376 def get_events(self):
377 return self.events
378
379 content = """<!-- not a comment --> &not-an-entity-ref;
380 <a href="" /> </p><p> &amp; <span></span></style>
381 '</script' + '>' </html> </head> </scripter>!"""
382 for element in [' script', 'script ', ' script ',
383 '\nscript', 'script\n', '\nscript\n']:
384 s = u'<script>{content}</{element}>'.format(element=element,
385 content=content)
386 self._run_check(s, [("starttag", "script", []),
387 ("data", content),
388 ("endtag", "script")],
389 collector=Collector)
390
Victor Stinner554a3b82010-05-24 21:33:24 +0000391 def test_malformatted_charref(self):
392 self._run_check("<p>&#bad;</p>", [
393 ("starttag", "p", []),
394 ("data", "&#bad;"),
395 ("endtag", "p"),
396 ])
Ezio Melotti5a888532014-02-01 21:20:22 +0200397 # add the [] as a workaround to avoid buffering (see #20288)
398 self._run_check(["<div>&#bad;</div>"], [
399 ("starttag", "div", []),
400 ("data", "&#bad;"),
401 ("endtag", "div"),
402 ])
Victor Stinner554a3b82010-05-24 21:33:24 +0000403
Senthil Kumaran3f60f092010-12-28 16:05:07 +0000404 def test_unescape_function(self):
405 parser = HTMLParser.HTMLParser()
406 self.assertEqual(parser.unescape('&#bad;'),'&#bad;')
407 self.assertEqual(parser.unescape('&#0038;'),'&')
408
Fred Drakebd3090d2001-05-18 15:32:59 +0000409
Ezio Melotti74592912011-11-08 02:07:18 +0200410
411class AttributesTestCase(TestCaseBase):
412
413 def test_attr_syntax(self):
414 output = [
415 ("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", None)])
416 ]
417 self._run_check("""<a b='v' c="v" d=v e>""", output)
418 self._run_check("""<a b = 'v' c = "v" d = v e>""", output)
419 self._run_check("""<a\nb\n=\n'v'\nc\n=\n"v"\nd\n=\nv\ne>""", output)
420 self._run_check("""<a\tb\t=\t'v'\tc\t=\t"v"\td\t=\tv\te>""", output)
421
422 def test_attr_values(self):
423 self._run_check("""<a b='xxx\n\txxx' c="yyy\t\nyyy" d='\txyz\n'>""",
424 [("starttag", "a", [("b", "xxx\n\txxx"),
425 ("c", "yyy\t\nyyy"),
426 ("d", "\txyz\n")])])
427 self._run_check("""<a b='' c="">""",
428 [("starttag", "a", [("b", ""), ("c", "")])])
429 # Regression test for SF patch #669683.
430 self._run_check("<e a=rgb(1,2,3)>",
431 [("starttag", "e", [("a", "rgb(1,2,3)")])])
432 # Regression test for SF bug #921657.
433 self._run_check(
434 "<a href=mailto:xyz@example.com>",
435 [("starttag", "a", [("href", "mailto:xyz@example.com")])])
436
437 def test_attr_nonascii(self):
438 # see issue 7311
439 self._run_check(
440 u"<img src=/foo/bar.png alt=\u4e2d\u6587>",
441 [("starttag", "img", [("src", "/foo/bar.png"),
442 ("alt", u"\u4e2d\u6587")])])
443 self._run_check(
444 u"<a title='\u30c6\u30b9\u30c8' href='\u30c6\u30b9\u30c8.html'>",
445 [("starttag", "a", [("title", u"\u30c6\u30b9\u30c8"),
446 ("href", u"\u30c6\u30b9\u30c8.html")])])
447 self._run_check(
448 u'<a title="\u30c6\u30b9\u30c8" href="\u30c6\u30b9\u30c8.html">',
449 [("starttag", "a", [("title", u"\u30c6\u30b9\u30c8"),
450 ("href", u"\u30c6\u30b9\u30c8.html")])])
451
452 def test_attr_entity_replacement(self):
453 self._run_check(
454 "<a b='&amp;&gt;&lt;&quot;&apos;'>",
455 [("starttag", "a", [("b", "&><\"'")])])
456
457 def test_attr_funky_names(self):
458 self._run_check(
459 "<a a.b='v' c:d=v e-f=v>",
460 [("starttag", "a", [("a.b", "v"), ("c:d", "v"), ("e-f", "v")])])
Ezio Melotti0f1571c2011-11-14 18:04:05 +0200461 self._run_check(
462 "<a $><b $=%><c \=/>",
463 [("starttag", "a", [("$", None)]),
464 ("starttag", "b", [("$", "%")]),
465 ("starttag", "c", [("\\", "/")])])
Ezio Melotti74592912011-11-08 02:07:18 +0200466
467 def test_entityrefs_in_attributes(self):
468 self._run_check(
469 "<html foo='&euro;&amp;&#97;&#x61;&unsupported;'>",
470 [("starttag", "html", [("foo", u"\u20AC&aa&unsupported;")])])
471
Ezio Melotti0f1571c2011-11-14 18:04:05 +0200472 def test_entities_in_attribute_value(self):
473 # see #1200313
474 for entity in ['&', '&amp;', '&#38;', '&#x26;']:
475 self._run_check('<a href="%s">' % entity,
476 [("starttag", "a", [("href", "&")])])
477 self._run_check("<a href='%s'>" % entity,
478 [("starttag", "a", [("href", "&")])])
479 self._run_check("<a href=%s>" % entity,
480 [("starttag", "a", [("href", "&")])])
481
482 def test_malformed_attributes(self):
483 # see #13357
484 html = (
485 "<a href=test'style='color:red;bad1'>test - bad1</a>"
486 "<a href=test'+style='color:red;ba2'>test - bad2</a>"
487 "<a href=test'&nbsp;style='color:red;bad3'>test - bad3</a>"
488 "<a href = test'&nbsp;style='color:red;bad4' >test - bad4</a>"
489 )
490 expected = [
491 ('starttag', 'a', [('href', "test'style='color:red;bad1'")]),
492 ('data', 'test - bad1'), ('endtag', 'a'),
493 ('starttag', 'a', [('href', "test'+style='color:red;ba2'")]),
494 ('data', 'test - bad2'), ('endtag', 'a'),
495 ('starttag', 'a', [('href', u"test'\xa0style='color:red;bad3'")]),
496 ('data', 'test - bad3'), ('endtag', 'a'),
497 ('starttag', 'a', [('href', u"test'\xa0style='color:red;bad4'")]),
498 ('data', 'test - bad4'), ('endtag', 'a')
499 ]
500 self._run_check(html, expected)
501
502 def test_malformed_adjacent_attributes(self):
503 # see #12629
504 self._run_check('<x><y z=""o"" /></x>',
505 [('starttag', 'x', []),
506 ('startendtag', 'y', [('z', ''), ('o""', None)]),
507 ('endtag', 'x')])
508 self._run_check('<x><y z="""" /></x>',
509 [('starttag', 'x', []),
510 ('startendtag', 'y', [('z', ''), ('""', None)]),
511 ('endtag', 'x')])
512
513 # see #755670 for the following 3 tests
514 def test_adjacent_attributes(self):
515 self._run_check('<a width="100%"cellspacing=0>',
516 [("starttag", "a",
517 [("width", "100%"), ("cellspacing","0")])])
518
519 self._run_check('<a id="foo"class="bar">',
520 [("starttag", "a",
521 [("id", "foo"), ("class","bar")])])
522
523 def test_missing_attribute_value(self):
524 self._run_check('<a v=>',
525 [("starttag", "a", [("v", "")])])
526
527 def test_javascript_attribute_value(self):
528 self._run_check("<a href=javascript:popup('/popup/help.html')>",
529 [("starttag", "a",
530 [("href", "javascript:popup('/popup/help.html')")])])
531
532 def test_end_tag_in_attribute_value(self):
533 # see #1745761
534 self._run_check("<a href='http://www.example.org/\">;'>spam</a>",
535 [("starttag", "a",
536 [("href", "http://www.example.org/\">;")]),
537 ("data", "spam"), ("endtag", "a")])
538
Ezio Melotti4b92cc32012-02-13 16:10:44 +0200539 def test_comments(self):
540 html = ("<!-- I'm a valid comment -->"
541 '<!--me too!-->'
542 '<!------>'
543 '<!---->'
544 '<!----I have many hyphens---->'
545 '<!-- I have a > in the middle -->'
546 '<!-- and I have -- in the middle! -->')
547 expected = [('comment', " I'm a valid comment "),
548 ('comment', 'me too!'),
549 ('comment', '--'),
550 ('comment', ''),
551 ('comment', '--I have many hyphens--'),
552 ('comment', ' I have a > in the middle '),
553 ('comment', ' and I have -- in the middle! ')]
554 self._run_check(html, expected)
555
556 def test_broken_comments(self):
557 html = ('<! not really a comment >'
558 '<! not a comment either -->'
559 '<! -- close enough -->'
560 '<!><!<-- this was an empty comment>'
561 '<!!! another bogus comment !!!>')
562 expected = [
563 ('comment', ' not really a comment '),
564 ('comment', ' not a comment either --'),
565 ('comment', ' -- close enough --'),
566 ('comment', ''),
567 ('comment', '<-- this was an empty comment'),
568 ('comment', '!! another bogus comment !!!'),
569 ]
570 self._run_check(html, expected)
571
Ezio Melotti6b7003a2011-12-19 07:28:08 +0200572 def test_condcoms(self):
573 html = ('<!--[if IE & !(lte IE 8)]>aren\'t<![endif]-->'
574 '<!--[if IE 8]>condcoms<![endif]-->'
575 '<!--[if lte IE 7]>pretty?<![endif]-->')
576 expected = [('comment', "[if IE & !(lte IE 8)]>aren't<![endif]"),
577 ('comment', '[if IE 8]>condcoms<![endif]'),
578 ('comment', '[if lte IE 7]>pretty?<![endif]')]
579 self._run_check(html, expected)
580
581 def test_broken_condcoms(self):
582 # these condcoms are missing the '--' after '<!' and before the '>'
583 html = ('<![if !(IE)]>broken condcom<![endif]>'
584 '<![if ! IE]><link href="favicon.tiff"/><![endif]>'
585 '<![if !IE 6]><img src="firefox.png" /><![endif]>'
586 '<![if !ie 6]><b>foo</b><![endif]>'
587 '<![if (!IE)|(lt IE 9)]><img src="mammoth.bmp" /><![endif]>')
588 # According to the HTML5 specs sections "8.2.4.44 Bogus comment state"
589 # and "8.2.4.45 Markup declaration open state", comment tokens should
590 # be emitted instead of 'unknown decl', but calling unknown_decl
591 # provides more flexibility.
592 # See also Lib/_markupbase.py:parse_declaration
593 expected = [
594 ('unknown decl', 'if !(IE)'),
595 ('data', 'broken condcom'),
596 ('unknown decl', 'endif'),
597 ('unknown decl', 'if ! IE'),
598 ('startendtag', 'link', [('href', 'favicon.tiff')]),
599 ('unknown decl', 'endif'),
600 ('unknown decl', 'if !IE 6'),
601 ('startendtag', 'img', [('src', 'firefox.png')]),
602 ('unknown decl', 'endif'),
603 ('unknown decl', 'if !ie 6'),
604 ('starttag', 'b', []),
605 ('data', 'foo'),
606 ('endtag', 'b'),
607 ('unknown decl', 'endif'),
608 ('unknown decl', 'if (!IE)|(lt IE 9)'),
609 ('startendtag', 'img', [('src', 'mammoth.bmp')]),
610 ('unknown decl', 'endif')
611 ]
612 self._run_check(html, expected)
613
Ezio Melotti0f1571c2011-11-14 18:04:05 +0200614
Fred Drakee8220492001-09-24 20:19:08 +0000615def test_main():
Ezio Melotti74592912011-11-08 02:07:18 +0200616 test_support.run_unittest(HTMLParserTestCase, AttributesTestCase)
Fred Drakee8220492001-09-24 20:19:08 +0000617
618
619if __name__ == "__main__":
620 test_main()