Update version number and release date.
[python/dscho.git] / Lib / test / test_sgmllib.py
blob6f4454f8f3fb1f3485dc2dfbee0ca2d3d41acabd
1 import pprint
2 import sgmllib
3 import unittest
4 from test import test_support
7 class EventCollector(sgmllib.SGMLParser):
9 def __init__(self):
10 self.events = []
11 self.append = self.events.append
12 sgmllib.SGMLParser.__init__(self)
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
29 # structure markup
31 def unknown_starttag(self, tag, attrs):
32 self.append(("starttag", tag, attrs))
34 def unknown_endtag(self, tag):
35 self.append(("endtag", tag))
37 # all other markup
39 def handle_comment(self, data):
40 self.append(("comment", data))
42 def handle_charref(self, data):
43 self.append(("charref", data))
45 def handle_data(self, data):
46 self.append(("data", data))
48 def handle_decl(self, decl):
49 self.append(("decl", decl))
51 def handle_entityref(self, data):
52 self.append(("entityref", data))
54 def handle_pi(self, data):
55 self.append(("pi", data))
57 def unknown_decl(self, decl):
58 self.append(("unknown decl", decl))
61 class CDATAEventCollector(EventCollector):
62 def start_cdata(self, attrs):
63 self.append(("starttag", "cdata", attrs))
64 self.setliteral()
67 class SGMLParserTestCase(unittest.TestCase):
69 collector = EventCollector
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()
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))
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())))
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),
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),
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")]),
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"),
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")]),
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"),
162 def test_processing_instruction_only(self):
163 self.check_events("<?processing instruction>", [
164 ("pi", "processing instruction"),
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"),
175 def test_bare_ampersands(self):
176 self.check_events("this text & contains & ampersands &", [
177 ("data", "this text & contains & ampersands &"),
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"),
185 def test_attr_syntax(self):
186 output = [
187 ("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", "e")])
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)
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")])
200 self.check_events("""<a b='' c="">""", [
201 ("starttag", "a", [("b", ""), ("c", "")]),
203 # Regression test for SF patch #669683.
204 self.check_events("<e a=rgb(1,2,3)>", [
205 ("starttag", "e", [("a", "rgb(1,2,3)")]),
208 def test_attr_funky_names(self):
209 self.check_events("""<a a.b='v' c:d=v e-f=v>""", [
210 ("starttag", "a", [("a.b", "v"), ("c:d", "v"), ("e-f", "v")]),
213 def test_illegal_declarations(self):
214 s = 'abc<!spacer type="block" height="25">def'
215 self.check_events(s, [
216 ("data", "abc"),
217 ("unknown decl", 'spacer type="block" height="25"'),
218 ("data", "def"),
221 def test_weird_starttags(self):
222 self.check_events("<a<a>", [
223 ("starttag", "a", []),
224 ("starttag", "a", []),
226 self.check_events("</a<a>", [
227 ("endtag", "a"),
228 ("starttag", "a", []),
231 def test_declaration_junk_chars(self):
232 self.check_parse_error("<!DOCTYPE foo $ >")
234 def test_get_starttag_text(self):
235 s = """<foobar \n one="1"\ttwo=2 >"""
236 self.check_events(s, [
237 ("starttag", "foobar", [("one", "1"), ("two", "2")]),
240 def test_cdata_content(self):
241 s = ("<cdata> <!-- not a comment --> &not-an-entity-ref; </cdata>"
242 "<notcdata> <!-- comment --> </notcdata>")
243 self.collector = CDATAEventCollector
244 self.check_events(s, [
245 ("starttag", "cdata", []),
246 ("data", " <!-- not a comment --> &not-an-entity-ref; "),
247 ("endtag", "cdata"),
248 ("starttag", "notcdata", []),
249 ("data", " "),
250 ("comment", " comment "),
251 ("data", " "),
252 ("endtag", "notcdata"),
254 s = """<cdata> <not a='start tag'> </cdata>"""
255 self.check_events(s, [
256 ("starttag", "cdata", []),
257 ("data", " <not a='start tag'> "),
258 ("endtag", "cdata"),
261 def test_illegal_declarations(self):
262 s = 'abc<!spacer type="block" height="25">def'
263 self.check_events(s, [
264 ("data", "abc"),
265 ("unknown decl", 'spacer type="block" height="25"'),
266 ("data", "def"),
269 def test_enumerated_attr_type(self):
270 s = "<!DOCTYPE doc [<!ATTLIST doc attr (a | b) >]>"
271 self.check_events(s, [
272 ('decl', 'DOCTYPE doc [<!ATTLIST doc attr (a | b) >]'),
275 # XXX These tests have been disabled by prefixing their names with
276 # an underscore. The first two exercise outstanding bugs in the
277 # sgmllib module, and the third exhibits questionable behavior
278 # that needs to be carefully considered before changing it.
280 def _test_starttag_end_boundary(self):
281 self.check_events("""<a b='<'>""", [("starttag", "a", [("b", "<")])])
282 self.check_events("""<a b='>'>""", [("starttag", "a", [("b", ">")])])
284 def _test_buffer_artefacts(self):
285 output = [("starttag", "a", [("b", "<")])]
286 self.check_events(["<a b='<'>"], output)
287 self.check_events(["<a ", "b='<'>"], output)
288 self.check_events(["<a b", "='<'>"], output)
289 self.check_events(["<a b=", "'<'>"], output)
290 self.check_events(["<a b='<", "'>"], output)
291 self.check_events(["<a b='<'", ">"], output)
293 output = [("starttag", "a", [("b", ">")])]
294 self.check_events(["<a b='>'>"], output)
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)
301 def _test_starttag_junk_chars(self):
302 self.check_parse_error("<")
303 self.check_parse_error("<>")
304 self.check_parse_error("</$>")
305 self.check_parse_error("</")
306 self.check_parse_error("</a")
307 self.check_parse_error("<$")
308 self.check_parse_error("<$>")
309 self.check_parse_error("<!")
310 self.check_parse_error("<a $>")
311 self.check_parse_error("<a")
312 self.check_parse_error("<a foo='bar'")
313 self.check_parse_error("<a foo='bar")
314 self.check_parse_error("<a foo='>'")
315 self.check_parse_error("<a foo='>")
316 self.check_parse_error("<a foo=>")
319 def test_main():
320 test_support.run_unittest(SGMLParserTestCase)
323 if __name__ == "__main__":
324 test_main()