[OpenSPARC-T2-SAM] / sam-t2 / devtools / v9 / lib / python2.4 / test / test_sgmllib.py

import pprint
import sgmllib
import unittest
from test import test_support


class EventCollector(sgmllib.SGMLParser):

    def __init__(self):
        self.events = []
        self.append = self.events.append
        sgmllib.SGMLParser.__init__(self)

    def get_events(self):
        # Normalize the list of events so that buffer artefacts don't
        # separate runs of contiguous characters.
        L = []
        prevtype = None
        for event in self.events:
            type = event[0]
            if type == prevtype == "data":
                L[-1] = ("data", L[-1][1] + event[1])
            else:
                L.append(event)
            prevtype = type
        self.events = L
        return L

    # structure markup

    def unknown_starttag(self, tag, attrs):
        self.append(("starttag", tag, attrs))

    def unknown_endtag(self, tag):
        self.append(("endtag", tag))

    # all other markup

    def handle_comment(self, data):
        self.append(("comment", data))

    def handle_charref(self, data):
        self.append(("charref", data))

    def handle_data(self, data):
        self.append(("data", data))

    def handle_decl(self, decl):
        self.append(("decl", decl))

    def handle_entityref(self, data):
        self.append(("entityref", data))

    def handle_pi(self, data):
        self.append(("pi", data))

    def unknown_decl(self, decl):
        self.append(("unknown decl", decl))


class CDATAEventCollector(EventCollector):
    def start_cdata(self, attrs):
        self.append(("starttag", "cdata", attrs))
        self.setliteral()


class SGMLParserTestCase(unittest.TestCase):

    collector = EventCollector

    def get_events(self, source):
        parser = self.collector()
        try:
            for s in source:
                parser.feed(s)
            parser.close()
        except:
            #self.events = parser.events
            raise
        return parser.get_events()

    def check_events(self, source, expected_events):
        try:
            events = self.get_events(source)
        except:
            import sys
            #print >>sys.stderr, pprint.pformat(self.events)
            raise
        if events != expected_events:
            self.fail("received events did not match expected events\n"
                      "Expected:\n" + pprint.pformat(expected_events) +
                      "\nReceived:\n" + pprint.pformat(events))

    def check_parse_error(self, source):
        parser = EventCollector()
        try:
            parser.feed(source)
            parser.close()
        except sgmllib.SGMLParseError:
            pass
        else:
            self.fail("expected SGMLParseError for %r\nReceived:\n%s"
                      % (source, pprint.pformat(parser.get_events())))

    def test_doctype_decl_internal(self):
        inside = """\
DOCTYPE html PUBLIC '-//W3C//DTD HTML 4.01//EN'
             SYSTEM 'http://www.w3.org/TR/html401/strict.dtd' [
  <!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.check_events(["<!%s>" % inside], [
            ("decl", inside),
            ])

    def test_doctype_decl_external(self):
        inside = "DOCTYPE html PUBLIC '-//W3C//DTD HTML 4.01//EN'"
        self.check_events("<!%s>" % inside, [
            ("decl", inside),
            ])

    def test_underscore_in_attrname(self):
        # SF bug #436621
        """Make sure attribute names with underscores are accepted"""
        self.check_events("<a has_under _under>", [
            ("starttag", "a", [("has_under", "has_under"),
                               ("_under", "_under")]),
            ])

    def test_underscore_in_tagname(self):
        # SF bug #436621
        """Make sure tag names with underscores are accepted"""
        self.check_events("<has_under></has_under>", [
            ("starttag", "has_under", []),
            ("endtag", "has_under"),
            ])

    def test_quotes_in_unquoted_attrs(self):
        # SF bug #436621
        """Be sure quotes in unquoted attributes are made part of the value"""
        self.check_events("<a href=foo'bar\"baz>", [
            ("starttag", "a", [("href", "foo'bar\"baz")]),
            ])

    def test_xhtml_empty_tag(self):
        """Handling of XHTML-style empty start tags"""
        self.check_events("<br />text<i></i>", [
            ("starttag", "br", []),
            ("data", "text"),
            ("starttag", "i", []),
            ("endtag", "i"),
            ])

    def test_processing_instruction_only(self):
        self.check_events("<?processing instruction>", [
            ("pi", "processing instruction"),
            ])

    def test_bad_nesting(self):
        self.check_events("<a><b></a></b>", [
            ("starttag", "a", []),
            ("starttag", "b", []),
            ("endtag", "a"),
            ("endtag", "b"),
            ])

    def test_bare_ampersands(self):
        self.check_events("this text & contains & ampersands &", [
            ("data", "this text & contains & ampersands &"),
            ])

    def test_bare_pointy_brackets(self):
        self.check_events("this < text > contains < bare>pointy< brackets", [
            ("data", "this < text > contains < bare>pointy< brackets"),
            ])

    def test_attr_syntax(self):
        output = [
          ("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", "e")])
          ]
        self.check_events("""<a b='v' c="v" d=v e>""", output)
        self.check_events("""<a  b = 'v' c = "v" d = v e>""", output)
        self.check_events("""<a\nb\n=\n'v'\nc\n=\n"v"\nd\n=\nv\ne>""", output)
        self.check_events("""<a\tb\t=\t'v'\tc\t=\t"v"\td\t=\tv\te>""", output)

    def test_attr_values(self):
        self.check_events("""<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.check_events("""<a b='' c="">""", [
            ("starttag", "a", [("b", ""), ("c", "")]),
            ])
        # URL construction stuff from RFC 1808:
        safe = "$-_.+"
        extra = "!*'(),"
        reserved = ";/?:@&="
        url = "http://example.com:8080/path/to/file?%s%s%s" % (
            safe, extra, reserved)
        self.check_events("""<e a=%s>""" % url, [
            ("starttag", "e", [("a", url)]),
            ])
        # Regression test for SF patch #669683.
        self.check_events("<e a=rgb(1,2,3)>", [
            ("starttag", "e", [("a", "rgb(1,2,3)")]),
            ])

    def test_attr_funky_names(self):
        self.check_events("""<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):
        s = 'abc<!spacer type="block" height="25">def'
        self.check_events(s, [
            ("data", "abc"),
            ("unknown decl", 'spacer type="block" height="25"'),
            ("data", "def"),
            ])

    def test_weird_starttags(self):
        self.check_events("<a<a>", [
            ("starttag", "a", []),
            ("starttag", "a", []),
            ])
        self.check_events("</a<a>", [
            ("endtag", "a"),
            ("starttag", "a", []),
            ])

    def test_declaration_junk_chars(self):
        self.check_parse_error("<!DOCTYPE foo $ >")

    def test_get_starttag_text(self):
        s = """<foobar   \n   one="1"\ttwo=2   >"""
        self.check_events(s, [
            ("starttag", "foobar", [("one", "1"), ("two", "2")]),
            ])

    def test_cdata_content(self):
        s = ("<cdata> <!-- not a comment --> &not-an-entity-ref; </cdata>"
             "<notcdata> <!-- comment --> </notcdata>")
        self.collector = CDATAEventCollector
        self.check_events(s, [
            ("starttag", "cdata", []),
            ("data", " <!-- not a comment --> &not-an-entity-ref; "),
            ("endtag", "cdata"),
            ("starttag", "notcdata", []),
            ("data", " "),
            ("comment", " comment "),
            ("data", " "),
            ("endtag", "notcdata"),
            ])
        s = """<cdata> <not a='start tag'> </cdata>"""
        self.check_events(s, [
            ("starttag", "cdata", []),
            ("data", " <not a='start tag'> "),
            ("endtag", "cdata"),
            ])

    def test_illegal_declarations(self):
        s = 'abc<!spacer type="block" height="25">def'
        self.check_events(s, [
            ("data", "abc"),
            ("unknown decl", 'spacer type="block" height="25"'),
            ("data", "def"),
            ])

    def test_enumerated_attr_type(self):
        s = "<!DOCTYPE doc [<!ATTLIST doc attr (a | b) >]>"
        self.check_events(s, [
            ('decl', 'DOCTYPE doc [<!ATTLIST doc attr (a | b) >]'),
            ])

    # XXX These tests have been disabled by prefixing their names with
    # an underscore.  The first two exercise outstanding bugs in the
    # sgmllib module, and the third exhibits questionable behavior
    # that needs to be carefully considered before changing it.

    def _test_starttag_end_boundary(self):
        self.check_events("""<a b='<'>""", [("starttag", "a", [("b", "<")])])
        self.check_events("""<a b='>'>""", [("starttag", "a", [("b", ">")])])

    def _test_buffer_artefacts(self):
        output = [("starttag", "a", [("b", "<")])]
        self.check_events(["<a b='<'>"], output)
        self.check_events(["<a ", "b='<'>"], output)
        self.check_events(["<a b", "='<'>"], output)
        self.check_events(["<a b=", "'<'>"], output)
        self.check_events(["<a b='<", "'>"], output)
        self.check_events(["<a b='<'", ">"], output)

        output = [("starttag", "a", [("b", ">")])]
        self.check_events(["<a b='>'>"], output)
        self.check_events(["<a ", "b='>'>"], output)
        self.check_events(["<a b", "='>'>"], output)
        self.check_events(["<a b=", "'>'>"], output)
        self.check_events(["<a b='>", "'>"], output)
        self.check_events(["<a b='>'", ">"], output)

        output = [("comment", "abc")]
        self._run_check(["", "<!--abc-->"], output)
        self._run_check(["<", "!--abc-->"], output)
        self._run_check(["<!", "--abc-->"], output)
        self._run_check(["<!-", "-abc-->"], output)
        self._run_check(["<!--", "abc-->"], output)
        self._run_check(["<!--a", "bc-->"], output)
        self._run_check(["<!--ab", "c-->"], output)
        self._run_check(["<!--abc", "-->"], output)
        self._run_check(["<!--abc-", "->"], output)
        self._run_check(["<!--abc--", ">"], output)
        self._run_check(["<!--abc-->", ""], output)

    def _test_starttag_junk_chars(self):
        self.check_parse_error("<")
        self.check_parse_error("<>")
        self.check_parse_error("</$>")
        self.check_parse_error("</")
        self.check_parse_error("</a")
        self.check_parse_error("<$")
        self.check_parse_error("<$>")
        self.check_parse_error("<!")
        self.check_parse_error("<a $>")
        self.check_parse_error("<a")
        self.check_parse_error("<a foo='bar'")
        self.check_parse_error("<a foo='bar")
        self.check_parse_error("<a foo='>'")
        self.check_parse_error("<a foo='>")
        self.check_parse_error("<a foo=>")


def test_main():
    test_support.run_unittest(SGMLParserTestCase)


if __name__ == "__main__":
    test_main()
Commit	Line	Data
920dae64 AT	1	import pprint
	2	import sgmllib
	3	import unittest
	4	from test import test_support
	5
	6
	7	class EventCollector(sgmllib.SGMLParser):
	8
	9	def __init__(self):
	10	self.events = []
	11	self.append = self.events.append
	12	sgmllib.SGMLParser.__init__(self)
	13
	14	def get_events(self):
	15	# Normalize the list of events so that buffer artefacts don't
	16	# separate runs of contiguous characters.
	17	L = []
	18	prevtype = None
	19	for event in self.events:
	20	type = event[0]
	21	if type == prevtype == "data":
	22	L[-1] = ("data", L[-1][1] + event[1])
	23	else:
	24	L.append(event)
	25	prevtype = type
	26	self.events = L
	27	return L
	28
	29	# structure markup
	30
	31	def unknown_starttag(self, tag, attrs):
	32	self.append(("starttag", tag, attrs))
	33
	34	def unknown_endtag(self, tag):
	35	self.append(("endtag", tag))
	36
	37	# all other markup
	38
	39	def handle_comment(self, data):
	40	self.append(("comment", data))
	41
	42	def handle_charref(self, data):
	43	self.append(("charref", data))
	44
	45	def handle_data(self, data):
	46	self.append(("data", data))
	47
	48	def handle_decl(self, decl):
	49	self.append(("decl", decl))
	50
	51	def handle_entityref(self, data):
	52	self.append(("entityref", data))
	53
	54	def handle_pi(self, data):
	55	self.append(("pi", data))
	56
	57	def unknown_decl(self, decl):
	58	self.append(("unknown decl", decl))
	59
	60
	61	class CDATAEventCollector(EventCollector):
	62	def start_cdata(self, attrs):
	63	self.append(("starttag", "cdata", attrs))
	64	self.setliteral()
65
66
67	class SGMLParserTestCase(unittest.TestCase):
68
69	collector = EventCollector
70
71	def get_events(self, source):
72	parser = self.collector()
73	try:
74	for s in source:
75	parser.feed(s)
76	parser.close()
77	except:
78	#self.events = parser.events
79	raise
80	return parser.get_events()
81
82	def check_events(self, source, expected_events):
83	try:
84	events = self.get_events(source)
85	except:
86	import sys
87	#print >>sys.stderr, pprint.pformat(self.events)
88	raise
89	if events != expected_events:
90	self.fail("received events did not match expected events\n"
91	"Expected:\n" + pprint.pformat(expected_events) +
92	"\nReceived:\n" + pprint.pformat(events))
93
94	def check_parse_error(self, source):
95	parser = EventCollector()
96	try:
97	parser.feed(source)
98	parser.close()
99	except sgmllib.SGMLParseError:
100	pass
101	else:
102	self.fail("expected SGMLParseError for %r\nReceived:\n%s"
103	% (source, pprint.pformat(parser.get_events())))
104
105	def test_doctype_decl_internal(self):
106	inside = """\
107	DOCTYPE html PUBLIC '-//W3C//DTD HTML 4.01//EN'
108	SYSTEM 'http://www.w3.org/TR/html401/strict.dtd' [
109	<!ELEMENT html - O EMPTY>
110	<!ATTLIST html
111	version CDATA #IMPLIED
112	profile CDATA 'DublinCore'>
113	<!NOTATION datatype SYSTEM 'http://xml.python.org/notations/python-module'>
114	<!ENTITY myEntity 'internal parsed entity'>
115	<!ENTITY anEntity SYSTEM 'http://xml.python.org/entities/something.xml'>
116	<!ENTITY % paramEntity 'name\|name\|name'>
117	%paramEntity;
118	<!-- comment -->
119	]"""
120	self.check_events(["<!%s>" % inside], [
121	("decl", inside),
122	])
123
124	def test_doctype_decl_external(self):
125	inside = "DOCTYPE html PUBLIC '-//W3C//DTD HTML 4.01//EN'"
126	self.check_events("<!%s>" % inside, [
127	("decl", inside),
128	])
129
130	def test_underscore_in_attrname(self):
131	# SF bug #436621
132	"""Make sure attribute names with underscores are accepted"""
133	self.check_events("<a has_under _under>", [
134	("starttag", "a", [("has_under", "has_under"),
135	("_under", "_under")]),
136	])
137
138	def test_underscore_in_tagname(self):
139	# SF bug #436621
140	"""Make sure tag names with underscores are accepted"""
141	self.check_events("<has_under></has_under>", [
142	("starttag", "has_under", []),
143	("endtag", "has_under"),
144	])
145
146	def test_quotes_in_unquoted_attrs(self):
147	# SF bug #436621
148	"""Be sure quotes in unquoted attributes are made part of the value"""
149	self.check_events("<a href=foo'bar\"baz>", [
150	("starttag", "a", [("href", "foo'bar\"baz")]),
151	])
152
153	def test_xhtml_empty_tag(self):
154	"""Handling of XHTML-style empty start tags"""
155	self.check_events("<br />text<i></i>", [
156	("starttag", "br", []),
157	("data", "text"),
158	("starttag", "i", []),
159	("endtag", "i"),
160	])
161
162	def test_processing_instruction_only(self):
163	self.check_events("<?processing instruction>", [
164	("pi", "processing instruction"),
165	])
166
167	def test_bad_nesting(self):
168	self.check_events("<a><b></a></b>", [
169	("starttag", "a", []),
170	("starttag", "b", []),
171	("endtag", "a"),
172	("endtag", "b"),
173	])
174
175	def test_bare_ampersands(self):
176	self.check_events("this text & contains & ampersands &", [
177	("data", "this text & contains & ampersands &"),
178	])
179
180	def test_bare_pointy_brackets(self):
181	self.check_events("this < text > contains < bare>pointy< brackets", [
182	("data", "this < text > contains < bare>pointy< brackets"),
183	])
184
185	def test_attr_syntax(self):
186	output = [
187	("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", "e")])
188	]
189	self.check_events("""<a b='v' c="v" d=v e>""", output)
190	self.check_events("""<a b = 'v' c = "v" d = v e>""", output)
191	self.check_events("""<a\nb\n=\n'v'\nc\n=\n"v"\nd\n=\nv\ne>""", output)
192	self.check_events("""<a\tb\t=\t'v'\tc\t=\t"v"\td\t=\tv\te>""", output)
193
194	def test_attr_values(self):
195	self.check_events("""<a b='xxx\n\txxx' c="yyy\t\nyyy" d='\txyz\n'>""",
196	[("starttag", "a", [("b", "xxx\n\txxx"),
197	("c", "yyy\t\nyyy"),
198	("d", "\txyz\n")])
199	])
200	self.check_events("""<a b='' c="">""", [
201	("starttag", "a", [("b", ""), ("c", "")]),
202	])
203	# URL construction stuff from RFC 1808:
204	safe = "$-_.+"
205	extra = "!*'(),"
206	reserved = ";/?:@&="
207	url = "http://example.com:8080/path/to/file?%s%s%s" % (
208	safe, extra, reserved)
209	self.check_events("""<e a=%s>""" % url, [
210	("starttag", "e", [("a", url)]),
211	])
212	# Regression test for SF patch #669683.
213	self.check_events("<e a=rgb(1,2,3)>", [
214	("starttag", "e", [("a", "rgb(1,2,3)")]),
215	])
216
217	def test_attr_funky_names(self):
218	self.check_events("""<a a.b='v' c:d=v e-f=v>""", [
219	("starttag", "a", [("a.b", "v"), ("c:d", "v"), ("e-f", "v")]),
220	])
221
222	def test_illegal_declarations(self):
223	s = 'abc<!spacer type="block" height="25">def'
224	self.check_events(s, [
225	("data", "abc"),
226	("unknown decl", 'spacer type="block" height="25"'),
227	("data", "def"),
228	])
229
230	def test_weird_starttags(self):
231	self.check_events("<a<a>", [
232	("starttag", "a", []),
233	("starttag", "a", []),
234	])
235	self.check_events("</a<a>", [
236	("endtag", "a"),
237	("starttag", "a", []),
238	])
239
240	def test_declaration_junk_chars(self):
241	self.check_parse_error("<!DOCTYPE foo $ >")
242
243	def test_get_starttag_text(self):
244	s = """<foobar \n one="1"\ttwo=2 >"""
245	self.check_events(s, [
246	("starttag", "foobar", [("one", "1"), ("two", "2")]),
247	])
248
249	def test_cdata_content(self):
250	s = ("<cdata> <!-- not a comment --> &not-an-entity-ref; </cdata>"
251	"<notcdata> <!-- comment --> </notcdata>")
252	self.collector = CDATAEventCollector
253	self.check_events(s, [
254	("starttag", "cdata", []),
255	("data", " <!-- not a comment --> &not-an-entity-ref; "),
256	("endtag", "cdata"),
257	("starttag", "notcdata", []),
258	("data", " "),
259	("comment", " comment "),
260	("data", " "),
261	("endtag", "notcdata"),
262	])
263	s = """<cdata> <not a='start tag'> </cdata>"""
264	self.check_events(s, [
265	("starttag", "cdata", []),
266	("data", " <not a='start tag'> "),
267	("endtag", "cdata"),
268	])
269
270	def test_illegal_declarations(self):
271	s = 'abc<!spacer type="block" height="25">def'
272	self.check_events(s, [
273	("data", "abc"),
274	("unknown decl", 'spacer type="block" height="25"'),
275	("data", "def"),
276	])
277
278	def test_enumerated_attr_type(self):
279	s = "<!DOCTYPE doc [<!ATTLIST doc attr (a \| b) >]>"
280	self.check_events(s, [
281	('decl', 'DOCTYPE doc [<!ATTLIST doc attr (a \| b) >]'),
282	])
283
284	# XXX These tests have been disabled by prefixing their names with
285	# an underscore. The first two exercise outstanding bugs in the
286	# sgmllib module, and the third exhibits questionable behavior
287	# that needs to be carefully considered before changing it.
288
289	def _test_starttag_end_boundary(self):
290	self.check_events("""<a b='<'>""", [("starttag", "a", [("b", "<")])])
291	self.check_events("""<a b='>'>""", [("starttag", "a", [("b", ">")])])
292
293	def _test_buffer_artefacts(self):
294	output = [("starttag", "a", [("b", "<")])]
295	self.check_events(["<a b='<'>"], output)
296	self.check_events(["<a ", "b='<'>"], output)
297	self.check_events(["<a b", "='<'>"], output)
298	self.check_events(["<a b=", "'<'>"], output)
299	self.check_events(["<a b='<", "'>"], output)
300	self.check_events(["<a b='<'", ">"], output)
301
302	output = [("starttag", "a", [("b", ">")])]
303	self.check_events(["<a b='>'>"], output)
304	self.check_events(["<a ", "b='>'>"], output)
305	self.check_events(["<a b", "='>'>"], output)
306	self.check_events(["<a b=", "'>'>"], output)
307	self.check_events(["<a b='>", "'>"], output)
308	self.check_events(["<a b='>'", ">"], output)
309
310	output = [("comment", "abc")]
311	self._run_check(["", "<!--abc-->"], output)
312	self._run_check(["<", "!--abc-->"], output)
313	self._run_check(["<!", "--abc-->"], output)
314	self._run_check(["<!-", "-abc-->"], output)
315	self._run_check(["<!--", "abc-->"], output)
316	self._run_check(["<!--a", "bc-->"], output)
317	self._run_check(["<!--ab", "c-->"], output)
318	self._run_check(["<!--abc", "-->"], output)
319	self._run_check(["<!--abc-", "->"], output)
320	self._run_check(["<!--abc--", ">"], output)
321	self._run_check(["<!--abc-->", ""], output)
322
323	def _test_starttag_junk_chars(self):
324	self.check_parse_error("<")
325	self.check_parse_error("<>")
326	self.check_parse_error("</$>")
327	self.check_parse_error("</")
328	self.check_parse_error("</a")
329	self.check_parse_error("<$")
330	self.check_parse_error("<$>")
331	self.check_parse_error("<!")
332	self.check_parse_error("<a $>")
333	self.check_parse_error("<a")
334	self.check_parse_error("<a foo='bar'")
335	self.check_parse_error("<a foo='bar")
336	self.check_parse_error("<a foo='>'")
337	self.check_parse_error("<a foo='>")
338	self.check_parse_error("<a foo=>")
339
340
341	def test_main():
342	test_support.run_unittest(SGMLParserTestCase)
343
344
345	if __name__ == "__main__":
346	test_main()