Python code coverage for Lib/test/test_sgmllib.py

#	count	content
1	1	import pprint
2	1	import re
3	1	import unittest
4	1	from test import test_support
5	1	sgmllib = test_support.import_module('sgmllib', deprecated=True)
6	n/a
7	n/a
8	2	class EventCollector(sgmllib.SGMLParser):
9	n/a
10	1	def __init__(self):
11	31	self.events = []
12	31	self.append = self.events.append
13	31	sgmllib.SGMLParser.__init__(self)
14	n/a
15	1	def get_events(self):
16	n/a	# Normalize the list of events so that buffer artefacts don't
17	n/a	# separate runs of contiguous characters.
18	30	L = []
19	30	prevtype = None
20	241	for event in self.events:
21	211	type = event[0]
22	211	if type == prevtype == "data":
23	151	L[-1] = ("data", L[-1][1] + event[1])
24	n/a	else:
25	60	L.append(event)
26	211	prevtype = type
27	30	self.events = L
28	30	return L
29	n/a
30	n/a	# structure markup
31	n/a
32	1	def unknown_starttag(self, tag, attrs):
33	26	self.append(("starttag", tag, attrs))
34	n/a
35	1	def unknown_endtag(self, tag):
36	9	self.append(("endtag", tag))
37	n/a
38	n/a	# all other markup
39	n/a
40	1	def handle_comment(self, data):
41	1	self.append(("comment", data))
42	n/a
43	1	def handle_charref(self, data):
44	0	self.append(("charref", data))
45	n/a
46	1	def handle_data(self, data):
47	161	self.append(("data", data))
48	n/a
49	1	def handle_decl(self, decl):
50	3	self.append(("decl", decl))
51	n/a
52	1	def handle_entityref(self, data):
53	0	self.append(("entityref", data))
54	n/a
55	1	def handle_pi(self, data):
56	1	self.append(("pi", data))
57	n/a
58	1	def unknown_decl(self, decl):
59	1	self.append(("unknown decl", decl))
60	n/a
61	n/a
62	2	class CDATAEventCollector(EventCollector):
63	1	def start_cdata(self, attrs):
64	2	self.append(("starttag", "cdata", attrs))
65	2	self.setliteral()
66	n/a
67	n/a
68	2	class HTMLEntityCollector(EventCollector):
69	n/a
70	1	entity_or_charref = re.compile('(?:&([a-zA-Z][-.a-zA-Z0-9]*)'
71	n/a	'\|&#(x[0-9a-zA-Z]+\|[0-9]+))(;?)')
72	n/a
73	1	def convert_charref(self, name):
74	2	self.append(("charref", "convert", name))
75	2	if name[0] != "x":
76	1	return EventCollector.convert_charref(self, name)
77	n/a
78	1	def convert_codepoint(self, codepoint):
79	1	self.append(("codepoint", "convert", codepoint))
80	1	EventCollector.convert_codepoint(self, codepoint)
81	n/a
82	1	def convert_entityref(self, name):
83	2	self.append(("entityref", "convert", name))
84	2	return EventCollector.convert_entityref(self, name)
85	n/a
86	n/a	# These to record that they were called, then pass the call along
87	n/a	# to the default implementation so that it's actions can be
88	n/a	# recorded.
89	n/a
90	1	def handle_charref(self, data):
91	1	self.append(("charref", data))
92	1	sgmllib.SGMLParser.handle_charref(self, data)
93	n/a
94	1	def handle_entityref(self, data):
95	1	self.append(("entityref", data))
96	1	sgmllib.SGMLParser.handle_entityref(self, data)
97	n/a
98	n/a
99	2	class SGMLParserTestCase(unittest.TestCase):
100	n/a
101	1	collector = EventCollector
102	n/a
103	1	def get_events(self, source):
104	30	parser = self.collector()
105	30	try:
106	1211	for s in source:
107	1181	parser.feed(s)
108	30	parser.close()
109	0	except:
110	n/a	#self.events = parser.events
111	0	raise
112	30	return parser.get_events()
113	n/a
114	1	def check_events(self, source, expected_events):
115	30	try:
116	30	events = self.get_events(source)
117	0	except:
118	n/a	#import sys
119	n/a	#print >>sys.stderr, pprint.pformat(self.events)
120	0	raise
121	30	if events != expected_events:
122	0	self.fail("received events did not match expected events\n"
123	n/a	"Expected:\n" + pprint.pformat(expected_events) +
124	0	"\nReceived:\n" + pprint.pformat(events))
125	n/a
126	1	def check_parse_error(self, source):
127	1	parser = EventCollector()
128	1	try:
129	1	parser.feed(source)
130	0	parser.close()
131	1	except sgmllib.SGMLParseError:
132	1	pass
133	n/a	else:
134	0	self.fail("expected SGMLParseError for %r\nReceived:\n%s"
135	0	% (source, pprint.pformat(parser.get_events())))
136	n/a
137	1	def test_doctype_decl_internal(self):
138	n/a	inside = """\
139	n/a	DOCTYPE html PUBLIC '-//W3C//DTD HTML 4.01//EN'
140	n/a	SYSTEM 'http://www.w3.org/TR/html401/strict.dtd' [
141	n/a	<!ELEMENT html - O EMPTY>
142	n/a	<!ATTLIST html
143	n/a	version CDATA #IMPLIED
144	n/a	profile CDATA 'DublinCore'>
145	n/a	<!NOTATION datatype SYSTEM 'http://xml.python.org/notations/python-module'>
146	n/a	<!ENTITY myEntity 'internal parsed entity'>
147	n/a	<!ENTITY anEntity SYSTEM 'http://xml.python.org/entities/something.xml'>
148	n/a	<!ENTITY % paramEntity 'name\|name\|name'>
149	n/a	%paramEntity;
150	n/a	<!-- comment -->
151	1	]"""
152	1	self.check_events(["<!%s>" % inside], [
153	1	("decl", inside),
154	n/a	])
155	n/a
156	1	def test_doctype_decl_external(self):
157	1	inside = "DOCTYPE html PUBLIC '-//W3C//DTD HTML 4.01//EN'"
158	1	self.check_events("<!%s>" % inside, [
159	1	("decl", inside),
160	n/a	])
161	n/a
162	1	def test_underscore_in_attrname(self):
163	n/a	# SF bug #436621
164	n/a	"""Make sure attribute names with underscores are accepted"""
165	1	self.check_events("<a has_under _under>", [
166	1	("starttag", "a", [("has_under", "has_under"),
167	1	("_under", "_under")]),
168	n/a	])
169	n/a
170	1	def test_underscore_in_tagname(self):
171	n/a	# SF bug #436621
172	n/a	"""Make sure tag names with underscores are accepted"""
173	1	self.check_events("<has_under></has_under>", [
174	1	("starttag", "has_under", []),
175	1	("endtag", "has_under"),
176	n/a	])
177	n/a
178	1	def test_quotes_in_unquoted_attrs(self):
179	n/a	# SF bug #436621
180	n/a	"""Be sure quotes in unquoted attributes are made part of the value"""
181	1	self.check_events("<a href=foo'bar\"baz>", [
182	1	("starttag", "a", [("href", "foo'bar\"baz")]),
183	n/a	])
184	n/a
185	1	def test_xhtml_empty_tag(self):
186	n/a	"""Handling of XHTML-style empty start tags"""
187	1	self.check_events("<br />text<i></i>", [
188	1	("starttag", "br", []),
189	1	("data", "text"),
190	1	("starttag", "i", []),
191	1	("endtag", "i"),
192	n/a	])
193	n/a
194	1	def test_processing_instruction_only(self):
195	1	self.check_events("<?processing instruction>", [
196	1	("pi", "processing instruction"),
197	n/a	])
198	n/a
199	1	def test_bad_nesting(self):
200	1	self.check_events("<a><b></a></b>", [
201	1	("starttag", "a", []),
202	1	("starttag", "b", []),
203	1	("endtag", "a"),
204	1	("endtag", "b"),
205	n/a	])
206	n/a
207	1	def test_bare_ampersands(self):
208	1	self.check_events("this text & contains & ampersands &", [
209	1	("data", "this text & contains & ampersands &"),
210	n/a	])
211	n/a
212	1	def test_bare_pointy_brackets(self):
213	1	self.check_events("this < text > contains < bare>pointy< brackets", [
214	1	("data", "this < text > contains < bare>pointy< brackets"),
215	n/a	])
216	n/a
217	1	def test_attr_syntax(self):
218	n/a	output = [
219	1	("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", "e")])
220	n/a	]
221	1	self.check_events("""<a b='v' c="v" d=v e>""", output)
222	1	self.check_events("""<a b = 'v' c = "v" d = v e>""", output)
223	1	self.check_events("""<a\nb\n=\n'v'\nc\n=\n"v"\nd\n=\nv\ne>""", output)
224	1	self.check_events("""<a\tb\t=\t'v'\tc\t=\t"v"\td\t=\tv\te>""", output)
225	n/a
226	1	def test_attr_values(self):
227	1	self.check_events("""<a b='xxx\n\txxx' c="yyy\t\nyyy" d='\txyz\n'>""",
228	1	[("starttag", "a", [("b", "xxx\n\txxx"),
229	1	("c", "yyy\t\nyyy"),
230	1	("d", "\txyz\n")])
231	n/a	])
232	1	self.check_events("""<a b='' c="">""", [
233	1	("starttag", "a", [("b", ""), ("c", "")]),
234	n/a	])
235	n/a	# URL construction stuff from RFC 1808:
236	1	safe = "$-_.+"
237	1	extra = "!*'(),"
238	1	reserved = ";/?:@&="
239	1	url = "http://example.com:8080/path/to/file?%s%s%s" % (
240	1	safe, extra, reserved)
241	1	self.check_events("""<e a=%s>""" % url, [
242	1	("starttag", "e", [("a", url)]),
243	n/a	])
244	n/a	# Regression test for SF patch #669683.
245	1	self.check_events("<e a=rgb(1,2,3)>", [
246	1	("starttag", "e", [("a", "rgb(1,2,3)")]),
247	n/a	])
248	n/a
249	1	def test_attr_values_entities(self):
250	n/a	"""Substitution of entities and charrefs in attribute values"""
251	n/a	# SF bug #1452246
252	1	self.check_events("""<a b=< c=<> d=&lt-> e='< '
253	n/a	f="&xxx;" g=' !' h='Ǵ'
254	n/a	i='x?a=b&c=d;'
255	1	j='&#42;' k='&#42;'>""",
256	1	[("starttag", "a", [("b", "<"),
257	1	("c", "<>"),
258	1	("d", "&lt->"),
259	1	("e", "< "),
260	1	("f", "&xxx;"),
261	1	("g", " !"),
262	1	("h", "Ǵ"),
263	1	("i", "x?a=b&c=d;"),
264	1	("j", "*"),
265	1	("k", "*"),
266	n/a	])])
267	n/a
268	1	def test_convert_overrides(self):
269	n/a	# This checks that the character and entity reference
270	n/a	# conversion helpers are called at the documented times. No
271	n/a	# attempt is made to really change what the parser accepts.
272	n/a	#
273	1	self.collector = HTMLEntityCollector
274	1	self.check_events(('<a title="“test”">foo</a>'
275	n/a	'&foobar;*'), [
276	1	('entityref', 'convert', 'ldquo'),
277	1	('charref', 'convert', 'x201d'),
278	1	('starttag', 'a', [('title', '“test”')]),
279	1	('data', 'foo'),
280	1	('endtag', 'a'),
281	1	('entityref', 'foobar'),
282	1	('entityref', 'convert', 'foobar'),
283	1	('charref', '42'),
284	1	('charref', 'convert', '42'),
285	1	('codepoint', 'convert', 42),
286	n/a	])
287	n/a
288	1	def test_attr_funky_names(self):
289	1	self.check_events("""<a a.b='v' c:d=v e-f=v>""", [
290	1	("starttag", "a", [("a.b", "v"), ("c:d", "v"), ("e-f", "v")]),
291	n/a	])
292	n/a
293	1	def test_attr_value_ip6_url(self):
294	n/a	# http://www.python.org/sf/853506
295	1	self.check_events(("<a href='http://[1080::8:800:200C:417A]/'>"
296	n/a	"<a href=http://[1080::8:800:200C:417A]/>"), [
297	1	("starttag", "a", [("href", "http://[1080::8:800:200C:417A]/")]),
298	1	("starttag", "a", [("href", "http://[1080::8:800:200C:417A]/")]),
299	n/a	])
300	n/a
301	1	def test_weird_starttags(self):
302	1	self.check_events("<a<a>", [
303	1	("starttag", "a", []),
304	1	("starttag", "a", []),
305	n/a	])
306	1	self.check_events("</a<a>", [
307	1	("endtag", "a"),
308	1	("starttag", "a", []),
309	n/a	])
310	n/a
311	1	def test_declaration_junk_chars(self):
312	1	self.check_parse_error("<!DOCTYPE foo $ >")
313	n/a
314	1	def test_get_starttag_text(self):
315	1	s = """<foobar \n one="1"\ttwo=2 >"""
316	1	self.check_events(s, [
317	1	("starttag", "foobar", [("one", "1"), ("two", "2")]),
318	n/a	])
319	n/a
320	1	def test_cdata_content(self):
321	1	s = ("<cdata> <!-- not a comment --> &not-an-entity-ref; </cdata>"
322	n/a	"<notcdata> <!-- comment --> </notcdata>")
323	1	self.collector = CDATAEventCollector
324	1	self.check_events(s, [
325	1	("starttag", "cdata", []),
326	1	("data", " <!-- not a comment --> &not-an-entity-ref; "),
327	1	("endtag", "cdata"),
328	1	("starttag", "notcdata", []),
329	1	("data", " "),
330	1	("comment", " comment "),
331	1	("data", " "),
332	1	("endtag", "notcdata"),
333	n/a	])
334	1	s = """<cdata> <not a='start tag'> </cdata>"""
335	1	self.check_events(s, [
336	1	("starttag", "cdata", []),
337	1	("data", " <not a='start tag'> "),
338	1	("endtag", "cdata"),
339	n/a	])
340	n/a
341	1	def test_illegal_declarations(self):
342	1	s = 'abc<!spacer type="block" height="25">def'
343	1	self.check_events(s, [
344	1	("data", "abc"),
345	1	("unknown decl", 'spacer type="block" height="25"'),
346	1	("data", "def"),
347	n/a	])
348	n/a
349	1	def test_enumerated_attr_type(self):
350	1	s = "<!DOCTYPE doc [<!ATTLIST doc attr (a \| b) >]>"
351	1	self.check_events(s, [
352	1	('decl', 'DOCTYPE doc [<!ATTLIST doc attr (a \| b) >]'),
353	n/a	])
354	n/a
355	1	def test_read_chunks(self):
356	n/a	# SF bug #1541697, this caused sgml parser to hang
357	n/a	# Just verify this code doesn't cause a hang.
358	1	CHUNK = 1024 # increasing this to 8212 makes the problem go away
359	n/a
360	1	f = open(test_support.findfile('sgml_input.html'))
361	1	fp = sgmllib.SGMLParser()
362	1	while 1:
363	9	data = f.read(CHUNK)
364	9	fp.feed(data)
365	9	if len(data) != CHUNK:
366	1	break
367	n/a
368	1	def test_only_decode_ascii(self):
369	n/a	# SF bug #1651995, make sure non-ascii character references are not decoded
370	1	s = '<signs exclamation="&#33" copyright="&#169" quoteleft="‘">'
371	1	self.check_events(s, [
372	1	('starttag', 'signs',
373	1	[('exclamation', '!'), ('copyright', '&#169'),
374	1	('quoteleft', '‘')]),
375	n/a	])
376	n/a
377	n/a	# XXX These tests have been disabled by prefixing their names with
378	n/a	# an underscore. The first two exercise outstanding bugs in the
379	n/a	# sgmllib module, and the third exhibits questionable behavior
380	n/a	# that needs to be carefully considered before changing it.
381	n/a
382	1	def _test_starttag_end_boundary(self):
383	0	self.check_events("<a b='<'>", [("starttag", "a", [("b", "<")])])
384	0	self.check_events("<a b='>'>", [("starttag", "a", [("b", ">")])])
385	n/a
386	1	def _test_buffer_artefacts(self):
387	0	output = [("starttag", "a", [("b", "<")])]
388	0	self.check_events(["<a b='<'>"], output)
389	0	self.check_events(["<a ", "b='<'>"], output)
390	0	self.check_events(["<a b", "='<'>"], output)
391	0	self.check_events(["<a b=", "'<'>"], output)
392	0	self.check_events(["<a b='<", "'>"], output)
393	0	self.check_events(["<a b='<'", ">"], output)
394	n/a
395	0	output = [("starttag", "a", [("b", ">")])]
396	0	self.check_events(["<a b='>'>"], output)
397	0	self.check_events(["<a ", "b='>'>"], output)
398	0	self.check_events(["<a b", "='>'>"], output)
399	0	self.check_events(["<a b=", "'>'>"], output)
400	0	self.check_events(["<a b='>", "'>"], output)
401	0	self.check_events(["<a b='>'", ">"], output)
402	n/a
403	0	output = [("comment", "abc")]
404	0	self.check_events(["", "<!--abc-->"], output)
405	0	self.check_events(["<", "!--abc-->"], output)
406	0	self.check_events(["<!", "--abc-->"], output)
407	0	self.check_events(["<!-", "-abc-->"], output)
408	0	self.check_events(["<!--", "abc-->"], output)
409	0	self.check_events(["<!--a", "bc-->"], output)
410	0	self.check_events(["<!--ab", "c-->"], output)
411	0	self.check_events(["<!--abc", "-->"], output)
412	0	self.check_events(["<!--abc-", "->"], output)
413	0	self.check_events(["<!--abc--", ">"], output)
414	0	self.check_events(["<!--abc-->", ""], output)
415	n/a
416	1	def _test_starttag_junk_chars(self):
417	0	self.check_parse_error("<")
418	0	self.check_parse_error("<>")
419	0	self.check_parse_error("</$>")
420	0	self.check_parse_error("</")
421	0	self.check_parse_error("</a")
422	0	self.check_parse_error("<$")
423	0	self.check_parse_error("<$>")
424	0	self.check_parse_error("<!")
425	0	self.check_parse_error("<a $>")
426	0	self.check_parse_error("<a")
427	0	self.check_parse_error("<a foo='bar'")
428	0	self.check_parse_error("<a foo='bar")
429	0	self.check_parse_error("<a foo='>'")
430	0	self.check_parse_error("<a foo='>")
431	0	self.check_parse_error("<a foo=>")
432	n/a
433	n/a
434	1	def test_main():
435	1	test_support.run_unittest(SGMLParserTestCase)
436	n/a
437	n/a
438	1	if __name__ == "__main__":
439	0	test_main()