| Fred Drake | bd3090d | 2001-05-18 15:32:59 +0000 | [diff] [blame] | 1 | """Tests for HTMLParser.py.""" |
| 2 | |
| 3 | import HTMLParser |
| 4 | import sys |
| 5 | import test_support |
| 6 | import unittest |
| 7 | |
| 8 | |
| 9 | class EventCollector(HTMLParser.HTMLParser): |
| 10 | |
| 11 | def __init__(self): |
| 12 | self.events = [] |
| 13 | self.append = self.events.append |
| 14 | HTMLParser.HTMLParser.__init__(self) |
| 15 | |
| 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 | |
| 62 | |
| 63 | class EventCollectorExtra(EventCollector): |
| 64 | |
| 65 | def handle_starttag(self, tag, attrs): |
| 66 | EventCollector.handle_starttag(self, tag, attrs) |
| 67 | self.append(("starttag_text", self.get_starttag_text())) |
| 68 | |
| 69 | |
| 70 | class TestCaseBase(unittest.TestCase): |
| 71 | |
| 72 | # Constant pieces of source and events |
| 73 | prologue = "" |
| 74 | epilogue = "" |
| 75 | initial_events = [] |
| 76 | final_events = [] |
| 77 | |
| 78 | def _run_check(self, source, events, collector=EventCollector): |
| 79 | parser = collector() |
| 80 | parser.feed(self.prologue) |
| 81 | for s in source: |
| 82 | parser.feed(s) |
| 83 | for c in self.epilogue: |
| 84 | parser.feed(c) |
| 85 | parser.close() |
| 86 | self.assert_(parser.get_events() == |
| 87 | self.initial_events + events + self.final_events, |
| 88 | parser.get_events()) |
| 89 | |
| 90 | def _run_check_extra(self, source, events): |
| 91 | self._run_check(source, events, EventCollectorExtra) |
| 92 | |
| 93 | def _parse_error(self, source): |
| 94 | def parse(source=source): |
| 95 | parser = HTMLParser.HTMLParser() |
| 96 | parser.feed(source) |
| 97 | parser.close() |
| 98 | self.assertRaises(HTMLParser.HTMLParseError, parse) |
| 99 | |
| 100 | |
| 101 | class HTMLParserTestCase(TestCaseBase): |
| 102 | |
| 103 | def check_processing_instruction_only(self): |
| 104 | self._run_check("<?processing instruction>", [ |
| 105 | ("pi", "processing instruction"), |
| 106 | ]) |
| 107 | |
| 108 | def check_simple_html(self): |
| 109 | self._run_check(""" |
| 110 | <!DOCTYPE html PUBLIC 'foo'> |
| 111 | <HTML>&entity;  |
| 112 | <!--comment1a |
| 113 | -></foo><bar><<?pi?></foo<bar |
| 114 | comment1b--> |
| 115 | <Img sRc='Bar' isMAP>sample |
| 116 | text |
| 117 | <!--comment2a-- --comment2b--> |
| 118 | </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><<?pi?></foo<bar\ncomment1b"), |
| 128 | ("data", "\n"), |
| 129 | ("starttag", "img", [("src", "Bar"), ("ismap", None)]), |
| 130 | ("data", "sample\ntext\n"), |
| 131 | ("comment", "comment2a-- --comment2b"), |
| 132 | ("data", "\n"), |
| 133 | ("endtag", "html"), |
| 134 | ("data", "\n"), |
| 135 | ]) |
| 136 | |
| 137 | def check_bad_nesting(self): |
| 138 | self._run_check("<a><b></a></b>", [ |
| 139 | ("starttag", "a", []), |
| 140 | ("starttag", "b", []), |
| 141 | ("endtag", "a"), |
| 142 | ("endtag", "b"), |
| 143 | ]) |
| 144 | |
| 145 | def check_attr_syntax(self): |
| 146 | output = [ |
| 147 | ("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", None)]) |
| 148 | ] |
| 149 | self._run_check("""<a b='v' c="v" d=v e>""", output) |
| 150 | self._run_check("""<a b = 'v' c = "v" d = v e>""", output) |
| 151 | self._run_check("""<a\nb\n=\n'v'\nc\n=\n"v"\nd\n=\nv\ne>""", output) |
| 152 | self._run_check("""<a\tb\t=\t'v'\tc\t=\t"v"\td\t=\tv\te>""", output) |
| 153 | |
| 154 | def check_attr_values(self): |
| 155 | self._run_check("""<a b='xxx\n\txxx' c="yyy\t\nyyy" d='\txyz\n'>""", |
| 156 | [("starttag", "a", [("b", "xxx\n\txxx"), |
| 157 | ("c", "yyy\t\nyyy"), |
| 158 | ("d", "\txyz\n")]) |
| 159 | ]) |
| 160 | self._run_check("""<a b='' c="">""", [ |
| 161 | ("starttag", "a", [("b", ""), ("c", "")]), |
| 162 | ]) |
| 163 | |
| 164 | def check_attr_entity_replacement(self): |
| 165 | self._run_check("""<a b='&><"''>""", [ |
| 166 | ("starttag", "a", [("b", "&><\"'")]), |
| 167 | ]) |
| 168 | |
| 169 | def check_attr_funky_names(self): |
| 170 | self._run_check("""<a a.b='v' c:d=v e-f=v>""", [ |
| 171 | ("starttag", "a", [("a.b", "v"), ("c:d", "v"), ("e-f", "v")]), |
| 172 | ]) |
| 173 | |
| 174 | def check_starttag_end_boundary(self): |
| 175 | self._run_check("""<a b='<'>""", [("starttag", "a", [("b", "<")])]) |
| 176 | self._run_check("""<a b='>'>""", [("starttag", "a", [("b", ">")])]) |
| 177 | |
| 178 | def check_buffer_artefacts(self): |
| 179 | output = [("starttag", "a", [("b", "<")])] |
| 180 | self._run_check(["<a b='<'>"], output) |
| 181 | self._run_check(["<a ", "b='<'>"], output) |
| 182 | self._run_check(["<a b", "='<'>"], output) |
| 183 | self._run_check(["<a b=", "'<'>"], output) |
| 184 | self._run_check(["<a b='<", "'>"], output) |
| 185 | self._run_check(["<a b='<'", ">"], output) |
| 186 | |
| 187 | output = [("starttag", "a", [("b", ">")])] |
| 188 | self._run_check(["<a b='>'>"], output) |
| 189 | self._run_check(["<a ", "b='>'>"], output) |
| 190 | self._run_check(["<a b", "='>'>"], output) |
| 191 | self._run_check(["<a b=", "'>'>"], output) |
| 192 | self._run_check(["<a b='>", "'>"], output) |
| 193 | self._run_check(["<a b='>'", ">"], output) |
| 194 | |
| 195 | def check_starttag_junk_chars(self): |
| 196 | self._parse_error("<") |
| 197 | self._parse_error("<>") |
| 198 | self._parse_error("</>") |
| 199 | self._parse_error("</$>") |
| 200 | self._parse_error("</") |
| 201 | self._parse_error("</a") |
| Fred Drake | bd3090d | 2001-05-18 15:32:59 +0000 | [diff] [blame] | 202 | self._parse_error("<a<a>") |
| 203 | self._parse_error("</a<a>") |
| 204 | self._parse_error("<$") |
| 205 | self._parse_error("<$>") |
| 206 | self._parse_error("<!") |
| 207 | self._parse_error("<a $>") |
| 208 | self._parse_error("<a") |
| 209 | self._parse_error("<a foo='bar'") |
| 210 | self._parse_error("<a foo='bar") |
| 211 | self._parse_error("<a foo='>'") |
| 212 | self._parse_error("<a foo='>") |
| 213 | self._parse_error("<a foo=>") |
| 214 | |
| 215 | def check_declaration_junk_chars(self): |
| 216 | self._parse_error("<!DOCTYPE foo $ >") |
| 217 | |
| 218 | def check_startendtag(self): |
| 219 | self._run_check("<p/>", [ |
| 220 | ("startendtag", "p", []), |
| 221 | ]) |
| 222 | self._run_check("<p></p>", [ |
| 223 | ("starttag", "p", []), |
| 224 | ("endtag", "p"), |
| 225 | ]) |
| 226 | self._run_check("<p><img src='foo' /></p>", [ |
| 227 | ("starttag", "p", []), |
| 228 | ("startendtag", "img", [("src", "foo")]), |
| 229 | ("endtag", "p"), |
| 230 | ]) |
| 231 | |
| 232 | def check_get_starttag_text(self): |
| 233 | s = """<foo:bar \n one="1"\ttwo=2 >""" |
| 234 | self._run_check_extra(s, [ |
| 235 | ("starttag", "foo:bar", [("one", "1"), ("two", "2")]), |
| 236 | ("starttag_text", s)]) |
| 237 | |
| 238 | def check_cdata_content(self): |
| 239 | s = """<script> <!-- not a comment --> ¬-an-entity-ref; </script>""" |
| 240 | self._run_check(s, [ |
| 241 | ("starttag", "script", []), |
| 242 | ("data", " <!-- not a comment --> ¬-an-entity-ref; "), |
| 243 | ("endtag", "script"), |
| 244 | ]) |
| 245 | s = """<script> <not a='start tag'> </script>""" |
| 246 | self._run_check(s, [ |
| 247 | ("starttag", "script", []), |
| 248 | ("data", " <not a='start tag'> "), |
| 249 | ("endtag", "script"), |
| 250 | ]) |
| 251 | |
| 252 | |
| 253 | test_support.run_unittest(HTMLParserTestCase) |